Identity and Emotion: Development Through Self-Organization (Studies in Emotion and Social Interaction)

Identity and Emotion Development through Self-Organization Recent ideas concerning the development of self and identity...

Author: Harke A. Bosma | E. Saskia Kunnen

39 downloads 1827 Views 940KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Identity and Emotion Development through Self-Organization

Recent ideas concerning the development of self and identity have stressed the importance of moving away from an approach which is mainly concerned with outcomes, to one which focuses instead on processes of development and, more specifically, on a relational perspective on these processes. Identity and Emotion focuses on the individual development of identity and the processes involved. By working from emotions and a dynamic systems perspective the book offers a new and exciting approach to human identity and its development across the lifespan. The contributors to the book are specialists in this approach, and offer challenging ideas on the development of identity as a self-organizing process. The book offers a wealth of new ideas and insights, but also concentrates on the ways these insights can be translated into research. Harke A. Bosma is Associate Professor of Developmental Psychology at the University of Groningen where he lectures in adolescent psychology with responsibility for research and teaching in this area. He has published in both Dutch and English and is co-editor of Coping and Self-Concept in Adolescence (1990) with A. E. Jackson and Identity and Development (1994) with T. L. G. Graafsma, H. D. Grotevant, and D. J. de Levita. E. Saskia Kunnen is Associate Professor of Developmental Psychology at the University of Utrecht where she researches and lectures in the areas of personal development in adolescence and adulthood, and application of dynamic systems theory and models in this field. She has published in both Dutch and English including articles in the International Journal of Behavioural Development and New Ideas in Psychology.

MMMM

STUDI ES I N E MOTI ON AN D SOCI A L IN T ER A CTION Second series Series editors Keith Oatley University of Toronto Antony Manstead University of Amsterdam This series is jointly published by the Cambridge University Press and the Editions de la Maison des Sciences de l’Homme, as part of the joint publishing agreement established in 1977 between the Fondation de la Maison des Sciences de l’Homme and the Syndics of the Cambridge University Press. Cette publication est publieé co-e´dition par Cambridge University Press et les Editions de la Maison des Sciences de l’Homme. Elle s’inte`gre dans le programme de co-e´dition e´tabli en 1977 par la Fondation de la Maison des Sciences de l’Homme et les Syndics de Cambridge University Press.

Titles published in the second series The psychology of facial expression 0 521 49667 5 hardback and 0 521 58796 4 paperback Edited by James A. Russell and Jose´ Miguel Fernańdez-Dois Emotions, the social bond, and human reality: part/whole analysis 0 521 58491 4 hardback and 0 521 58454 7 paperback Thomas J. Scheff Intersubjective communication and emotion in early ontogeny 0 521 62257 3 hardback and 2 7351 07728 hardback (France only) Edited by Stein Bra˚ten Emotion across languages and cultures: diversity and universals 0 521 59042 6 hardback and 0 521 59971 7 paperback Anna Wierzbicka Communicating emotion: social, moral and cultural processes 0 521 55315 6 hardback and 0 521 55741 0 paperback Sally Planalp The social context of nonverbal behavior 0 521 58371 3 hardback and 0 521 58666 6 paperback Edited by Pierre Philippot, Robert S. Feldman, and Erik J. Coats

Feeling and thinking: the role of affect in social cognition 0 521 64223 X hardback Edited by Joseph P. Forgas Gender and emotion: social psychological perspectives 0 521 63015 0 hardback and 0 521 63986 7 paperback Edited by Agneta H. Fischer Causes and consequences of feelings 0 521 63325 7 hardback and 0 521 63363 X paperback Leonard Berkowitz Emotions and beliefs: how feelings influence thoughts 0 521 77138 2 hardback and 0 521 78734 3 paperback Edited by Nico H. Frijda, Antony S. R. Manstead, and Sacha Bem For a list of titles in the First Series in Studies in Emotion and Social Interaction, see end of book.

Identity and Emotion Development through Self-Organization

edited by

Harke A. Bosma and E. Saskia Kunnen

Editions de la Maison des Sciences de l’Homme Paris

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 2RU, UK Editions de la Maison des Sciences de l’homme 54 Boulevard Raspail, 75270 Paris Cedex 06, France Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521661850 © Maison des Sciences de l’Homme and Cambridge University Press 2001 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2001 This digitally printed first paperback version 2005 A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Identity and emotion: development through self-organization / edited by Harke A. Bosma & E. Saskia Kunnen. p. cm. – (Studies in emotion and social interaction) Includes bibliographical references and index. ISBN 0 521 66185 4 1. Identity (Psychology) 2. Personality and emotions. I. Bosma, Harke, 1945– II. Kunnen, E. Saskia. III. Series. BF697.I349 2001 155.2–dc21 00-046743 ISBN-13 978-0-521-66185-0 hardback ISBN-10 0-521-66185-4 hardback ISBN-13 978-0-521-02156-2 paperback ISBN-10 0-521-02156-1 paperback

Contents

List of contributors Preface

page ix xiii 1

1 Introduction E. Saskia Kunnen, Harke A. Bosma, Cor P. M. Van Halen, and Matty Van der Meulen

2 Developments in self-concept theory and research: affect, context, and variability

10

Matty Van der Meulen

Commentary: the self-concept is dead, long live . . . which construct or process? Differentiation and organization of self-related theories

33

Annerieke Oosterwegel

39

3 The self and emotions Nico H. Frijda

Commentary: the self and emotions

58

Seymour Epstein

4 Fish, foxes, and talking in the classroom: introducing dynamic systems concepts and approaches

64

Paul L. C. Van Geert

Commentary: fish, foxes, identity, and emotion

89

Linda A. Camras and George F. Michel

5 A relational perspective on the development of self and emotion

93

Alan Fogel

Commentary: the personal experience of coherence Jeroen Jansz

vii

115

viii

Contents 6 Affective processes in a multivoiced self

120

Hubert J. M. Hermans and Els Hermans-Jansen

Commentary: affective processes in a multivoiced self in action

141

Leni Verhofstadt-Dene`ve

7 Old–new answers and new–old questions for personality and emotion: a matter of complexity

151

Jeannette Haviland-Jones, David Boulifard, and Carol Magai

Commentary: emotions as sources of information about the self

172

Peter G. Heymans

8 Cognitive–emotional self-organization in personality development and personal identity

177

Marc D. Lewis and Michel Ferrari

Commentary: two faces of identity

199

Carol Magai

9 A self-organizational approach to identity and emotions: an overview and implications

202

E. Saskia Kunnen, Harke A. Bosma, Cor P. M. Van Halen, and Matty Van der Meulen

References Author index Subject index

231 259 265

Contributors

Harke A. Bosma Department of Developmental Psychology, University of Groningen, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands David Boulifard Department of Psychology, State University of New Jersey, New Brunswick, NJ 08903, USA Linda A. Camras Department of Psychology, DePaul University, 2219 N. Kenmore Ave., Chicago, IL 60614, USA Seymour Epstein Department of Psychology, University of Massachusetts at Amhurst, Amhurst, MA 01003, USA Michel Ferrari Department of Human Development and Applied Psychology, Ontario Institute for Studies in Education, University of Toronto, 252 Bloor Street West, Toronto, Ontario M5S 1V6, Canada Alan Fogel Department of Psychology, University of Utah, 390 S. 1530 E., Salt Lake City, UT 84112-0251, USA Nico H. Frijda Department of Psychonomics, University of Amsterdam, Roeterstraat 15, 1018 WB Amsterdam, The Netherlands Jeanette Haviland-Jones Department of Psychology, State University of New Jersey, New Brunswick, NJ 08903, USA ix

x

Contributors

Hubert J. M. Hermans Department of Clinical and Personality Psychology, University of Nijmegen, PO Box 9104, 6500 HE Nijmegen, The Netherlands Els Hermans-Jansen Center for Personality-Research, Bosweg 18, 6571 CD Berg en Dal, The Netherlands Peter G. Heymans Department of Developmental Psychology, University of Utrecht, Heidelberglaan 2, 3584 CS Utrecht, The Netherlands Jeroen Jansz Department of Political and Social-Cultural Sciences, University of Amsterdam, Oude Hoogstraat 24, 1012 CE Amsterdam, The Netherlands E. Saskia Kunnen Department of Developmental Psychology, University of Utrecht, Heidelberglaan 2, 3584 CS Utrecht, The Netherlands Marc D. Lewis Department of Human Development and Applied Psychology, Ontario Institute for Studies in Education, University of Toronto, 252 Bloor Street West, Toronto, Ontario M5S 1V6, Canada Carol Magai Department of Psychology, Long Island University, Brooklyn, NY 11201, USA George F. Michel Department of Psychology, DePaul University, 2219 N. Kenmore Ave., Chicago, IL 60614, USA Annerieke Oosterwegel Department of Developmental Psychology, University of Utrecht, PO Box 80.140, Heidelberglaan 2, 3584 CS Utrecht, The Netherlands Paul L. C. Van Geert Department of Psychology, University of Groningen Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands Cor P. M. Van Halen Department of Developmental Psychology, University of Groningen, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands

Contributors Matty Van der Meulen Department of Psychology, University of Groningen Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands Leni Verhofstadt-Dene`ve Department of Developmental and Personality Psychology, University of Ghent, Henri Dunantlaan 2, B-9000 Ghent, Belgium

xi

MMMM

Preface

Recent ideas concerning the development of self and identity have stressed the importance of moving away from an approach which is mainly concerned with outcomes, to one which focuses instead on processes of development and, more specifically, on a relational perspective on these processes. This change has also led to increased attention to the role of emotions in the development of self and identity. These developments offer new possibilities and challenges for theory and research. However, they also lead to new concerns and questions at a theoretical, as well as a methodological level. In 1996, a workshop on the development of self and identity was organized with the explicit intention of focusing on these new trends. The main topics of the workshop were the conceptualization of the development of the person as an emotional, relational, and selforganizing process and the way in which such a dynamic conceptualization can be translated into empirical research employing methodological approaches which are adapted to the study of dynamic processes in self-stability and change. During the intense and lively workshop discussions, new ideas were developed, and serious attempts were made to clarify and elaborate the development of self and identity as an inherently emotional process embedded within a relational context. This book can be seen as a next step in this discussion. Most of the contributors to this volume were participants in the workshop. Using the workshop discussions as a starting point, they were asked to elaborate their perspective both theoretically and methodologically. Their ideas and the comments provided by others reflect and extend the nature of the workshop discussions and provide an illustration of the self-organizing, dialogical, and open approach which is advocated in this volume. In the organization of the original workshop and the preparation of this volume we have received considerable support from individuals and xiii

xiv

Preface

organizations. Here we wish to express our gratitude for their help. The workshop was supported financially by the Dutch Science Foundation (NWO), the Association for Developmental Psychological Research, and the Departments of Developmental Psychology of the University of Utrecht and of the University of Groningen. The latter department provided additional help by paying for the transcription of the audio-taped workshop discussions. This was greatly appreciated since the resulting transcripts not only brought the workshop discussions back to mind, but also helped the authors to prepare their chapters for the book. The Departments of Developmental Psychology of Utrecht and Groningen Universities also gave financial support for the preparation of the book itself. A variety of individuals provided us with essential help and we wish to express our deep thanks to each of them. Nel Wiersma transcribed the workshop discussions and also helped to abbreviate some of the contributions. Fiona Buiter corrected the English of the non-native authors. Leen Van Geert provided editorial assistance in the preparation of the manuscript in its final form. Their help was truly indispensable. We are also very grateful to our publishers, Catherine Max and Sarah Caro, for their trust in the project and the series editor Tony Manstead for his very careful and supportive review of the manuscript. This book would never have been prepared without the enthusiastic and creative efforts of the presenters/authors. They all kept on schedule and conscientiously revised their texts on the basis of the feedback we provided. We are thankful (and proud) that they were all willing to share their scholarship in their contributions to the book. Finally, we want to give our special thanks to Matty Van der Meulen and Cor Van Halen, who helped us organize the workshop, who greatly contributed to the introductory and concluding chapters of the book, and who gave us the essential comradely support during times when the completion of the book seemed a distant prospect. Harke A. Bosma and E. Saskia Kunnen

CH A P T E R 1

Introduction E. Saskia Kunnen, Harke A. Bosma, Cor P. M. Van Halen, and Matty Van der Meulen

Over the years, the topics of self and identity have received a great deal of attention in the field of psychology. The literature is replete with investigations into self-concept, people’s perceptions, ideas, and feelings about themselves, and into identity, people’s perceptions of their own sameness and continuity (Oosterwegel and Wicklund, 1995). Although researchers in the field choose to focus on different facets of self and identity, broad theoretical trends can be identified. Traditionally, theorists have conceptualized self and identity as cognitive structures (Hattie, 1992; Marcia, Waterman, Matteson, Archer, and Orlofsky, 1993). These structures have mostly been regarded as stable mental representations that – once they have become crystallized through the repeated processing of personal information – control our further behavior (e.g. Markus and Wurf, 1987). As a result, the phenomena that are seen as indicative of self and identity are implicitly reduced to a self-concept: a set of beliefs about oneself. This set of beliefs, moreover, is considered to have dynamic implications for the regulation of our actions. The self-concept is thought to serve as an interpretative framework that integrates our personal experiences and as a regulative basis to guide further behavior. However, since the mental representations that constitute the self-concept are seen as stable carriers of personal information, deeply engraved in our memory, the traditional approaches are more suitable for accounting for aspects like stability and continuity than for the dynamics that emerge in self and identity. For instance, how do aspects of self and identity develop over time, why do they change with different situations and circumstances, how are they affected by emotional states or emotions as such? Cognitivists have tried to solve this dilemma by postulating increasingly complex representational structures underlying the self-concept, operating through intra-individual mechanisms of information processing. At this point, however, a purely cognitive approach to self and identity runs into serious conceptual problems. Neither the supposed representational complexity, nor the supposed information processing mechanisms can be inferred directly from what people tell about themselves. 1

2

E. Saskia Kunnen, Harke A. Bosma et al.

Thus, the self-concept is no longer what people feel, think, or say about themselves, but has become a hypothetical, almost homunculary construct with a life of its own. This comes close to a kind of ‘‘agency smuggling,’’ as Gergen once put it (1984). In fact, a cognitivistic view turns us into rule-instructed operators without clarifying how we adapt to the ever-changing contexts of everyday life or even initiate changes (e.g. Bruner, 1990). Concepts like self and identity have thus become abstracted from the behavior that we exhibit in relation to the natural circumstances in which we have learned to function. Empirical research also suggests that the common conceptualizations have strong limitations. Given the assumed functional importance and personal relevance of self and identity, the empirically found relations between cognitive representations about oneself and constructs such as well-being, adaptation, motivation, and performance behavior, turn out to be less straightforward than theoretically expected. One reason may be that the prevailing approach is too static and cognitivistic. If we consider how opinions about oneself may play a role in psychological functioning, it is quite clear that cognitive opinions alone do not indeed tell the whole story. For example, negative self-evaluations are important only if they concern topics that matter to us (Harter, 1999; James, 1890; Tesser, 1988). We could happily admit that we are very bad at playing tennis, as long as we do not care about our tennis performance. For someone else, the evaluation that he is not as good as the top-ten players in the world may be devastating. Thus, self-evaluations appear especially important if they are about our basic concerns, and, thus, are connected with emotions. The type of emotions involved may differ greatly. Take, for example, an ambitious player who is uncertain about his capacities. Baumeister, Smart, and Boden (1996) describe how people with a high but brittle self-esteem may act aggressively in situations that are perceived as threatening to their self-esteem. A certain player may present himself as a great player and become very angry if he loses a game, attributing the loss to anything but himself. Another may feel sad and see himself as a lousy player, after losing an important match. But even for the most ambitious tennis player, his evaluation of his tennis performance or his identity as a tennis player does not matter always and everywhere. In his relation with his intellectual mother, for instance, he might even be a bit embarrassed about his tennis reputation. Whether it matters, depends on the situation, has to do with the specific relationship involved. Thus, opinions about oneself are intertwined with emotions, arise in relations, and may be more or less stable. They emerge, become relevant, motivate, and change in the ever-changing relations. Focusing

Introduction

3

solely on non-contextual cognitive statements about oneself may be too narrow an approach. From the history of thinking about self and identity it becomes evident that the cognitive approach has not always predominated. Founding fathers of self and identity psychology, such as William James (1890) and Erik Erikson (1950, 1968), described much broader concepts. They perceived self and identity as dynamic phenomena, which include cognitions, but also emotions and perceptions, and which are always embedded in the person’s relationship with the context. The more restricted, cognitive focus became dominant afterwards, in all probability as a consequence of the demands of formalization and standardization since the time of behaviorism. Since then, self-related research has focused mainly on self-representations, often in written form. In identity research the structuralistic status approach of Marcia (e.g. Marcia, 1980) has dominated for more than thirty years. The recent chapter of Harter in the Handbook of Child Psychology (1998) clearly shows the dilemma. Harter begins by stating that emotions do play an important role in the self-concept, and that research should pay attention to them. However, the current knowledge in this field, as Harter notices, mainly concerns cognitive aspects. That is why her overview of the research and her further elaboration are cognitive too. But, things appear to be changing. Newer approaches to the study of self and identity have challenged aspects or implications of this cognitive approach and now focus on factors that have been hitherto neglected. Demo (1992), for example, criticized the notion that the self-concept is always stable. Identity researchers such as Bosma, Graafsma, Grotevant, and De Levita (1994) have begun to conceptualize identity as a process of ongoing adaptation and have made suggestions for a relational and interdisciplinary approach to identity development. Recent research efforts (e.g. Baumeister, 1998) started to investigate the role of emotions and situational influences on the development of self and identity. In a similar way many other authors have begun to address the limitations of a merely cognitive approach to self and identity (see chapter 2 for a comprehensive overview). Although these recent developments have resulted in new insights and have provided new perspectives, fundamental questions remain. For example, the role of emotions in self and identity requires further explanation. How do emotions and cognitions influence each other? What direction or directions does this influence take? In terms of contextual factors, what kinds of events tend to be relevant to an individual’s sense of self and why do the same events often impact on different individuals in different ways? Furthermore, we lack insight into the processes that can account for stability and change in

4


self-perception and self-evaluation over different time spans. What processes are responsible for long-term developmental change from infancy into adulthood? Is fundamental and enduring change possible in adulthood? If so, what mechanisms are responsible for such change? These fundamental questions remain largely unanswered. One major impediment for addressing them is the lack of a framework in which the role of emotions, the role of the context, and issues of stability and change in self and identity are considered in combination. The more recent approaches offer new perspectives, but it is unclear how they relate to each other. Moreover, a majority of these approaches are still firmly rooted in the conventional view of self and identity as a cognitive, internal structure. Emotions and context are primarily regarded as correlates of this structure or influences upon its expression. And for these structures increasingly complicated conceptualizations have been suggested (compare Higgins, 1987; Markus and Wurf, 1987). Recently, conceptualizations of self and identity as an internal, entitylike structure have been criticized more radically. Several theorists have begun to tackle self and identity from a completely different perspective. Instead of working with an established construct of ‘‘self’’ or ‘‘identity,’’ and trying to relate emotions and context to it, they start from the opposite direction: emotions and context are seen as formative conditions from which self and identity emerge in a self-organizational process. Fogel (1993), for example, considers relationships fundamental to all development and discusses how the self evolves in a social and emotional context. Hermans (1996) also takes relationships as a starting point and has argued for a ‘‘dialogical interchange’’ instead of an ‘‘information processing’’ perspective on the self. Lewis’ (1995) theory of personality development is based on principles of self-organization, and it describes how stable characteristics emerge from feedback and coupling in cognition–emotion interactions. Haviland (Haviland, Davidson, Ruetsch, Gebelt, and Lancelot, 1994) and Magai (Magai and McFadden, 1995) turn to emotional processes to explain identity and personality development. Most of these authors apply principles of non-linear dynamic systems theory, e.g. the notion of self-organization, to the study of developmental processes. Groundbreaking work in this area has recently been done by Van Geert (1991, 1994). These new and more radical perspectives provide promising insights that may serve to begin to answer many of the questions listed earlier. They also offer intriguing opportunities for the integration into a much broader theoretical framework based on self-organizational perspectives. The aim of this volume is the discussion and further elaboration of these new ideas. In fact, according to many of the authors in this volume, emotions – central in one’s self-experience – emerge, change,

Introduction

5

and dissipate in relationships as a dynamic, self-organizing system. The authors suggest conceptualizing self and identity as rooted in emotion, emerging in relationships, and developing as a dynamic, self-organizing system. Most of the authors of the chapters in this book share an active interest in a dynamic perspective on self and identity. In such a perspective the various elements – cognitions, emotions, and context – need no longer be considered separately, but can be viewed as a complex, interacting network. Processes within such a network may give rise to ‘‘self’’ phenomena: conscious and non-conscious, reflective and non-reflective self-experiences. If these experiences become stabilized in this network, and relevant for the person, they can be seen as part of the identity. Such a process approach has the advantage that the embeddedness in the context and the changeability are inherently given. This approach has yet to explain stability. Several chapters in this volume will explicitly focus on this topic. Here we wish to warn against a source of conceptual confusion. ‘‘Self-organization’’ in psychology can have two completely different meanings: it can refer (a) to the structure/organization of the self, and (b) to a very general process in which higher-order phenomena emerge from interacting lower-order components (Lewis and Granic, 1999a). The first meaning is most common in psychology, the second stems from Dynamic Systems Theory and could also be referred to as ‘‘auto-organization.’’ These two meanings have as much in common as ‘‘wine table’’ and ‘‘table wine’’ (Marc Lewis, personal communication, April 2000). Although most authors in this volume use ‘‘self-organization’’ in the sense of ‘‘autoorganization,’’ it is sometimes used in the sense of ‘‘the organization of the self.’’ A conceptualization of self and identity as rooted in emotion, emerging in relationships, and developing as a dynamic, self-organizing system has implications for the methods that are used for assessing them. Most current operationalizations of self and identity are based on the idea of a stable cognitive structure, most clearly exemplified by questionnaires and standardized interviews with questions concerning the person’s ideas and opinions about him- or herself (see Byrne, 1996, for an overview). More open methods, such as open essay questions and ‘‘Who are you?’’ interviews, are also common. In contrast to the standardized questionnaires and interviews, the more open methods allow the subjects to give their own self-description. Identitymeasures mostly concern semi-open or fully standardized interviews and questionnaires also (e.g. Marcia et al., 1993). The assumptions behind most of these instruments are that the construct to be measured is context-independent, stable, and mainly cognitive. Such instruments can hardly be of use

6


when self and identity are seen as dynamic, self-organizing systems. Measurement will therefore be a recurrent issue throughout this book. In some chapters alternative measurement methods will be presented that are more in line with the new conceptualizations. In the concluding chapter we will come back to this topic and discuss suggestions for methodology and methods that are more appropriate. Another recurrent issue concerns the terminology that is used with regard to self and identity. This terminology has always been a source of inconsistency and confusion (e.g. Bosma, 1995). As the reader of the different chapters will note, there is a huge diversity of concepts that all have to do with how people perceive themselves within their context. All these concepts can be classified in two main groups with very different theoretical backgrounds. The first group consists of a broad collection of ‘‘self’’ terms: self-concept, self-system, self-schema, the self, the Self, self-evaluation, self-esteem. James, Cooley, and Mead are seen as the founding fathers of this theoretical stream. Different ‘‘self’’ terms connote different theoretical backgrounds. The term ‘‘selfconcept’’ refers to a cognitive construct, while self-system definitions suggest broader and more dynamic connotations. The other group consists of ‘‘identity’’ terms, like ‘‘ego’’ or ‘‘self,’’ but the most frequently used term in this group is ‘‘identity’’ itself. Erikson (influenced by Freud as well as by James) is the founding father of this theoretical stream. In the ways these concepts are used (also in this volume) it is often not very clear what the differences are between self and identity. Regardless of the different and often completely separate theoretical backgrounds, however, there are good reasons to assume that, in fact, both groups of theories and concepts concern more or less the same empirical phenomena. This is quite evident when the focus is on selfconception problems and identity problems (Van der Werff, 1990; Van Halen, in preparation). For this reason ‘‘self’’ and ‘‘identity’’ will be used interchangeably in this book. The aim of this book is to discuss and elaborate a dynamic systems perspective of self, identity, and emotion. We think that the application of such a perspective can help to overcome certain limitations of the cognitive approach. It can help to integrate emotions and the context in the study of self and identity and it offers promising possibilities for the conceptualization of stability and change. To achieve this aim, the book is organized according to a certain ground plan. After this introduction the book continues with a contribution (chapter 2) about three problematic issues in recent self-concept theory and research. In this chapter, Van der Meulen will show that the perspectives from which the authors in this book approach ‘‘the self’’ are not merely applications of modern paradigms in psychology to a particular

Introduction

7

domain, ‘‘the self.’’ They can (also) be understood as a continuation and extension of recurring arguments within self-concept theory and research over the years in different branches of psychology. Van der Meulen discusses how the former conception of a stable cognitive self-concept has been challenged, and how the role of emotions and of context, and the issue of stability versus variability have become three major issues in the theoretical discussion. The overview makes clear that all these new ideas have resulted in a shattered picture with many loose ends at this moment. She concludes that till now the emotional, dynamic, and relational aspects of self-concept and identity have been addressed in isolation. We see the integration of the three issues in theory and research as one of the major challenges in the psychology of self and identity. The discussion of such an integration is the main aim of this volume. Chapter 2, thus, sets the agenda for the rest of the book. It is followed by chapters 3 and 4 focusing on emotions and dynamic systems theory in a general way, and by chapters 5, 6, 7, and 8 offering theory and research, in which the issues are integrated in a particular way. In chapter 3, Frijda discusses emotions. Emotions, according to Frijda’s componential emotion theory (1986), are always about something: they emerge in the person’s relationship with the world. In addition, emotions signal that one’s own person is at stake. Moreover, emotions can be conceptualized as fluid processes rather than structures or entities. Frijda addresses the nature of emotions and discusses how emotions are related to the self. He argues that emotions do not require a representation of self, because they include responses to perceived events in which the self is not explicitly appraised. However, from an early age emotions imply what William James called a notion of ‘‘I’’ – the center of experience and action. Van Geert (chapter 4) presents a general and non-technical introduction to some of the major concepts of dynamic systems theory and its application to developmental psychology. Various contributions in this book rely heavily on the theory of dynamic systems. Readers unfamiliar with either dynamic systems theory or its application to developmental psychology will probably find the proliferation of technical terms confusing rather than revealing. Because dynamic systems theory fits in naturally with the basic concepts and models of developmental psychology, it may serve as an adequate theoretical and methodological tool in the study of developmental phenomena. Van Geert’s main aim is to demonstrate that these concepts and models, exotic as they may seem, function as convenient and handy tools for conceptualizing and studying processes of development and of the development of self and identity in particular.

8


Chapter 5 is about the intrinsic relatedness of person and context, in other words, about a relational perspective of self and emotion. The premises in Fogel’s chapter are that psychological experience always implies a relation, and that this experience is always dynamic and changing. Still, people have a sense of uniqueness and permanence through time, because they experience their changing relationships according to different types of emotions. These emotions provide them with information about the self. The experience and the type of emotional process varies at different time scales. This chapter forms a bridge to the following chapters. Its main focus is on social-relational processes, but ideas about emotions, self, and the dynamics of change are incorporated as well. This chapter, therefore, also offers a recent approach which integrates the emotional, dynamic, and relational aspects of self and identity. The following three chapters each describe a specific approach to the dynamics of self, identity, and emotion and provide illustrations of and empirical support for the model by means of examples and case studies. Hermans and Hermans-Jansen (chapter 6) consider the relationship between self – viewed as multivoiced and dialogical – and emotion. They argue that during human development a person’s affective responses are increasingly influenced by the relationship the person has with him- or herself. In chapter 7, Haviland-Jones, Boulifard, and Magai discuss the function of emotions in identity. Both emotional experiences and identity show discontinuous and non-linear developmental changes. The authors assert that the application of dynamic systems modeling will give a better insight into these processes. In chapter 8, Lewis and Ferrari address the problem of the continuity of identity despite ongoing change in the person and the world. On the basis of principles of self-organization which can be observed in natural systems, they discuss personality and identity self-organization. Central in this process is the developmental consolidation of recursively interacting cognition–emotion elements. Each of the chapters 2 through 8 will be followed by a commentary, written by a distinguished scholar with regard to the topic addressed in the chapter. These scholars were asked, first, to comment on the ideas expressed in the chapter in the light of the aims of the volume, and, second, to give their own ideas about these aims. As a consequence of this twofold request some commentaries have the form of real discussions, while others mainly present perspectives in addition or complementary to the topic of the chapter and the book. Some authors have replied to the commentaries on their chapters. These replies are not included in the book, but their main points are discussed in the final chapter. They are referred to in the text and included in the list of

Introduction

9

references. As the combination of chapters and commentaries will show, the attempts to combine emotion theory, dynamic systems thinking, and recent developments in theory and research on self and identity are very new and still leave much room for debate. We do not want to give the impression that with this volume all of the problems in this domain have been solved. On the contrary, the critical commentaries of outstanding scholars on the various contributions to the book are meant to broaden the discussion beyond the work and ideas of the authors of the different chapters. By asking the commentators to discuss the ideas in the chapter, we hope to get an overview of discussion points, contradictory viewpoints, unsolved problems, and unanswered questions. In this sense, the commentators can be seen as a forum of highly specialized colleagues. Because the topic is so new, it is important to get an overview of the diversity of the many viewpoints, before boiling down our thinking about the topic into a comprehensive body of theoretical statements. In chapter 9, the final chapter of the book, we will discuss how emotion, dynamic systems thinking, and a relational perspective are related to each other in the different theoretical and empirical approaches to the study of self and identity development. The various authors all have their own focus and the chapter will elaborate how their approaches, in combination, offer a broad and lively picture of how self and identity can be seen as rooted in emotion, emerging in relationships, and developing as a dynamic, self-organizing system. The chapter will also discuss to what extent they complement each other, and to what extent contradictions exist between the different viewpoints. The resulting overview will present the current knowledge in this new field.

C HA P T E R 2

Developments in self-concept theory and research: affect, context, and variability Matty Van der Meulen Theory and research on constructs such as the self-concept, which have a considerable history in psychology, are inevitably influenced by prevailing scientific opinions and developments at a particular point in time (Baumeister, 1987; Logan, 1987). In the sixties and seventies the self-concept was a rather unproblematic construct, predominantly handled as a trait: a relatively stable, generalized, cognitive set or system of descriptive features, characteristic of a particular individual. The majority of methods still used to investigate the self-concept, mainly variations of self-esteem questionnaires, underline this view (Byrne, 1996; Wylie, 1989). This solid picture has, however, been questioned during the last two decades from different angles. Markus and Wurf’s (1987) proposal for a dynamic self-concept, in which situational influences are taken into account, has been very influential in this process. Not surprisingly, the focusing on situational aspects of the self-concept undermines its supposed stability. Furthermore, the strictly cognitivistic interpretation of the self-concept has been differentiated (Byrne, 1996; Damon and Hart, 1988; Greenwald and Pratkanis, 1984), most explicitly by Epstein (1993a). In this chapter the attention is focused on the foregoing three issues. The traditional conceptualization of the self-concept will be examined in the light of recent developments: (a) the cognitive view of the selfconcept will be confronted with the role of emotion, (b) the idea of a generalized construct faces the problem of how to take the context into account, and (c) the assumed stability of the self-concept has to stand up against observed situational and temporal variability of the selfconcept. These issues were originally highlighted by James (1890), Cooley (1902), and Mead (1934), the nestors of self-concept literature. The renewed interest in self-concept psychology on these issues can further be placed within the broader context of current psychological theorizing and empirical research. Emotion, context, and variability are topics of general interest in developmental, social, and personality psychology, as will be indicated below. 10

Developments in self-concept theory and research

11

These three issues are not the only ones in self-concept theory and research. Other recurring issues, beyond the scope of this chapter, are for instance: the self-as-subject versus the self-as-object, the global self-concept versus hierarchical multidimensional models, real versus ideal self-concept, true versus false selves, and the motivational or executive function of the ‘‘self’’ (Baumeister, 1998; Byrne, 1996; Harter, 1996, 1998; Marsh, Byrne, and Shavelson, 1992). At this point a description of the phenomenon we are dealing with in this chapter is called for. To avoid terminological complications and lengthy explanations, I will use the term ‘‘self-concept’’ to refer to ‘‘the set of beliefs a person has about him- or herself’’ (Harre´, 1987). Contrary to common practice, I will not use the terms self-concept and self-esteem interchangeably. The term ‘‘self-esteem’’ is confined to the evaluation of one’s entire person, that is global self-esteem, or of specific aspects or components of oneself. These terms are used here as collective terms, referring to constructs the characteristics of which will be dealt with in the rest of this chapter. (For definitions and self-concept models see Bracken, 1996; Brinthaupt and Lipka, 1992; Byrne, 1996; Damon and Hart, 1988; Greenwald and Pratkanis, 1984; Hattie, 1992; Oosterwegel, 1992; and chapter 1 of this volume.) In the following sections the question of affect, context, and variability in relation to the self-concept will be discussed separately. In the final section these three issues will be considered in combination and the implications for the conceptualization, measurement, and future of the self-concept will be examined. The place and role of emotions in the self-concept Although the cognitive foundation of the self-concept has not been a major subject of debate over the years, it has been present from the start. In the early years of self-psychology James (1890) and also Cooley (1902) included self-feelings as an important part of the ‘‘self.’’ According to James, self-feelings are aroused when one’s success is not in accordance with one’s pretensions. For Cooley self-feelings are the accompanying emotions when individuals imagine how they appear to others, and how others would judge that appearance. Self-feelings arise out of a desire for appreciation or an attack on one’s ‘‘me.’’ Thirty years later, however, Mead (1934) emphasized that ‘‘self’’ has to do with thinking in the first place and is therefore a cognitive construct, rather than an emotional phenomenon. Nowadays one finds a renewed interest in emotions in several domains of psychology, e.g. intelligence (Goleman, 1995). This interest is reflected for instance in Eisenberg’s (1998) introduction to the latest

12


edition of the Handbook of Child Psychology. She writes that emotion is one of the emerging themes in social and personality development, and therefore ‘‘an integral aspect of conceptions of the self’’ (p. 3). This is, however, not yet generally accepted. On the one hand, there is still a fair amount of agreement about the interpretation of the self-concept as a cognitive construct (Hattie, 1992; Marsh, Byrne, and Shavelson, 1992; Oosterwegel, 1992). On the other hand, several authors consider emotions and affect important factors in relation to the self-concept (e.g. Epstein, 1993a; Greenwald and Pratkanis, 1984). How emotions have been given a place and role in the self-concept will be worked out hereafter. The terms emotion and affect include different features of emotions: emotional states, emotional experiences, or emotional expressions (Lewis, 1993). Furthermore, emotions indicate an individual’s relationship with the environment and a readiness for action, in particular when matters of importance are at stake (Saarni, Mumme, and Campos, 1998; Fischer and Tangney, 1995). Affective responses to self-referent stimuli James and Cooley observed that one’s own person – one’s appearance, behavior, or possessions – is an object of special significance: one’s reflections in the mirror give rise to a peculiar interest, one’s children are unquestionably the prettiest and brightest. A number of studies provide empirical evidence for these everyday observations. When children have acquired objective self-awareness, from the second half of the second year onwards (Lewis and Brooks-Gunn, 1979), they show self-conscious emotional reactions: to their reflections in the mirror (Lewis, Sullivan, Stanger, and Weiss, 1989; Schneider-Rosen and Cicchetti, 1991), to failure (Kagan, 1981) and success (Bullock and Lu¨tkenhaus, 1990), and to overpraise and exposure in interaction situations (Lewis et al., 1989). When in the third year children are able to evaluate their own behavior in relation to rules and standards, they may experience self-conscious evaluative emotions like pride and shame (Lewis, 1993; Stipek, Recchia, and McClintic, 1992). That one’s own person is not a neutral object is also illustrated by people’s judgment of attributes which are not consciously or immediately recognized as referring to themselves. People, for example, evaluate distorted pictures of themselves, their handwriting, or recordings of their own voice less neutrally than comparable information of others (Huntley, 1940; Van der Werff, 1967; Wolff, 1943). Though these phenomena are intriguing, one of the problems in these studies is to present stimuli that are self-referent and at the same time sufficiently unrecognizable. This is less of a problem in a study by Hoorens and Nuttin


13

(1993), who found that letters contained in one’s own name are liked better than other letters. Thus the point is not whether emotions have to do with one’s own person or one’s self-concept, but rather what precisely their relationship to the self-concept is. Types of relationships between the self-concept and emotions The relationship between the self-concept and emotions has been given shape in different ways and degrees in the self-concept literature, running from relations of emotions with early self-awareness to a close association or intertwinement of cognition and emotion within the self-concept. (a) Self-awareness as a prerequisite for emotional experience People may not perceive their own emotions or may be unaware of their emotional states. Emotional experience, that is the meaning individuals attach to their emotional state and expressions, requires a particular cognitive ability, i.e. self-awareness, the capability of attending to oneself (Lewis, 1993). This capacity usually emerges in the second half of the second year (Kagan, 1981; Lewis and Brooks-Gunn, 1979) and is inferred from particular self-referential behavior, like pointing to self-reflections in a mirror. Lewis et al. (1989) found that embarrassment, a self-conscious emotion, occurred more in young children who showed signs of self-recognition. The capacity to reflect upon oneself is necessary to recognize one’s own emotions and also to communicate about them (Saarni, 1997). (b) Emotions as indicators of self-awareness Self-awareness is thus considered to be a prerequisite for emotional experience. Conversely, emotional expressions have been used as indications for the existence of self-awareness. In young children emotional reactions such as ‘‘outcome reactions’’ after completing a building task (Bullock and Lu¨tkenhaus, 1990) or signs of distress in reaction to the behavior of a model (Kagan, 1981) have been interpreted as such. (c) Emotions as a dimension of the self-concept Some recent multidimensional, hierarchical models of the self-concept contain dimensions such as Emotional Self-Concept, Emotional Stability, Self-Confidence, or Affect (Bracken, 1996; Hattie, 1992; Marsh and Hattie, 1996; Marsh et al., 1992). Such models have been developed and

14


extensively investigated through factor-analyzing self-description questionnaires. Moreover, classification systems of open, experiential methods, like the ‘‘Who are you?’’ interview, may have a content category Feelings and Emotions (l’Ecuyer, 1992). (d) Self-esteem as the affective component of the self-concept In the literature self-esteem is usually seen as representing the affective dimension of the self-concept (Damon and Hart, 1988; Keith and Bracken, 1996). In contrast to Emotional Self-Concept as one of the dimensions of the self-concept, self-esteem may concern various domains of one’s own person or one’s own person in general. This is often operationalized through separate subscales in self-esteem questionnaires: e.g. Scholastic Competence, Social Acceptance, Athletic Competence, Physical Appearance, Behavioral Conduct, and Global Self-Worth in Harter’s Self-Perception Profile for Children (Harter, 1985). (e) Causal relationships between self-concept and emotions Though not everyone considers emotions an integral part of the selfconcept, it is clear that certain evaluations or appraisals of one’s own person can trigger emotions. A first example is the kind of emotional reactions experienced after success or failure, e.g. in young children as described above (Stipek et al., 1992). Also adult self-conscious emotions like shame, guilt, embarrassment, and pride have recently become subjects of investigation (Tangney and Fischer, 1995). Guilt and pride arise, for instance, when one holds oneself responsible for wrongdoings or for socially approved outcomes respectively. When high but fragile self-esteem is threatened (Baumeister, Smart, and Boden, 1996), or in adult narcissism (Rhodewalt and Morf, 1998), emotional outbursts and behaviors may become quite violent and aggressive. A second type of relationship can be found in the literature on self-esteem, depression, and suicide. Depression and suicide are associated with chronically low self-esteem (Baumeister, 1993). Though global self-worth and depressed affect correlate highly, these are two distinct constructs that influence each other reciprocally. This occurs through different pathways, in which depressed affect may be blended with anger (Harter and Marold, 1989). Harter and Marold suggest that low self-worth precedes depressed affect when anger is directed towards oneself. Conversely, in cases where depressed affect precedes low self-worth anger is predominantly directed towards others. A third type of relationship concerns emotional problems that result from discrepancies between divergent perspectives on one’s own per-


15

son. In Higgins’ work specific patterns of discrepancies between ‘‘actual,’’ ‘‘ideal,’’ and ‘‘ought’’ self give rise to specific emotional reactions and also health problems (Higgins, 1987; Higgins, Vookles, and Tykocinski, 1992). (f) Emotions and cognitions intertwined within the self-concept A further step is to see the cognitive foundation of the self-concept closely associated or inextricably intertwined with emotions or affect. Epstein is the most explicit advocate of the association between the self-concept and affect. This has been worked out in his cognitive– experiential self-theory (CEST; e.g. 1973, 1991a, 1993a). In this theory two distinct systems of information processing are assumed, a rational system and an experiential system. The rational system operates according to analytic, logical rules, mediated by conscious appraisals of events, and therefore works more slowly. The experiential system processes on a more holistic, context-specific level, in an implicit way, therefore works more rapidly, and is particularly related to emotional experiences. With regard to the ‘‘self’’ as an object, both systems operate simultaneously, but independently. The rational system concerns cognitions about one’s own person that can be retrieved and described by regular self-report methods. The experiential system regards selfreferent cognitions derived from emotionally significant experiences, which may not always be explicit, but which nevertheless may have a significant influence on behavior. Self-referent information obtained from these two distinct systems may be discrepant, e.g. high reported self-esteem versus low self-esteem behavior. It is only in the experiential system that cognition and emotion within a person’s self-theory are strongly associated, in cognitive–affective units or modules. The cognitive–experiential self-theory has acted as an explanatory framework for several phenomena, for instance the acquisition and maintaining of negative self-schemata, and depressive realism (Epstein, 1992a; Pacini, Muir, and Epstein, 1998). In Greenwald and Pratkanis’ (1984) conception the association between cognition and affect goes further: both are part of the selfconcept. They consider ‘‘self’’ as the object of an attitude, more precisely, as an attitudinal schema. Such a schema consists of (a) a cognitive structure, (b) behavioral tendencies, and (c) affect, the central component of an attitude. Parallels between the findings in attitude research and self-concept research support this view. Summing up, several authors agree that emotions are important for the self-concept, but the divergent approaches above illustrate that so far theory and research provide no clear-cut standpoint about the nature of the relationship. Self-concept and emotions are predominantly

16


viewed and investigated as two constructs which may be related to each other, influence each other reciprocally, sometimes correlate highly, but which remain distinct. In Epstein’s self-theory emotions are associated with but not part of the experiential self-system, though the idea of cognitive–affective modules suggests a strong interweaving. An example of emotions as part of the self-concept is Greenwald and Pratkanis’ approach. In their attitudinal schema both cognition and affect are represented. Another candidate for this is the widespread view that self-esteem – global self-evaluation or evaluation of specific components – represents the affective dimension of the self-concept. However, this last case is disputable. It is a question whether or to what degree affect or emotions are involved in self-evaluation (Baumeister, 1998). James (1890) was the first to make a distinction between cold intellectual self-estimation and the passionate warmth of self-feeling. Evaluation refers, strictly speaking, to the cognitive component of an emotion, the appraisal (Frijda, 1986; Hattie, 1992; Oosterwegel, 1992), which may arouse emotions but is not emotional itself. Epstein and also Greenwald and Pratkanis base their theoretical expositions partly on information-processing models and research. This domain may provide further insight into the relationship between the self-concept and emotions. Processing of self-relevant material The question here is whether or how affect influences the processing of information about one’s own person. Studies by Kuiper and Rogers (Kuiper and Rogers, 1979; Rogers, Kuiper, and Kirker, 1977) and Markus and co-workers (Markus, 1977; Markus and Sentis, 1982) show that self-relevant information is selected, processed, and retrieved more easily than other information. Processes like that have been called ‘‘self-reference effects’’ (Greenwald and Pratkanis, 1984). Such effects can be attributed to four specific features of self-schemata (Markus and Sentis, 1982): compared to other cognitive structures, self-schemata are more extensive and complex, better connected, more frequently activated, and more associated with affect. The role of affect is thought to be the most important of these features, but precisely why and how this functions remains unclear. Epstein’s experiential system, in which affect has a prominent role, provides a possible explanation for the specific processing of self-relevant stimuli. One of his assumptions is that the experiential system processes information relatively rapidly and in an uncontrolled way, at a preconscious level of awareness. Epstein stresses that there are different levels of processing. Others point out that different channels of processing should be considered:


17

propositional/verbal, affective/emotional, behavior/motor (Ashmore and Ogilvie, 1992). In considering the processing of affect-loaded stimuli, the controversial issue about the primacy of affect or cognition comes up again. In discussing multiprocess models for emotional experience, Robinson (1998) concludes that what has primacy depends on the type of information. For information that is valenced and at the same time urgent, preattentive mechanisms lead to a rapid response, without conscious control, as in fear and anxiety. In controlled processing the role of valence is to direct the attention. Self-relevant stimuli are examples of valenced stimuli, which may give rise to both types of processing, depending on the urgency of the situation. The problem for empirical research, according to Robinson, is how to disentangle different types of processing, in particular when they seem to interact so highly. Basic concerns, values, and needs as instigators A first observation was that one’s own person is an object of special significance and affective responses. After that several ways and degrees of the relationship between the self-concept and emotions were described. Research and models of processing information about one’s own person were also briefly mentioned. A last point now is to find an explanation why people react affectively to aspects of their own person, to self-referent stimuli. The explanations vary in the terms that are used, but they all boil down to the assumption of one or more underlying basic concerns, values or needs, which have the function to protect or enhance selfregard or self-esteem (Allport, 1955; Epstein, 1991a; Frijda, 1986; Greenwald and Pratkanis, 1984; Kaplan, 1986; Sedikides and Strube, 1997). Positive emotions are associated with the approximation of such concerns, negative emotions with the threatening or thwarting of them. Such emotions signal a self-regarding attitude, which probably has an adaptive function (Baumeister, 1998; Greenwald and Pratkanis, 1984; Sedikides and Strube, 1997). Conclusion: the function of emotions for the self-concept Early self-psychologists such as James and Cooley explicitly stated that emotions, or self-feelings, were an important part of the ‘‘self.’’ In the following decades, however, emotions have not held a prominent place in mainstream self-psychology theory and research. Admittedly, several authors write that emotions are of importance and are involved in the self-concept, but theoretical elaborations of this idea or empirical

18


research into it have been scarce. Eisenberg (1998) announced that emotions are considered to be an integral aspect of the conceptions of self. Though agreeing with this, Harter (1996) admits in a historical overview that she still predominantly handles the self-concept as a cognitive construct. Despite the unclear status of emotions, people’s reactions towards self-relevant stimuli indicate that one’s own person is not a neutral object, but ‘‘hot’’ information accompanied by affect. This has its effects in the attention for and the processing of self-relevant material. How exactly this works needs further explication and investigation. The most elaborate ideas on this subject can be found in the work of Epstein (e.g. 1991a, 1993a) and the literature on self-conscious emotions (Tangney and Fischer, 1995). Several functions are assigned to the self-concept: e.g. the processing of information about oneself, the motivation and guidance of behavior (Harter, 1998; Higgins, 1996), the integration of developmental processes, such as physiological and cognitive functioning (Baltes, Lindenberger, and Staudinger, 1998). Clarifying the place and role of emotions may possibly specify and extend these functions of the selfconcept. Emotions, for instance, may indicate what aspects of one’s own person and what situational elements are important for an individual (Saarni et al., 1998; Fischer and Tangney, 1995). What is desirable in the discussions around this issue is a more extensive treatment of recent emotion theory and research than has been the case so far in selfconcept literature. The significance of context for the self-concept The significance of context in the emergence and development of the self-concept has been recognized from the earliest psychological involvement with this subject by the classic authors James, Cooley, and Mead. Notwithstanding the importance assigned to context, the selfconcept was for a long time considered to be a fairly context-free phenomenon, consisting of a set of descriptive features which applied to one’s own person in general, across situations. This view of the self-concept has been challenged lately under the influence of a reassessment of the relationship between context and development in general, among others things. The nature of context During the last two decades theory and research on the self-concept have been directed more to the influence of the situation, environment, or context. Emphasis on the role of context in psychological functioning


19

is not limited to the self-concept literature. In developmental psychology, for instance, the environment within which individuals grow up and live, are ‘‘embedded in’’ so to speak, has always been considered an important factor (see Magnusson and Stattin, 1998; Moen, Elder, and Lu¨scher, 1995 for overviews of the role of context in psychological research and developmental psychology). Context is a global construct, which needs further specification. In the first place, different levels of context can be distinguished. Bronfenbrenner and Crouter (1983) provide a taxonomy containing four hierarchical context systems: microsystems (single settings), mesosystems (systems of microsystems), exosystems (larger social structures), and macrosystems (society at large). Second, context can be viewed either as a relatively stable variable, exerting a constant influence, like socio-economic status, or as a changing variable during the course of development. In the latter case, Kindermann and Skinner (1992) distinguish three different models: (1) one-way influences: contexts which themselves change, but which are not influenced by the subjects under study, e.g. birth of a sibling; (2) reciprocal influences: context and subjects induce change in each other through feedback mechanisms, but the change in context is not particularly tailored to the subjects, e.g. mutual socialization in peer groups; and (3) adjusted contextual change: changes in the context are attuned to developmental changes in the subjects, e.g. parental childrearing goals and strategies. Third, for a clear understanding of the role of the context, the processes through which the context influences the behavior under study (and vice versa) need to be explained. What elements of the context are relevant to a particular person at a particular time? What seems to be most important is not the objectively observable context, i.e. the actual context, but the meaning context has for the individual, i.e. the perceived context. In that case, context becomes a psychological variable, part of the characteristics of the individual and less a separately measurable variable (Magnusson and Stattin, 1998; Sansone and Berg, 1993). In such interactional subject–context models subjects are no passive receivers of contextual influence, but active agents participating in their own development (e.g. Brandtsta¨dter, 1998). Types of relationship between self-concept and context In the self-concept literature context has been assigned a twofold role: (1) context plays an important part in the emergence and development of the self-concept, and (2) context determines or influences the current self-concept, the self-concept at a particular time and place.

20


There exists considerable agreement that context, at the level of interactions with the primary carers, is a prerequisite for the emergence of self-awareness in infants (Lewis and Brooks-Gunn, 1979; Pipp, 1993; Trevarthen, 1993). In the further development of the self-concept, significant others – parents, peers, teachers, partners, or colleagues – play a role through their feedback in direct interactions or, more indirectly, through social comparison (e.g. Demo and Savin-Williams, 1992; Feiring and Taska, 1996; Harter, 1996, 1998). In social comparison the social groups to which persons belong or compare themselves with, the reference groups, are particularly important (Juhasz, 1992). Over the life span so-called normative contextual transitions, such as entry into school, the transition from school to work, military service, marriage, or a first child (Gecas and Mortimer, 1987; Leifer, 1980; Mummendey, 1988), may affect individuals’ self-concept. In addition to that, stressful life events or otherwise important turning points may exert their influence (Clausen, 1995; Prout and Prout, 1996). Throughout these contextual influences – in infant–carer interactions, socialization practices, peer interactions, normative life transitions, or stressful life events – culture resonates. Culture may be classified as a macrosystem (Bronfenbrenner and Crouter, 1983), but it manifests itself, gets its meaning, in concrete practices and tools, and in interactions between people (Goodnow, 1996), especially in the relationships between infants and carers (Fogel, 1993). Emerging through interactions, the self-concept incorporates the meaning a cultural community attaches to it (Schweder, Goodnow, Hatano, LeVine, Markus, and Miller, 1998). Cultures may differ in the way the self-concept is perceived, but a preliminary question is whether the self-concept is a universal phenomenon. Two aspects seem to exist across cultures: a sense of self as physically distinct from others and a sense of continuity over time (Hart and Edelstein, 1992). Other aspects may differ considerably between cultures. A salient distinction is one between cultures in which the self-concept is individualistic or independent versus cultures in which the self-concept is relational or interdependent (Markus and Kitayama, 1991; Schweder et al. 1998; Triandis, 1989). This distinction has important consequences for other aspects of the self-concept, like the perception, processing, and retrieval of information about one’s own person, the nature of self-conscious emotions, and motives of self-preservation or self-enhancement. In social psychology the interest in context concerns the specific set of elements of the self-concept, which becomes actualized in a particular situation at a particular moment. Markus and Wurf (1987) have introduced the term ‘‘working self-concept’’ to refer to this. This notion of a


21

working self-concept can explain why persons may have different self-concepts in different situations. An example of empirical evidence for this can be found in the work of Ashmore and Ogilvie (1992). They studied variations in self-concept, in what they call self-with-other representations. With a specific clustering method (HICLAS; see De Boeck and Rosenberg, 1988), clusters of self-ratings on a set of features in combination with certain target persons can be obtained. To get away from the idea that one’s self-concept is in constant flux, as Gergen represents it (1984), Markus and Wurf (1987) and others (Damon and Hart, 1988; Hattie, 1992) distinguish ‘‘core’’ aspects of the self-concept, which are relatively stable over situations, and more peripheral aspects, which are tied to specific contexts. To the core aspects could belong factual information, such as ascribed characteristics, major roles and memberships, and important or salient self-conceptions (Markus and Nurius, 1986). Authors such as Fogel (1993) and Hermans and Kempen (1993) also search for some kind of unity or synthesis. For Fogel ‘‘self’’ is something that is constantly being created in dialogue. He tries to find a kind of unity or cohesion, not in core aspects, but in common informational themes that arise across separate dialogical relationships. Hermans and Kempen assign to the ‘‘self’’ the function of synthesizing the self as a whole, to counterbalance for parts of the self which become dominant in particular situations. The term working self-concept has found its equivalence applied to the context in ‘‘working context.’’ Sansone and Berg (1993) use this term to make clear that aspects of the context influence a person’s behavior or activity in so far as they have a specific meaning for that particular person. In Sansone and Berg’s view there exists a reciprocal influencing: elements of the context which are relevant to the individual define the working context, but the working context in turn defines which elements become part of the working self-concept. Together they form what Sansone and Berg call ‘‘the activated life space.’’ Which elements of the context are relevant at a certain moment may be inferred sometimes from a person’s emotional reactions (cf. Epstein, 1991a; Fischer and Tangney, 1995; Robinson, 1998; Saarni, 1997). The notion of a working self-concept has consequences for the level of self-esteem. The working self-concept consists of a variable, contextspecific set of elements, each of which can be valued differently by a person. Therefore the evaluation of such sets may change from one situation to another. Research with the Experience Sampling Method, developed by Csikszentmihalyi, nicely illustrates this. With this method subjects carry a beeper and have to fill in self-report questionnaires at randomly signaled moments several times a day for some weeks. Using this method, Wells (1992) found that young professional

22


women’s current on-going self-esteem was higher when they were at work or engaged in leisure activities than when they were at home taking care of their children. A similar type of study with this method was carried out with adolescents to investigate how their self-esteem was influenced by physical setting, activities they were engaged in or persons they interacted with (Jaquish and Savin-Williams, 1981; Savin-Williams and Demo, 1983). Conclusion: the dynamic interaction between the self-concept and context Interactional or transactional models of individual and context relationship (Magnusson and Stattin, 1998; Sameroff and Fiese, 1990) are widely accepted, also for a phenomenon like the self-concept (Stratton, 1988). In line with this, the self-concept is often viewed as a dynamic system (e.g. Baltes et al., 1998; Hermans and Kempen, 1993; Markus and Wurf, 1987; Oosterwegel, 1992). Exactly how such a system functions, and what kind of interchanges between individual and context take place, can be worked out very differently. This depends, first, on one’s general conceptualization of the selfconcept, or the aspects of the self-concept under study. For the interpersonal self-concept (Fogel, 1993; Neisser, 1993), the interchange occurs very directly, through mother-and-child attunement. In the working self-concept the core aspects are more or less stable over situations. The interchange between context and self-concept in the first place concerns the more peripheral aspects. For multidimensional, hierarchical models (Hattie 1992; Marsh and Hattie, 1996), context exerts the most influence on concrete, situational self-concept aspects at the bottom level of the model. Here the system is most flexible or, so to speak, ‘‘dynamic.’’ Second, the level of context and whether context itself is considered to be fixed or changeable, both characterize this interchange. At a microlevel, the interchange between context and an individual’s selfconcept seems to occur directly in child–carer interactions. In the relationship between working self-concept and working context both are open systems which reciprocally constitute each other. At the macrolevel, culture seems to be a variable to be reckoned with. Culture must not be viewed here as a kind of abstract variable exerting one-way influence on the self-concept, but as something which is expressed in interactions at every contextual level and, in turn, is defined by these. Third, the relationship between context and individual is also typified by what is precisely meant by the term ‘‘dynamic.’’ The term ‘‘dynamic’’ may be used in a loose way, meaning malleable or flexible as opposed to stable or fixed. It refers in particular to processes bringing about change. For Markus and Wurf (1987), the self-concept is dynamic


23

in the sense that different sets of self-representations become active in different situations. Moreover, dynamic also points to the agentive features of these self-representations or self-structures. In that case dynamic means the function of self-structures in processing input and output of information and behavior. For Oosterwegel (1992), the term dynamic refers to the processes of the organization of the self-concept or, in her terminology, the self-system. With this she means the way in which various aspects of the self-concept, and divergent perspectives one can take towards it, become differentiated or integrated in order to maintain a kind of balance. Dynamic is a term that is also used more specifically, in relation to dynamic systems theory. In order to understand the complex functioning of the person–environment system in general, Magnusson and Stattin (1998) advocate a holistic viewpoint and a dynamic systems approach. Hermans and Kempen (1993) use dynamic systems theory to clarify the function of the synthesizing ‘‘self.’’ An advantage of this theory is that dynamic systems are not closed, but open to their environment. In addition, the principle of self-organization, the iterative feedback of the system’s output into itself, might explain how synthesizing processes between the multiple positions one can take towards oneself could work. The relationship between the self-concept and context has been given shape very differently in the course of time. Instead of two separate variables that somehow correlate, this relationship is nowadays viewed as a dynamic interaction between two interdependent systems. This nevertheless leaves several questions to be answered. What aspects of the self-concept relate to what levels of the context, how do processes of interaction precisely work theoretically, and what empirical evidence can be found for that? Principles of dynamic systems theory may be applied heuristically, to form ideas about this, or empirically, to test specific models. It is obvious that suitable methods and measurements are needed to investigate this. Stability and variability within the self-concept As has been shown by the previous section, an interactional individual– context view has certain implications for the stability and variability of the self-concept. Stability and variability of personality characteristics, the difference between traits and states, is an old question, which is a topical subject again these days (e.g. Baltes et al., 1998). James made a distinction between stable self-feeling, the average overtone independent of objective reason, and self-feelings as a result of one’s actual success or failure. Several authors agree now that both stability and

24


variability characterize the self-concept (Brinthaupt and Lipka, 1992; Demo, 1992; Markus and Wurf, 1987). Others restrict the use of the term self-concept to enduring, relatively stable dimensions (Marsh et al., 1992). The prevailing viewpoint on the construct thus sets limits to the very possibility of stability and variability of the self-concept beforehand. In speaking about stability or variability, it is important to make a distinction between the self-concept and self-esteem. A second point to take into account is the time-perspective. It makes a difference whether short-term fluctuation or variability is meant or long-term developmental change. Life-span developmental change of the self-concept The development of the self-concept has been described with some frequency for the period up to and including adolescence or for the entire life span (e.g. Baltes et al., 1998; Damon and Hart, 1988; Demo, 1992; Harter, 1983, 1998; Hattie, 1992). Inherent to a life-span perspective is a focus on changes which occur over relatively long periods of time and which concern traitlike characteristics (Nesselroade, 1991). Self-concept studies covering several age periods usually concern the changing content or dimensions of the self-concept, but also more structural aspects, like the organization of self-concept dimensions. Recently more attention is requested for processes of change. A processual view on change does not consider the self-concept to unfold itself in isolation, but in person–environment interaction, as discussed in the previous section. The number and nature of self-concept dimensions change over the years. Young children’s self-conceptions concern a few concrete observable dimensions, like abilities, activities, and possessions, while adolescents’ self-conceptions refer to more abstract dimensions, like personality characteristics and attributes related to various roles (Harter, 1998). Within certain age periods the main dimensions are considered to remain relatively stable. In the conceptualization of the working self-concept (Markus and Wurf, 1987) more enduring change concerns the entire set or universe of self-conceptions, which form the basis for the working self-concept. The meaning of existent self-conceptions may change or new elements may be added to the set. Within the working self-concept certain ‘‘core’’ elements may possess a certain cross-situational and temporal stability. In addition to change in content or dimensions, change in structural aspects of the self-concept has been investigated. Structure may refer to characteristics of separate dimensions, like abstractness and centrality,


25

as well as the relationship between dimensions, like consistency and differentiation (Bailey, 1970). Studies of structural aspects are scarce. Damon and Hart (1988) studied the level of self-understanding in several age groups. Oosterwegel (1992) investigated differentiation and discrepancies between several perspectives on the self-concept – own real and ideal perspective and that of others – in children of six to sixteen years old. Structural aspects may change in age periods where developmental transitions occur, but also remain stable for some time. Strauman (1996), for instance, found that the content of self-conceptions of students changed over the period of three years, but that discrepancies between their self-conceptions remained stable. Changes in content and structure of the self-concept are related to cognitive development and life transitions in the social environment of the individual. It is a question which of the two factors exerts more influence on self-concept change. Hattie and Marsh (1996) find that events people are confronted with in the course of their lives are more influential than cognitive development. Demo (1992), on the contrary, thinks that developmental changes outdo variability due to situational differences. Baltes et al. (1998) present a model of interacting sources, related to age, history, and non-normative occurrences, providing opportunities and constraints influencing the development of ‘‘self and personality.’’ Such a model makes clear that the answer to the question of which factors influence the development of the self-concept requires ingenious research designs and at least a study of complex processes. Short-term situational variability of the self-concept In the section on context, the variability of the self-concept related to different roles and situations has already been discussed. What is variable in the idea of the working self-concept is the set of selfconceptions that is activated in particular situations. Within this set some elements recur quite often, i.e. the stable core aspects, others are more tied to the specific situation or context. Marsh and others (Hattie, 1992; Hattie and Marsh, 1996; Marsh et al., 1992) in the first place stress the stability of the self-concept. The hierarchical models they propose, however, leave open the possibility of variability. This variability occurs at the bottom of the hierarchy, where concrete aspects of the self-concept are placed. The more abstract dimensions at the top are thought to be relatively stable. L’Ecuyer (1992) also presents a hierarchical model, based on theoretical considerations and the coding of self-descriptive interviews with subjects over the entire life span. In this model, hierarchy is represented

26


as increasing abstraction, starting with daily experiences that are directly felt, then perceived, and finally conceptualized. The daily experiences may be manifold, but in l’Ecuyer’s view all experiences can ultimately be classified in five principal dimensions – Material Self, Personal Self, Adaptive Self, Social Self, and Self/Nonself – which remain the same for all age groups. In the above views variability is one of the characteristics of the self-concept. Variability concerns peripheral, concrete, directly experienced aspects, which are contextually grounded. Core aspects and abstract dimensions refer to stability and continuity over time and thus are the aspects that are usually studied over longer time periods. Variability, however, also offers possibilities to be studied in a life-span perspective. Certain age trajectories, like adolescence, may show more variability than others. Self-esteem: fluctuations along a baseline Self-esteem has been defined in this chapter as the evaluation of one’s entire person or of specific components of oneself. It is usually measured by self-report questionnaires, several of which now possess satisfactorily theoretical and psychometric qualities (see for overviews Byrne, 1996; Hattie, 1992; Keith and Bracken, 1996; Wylie, 1989). These instruments ask for an opinion about one’s own person in general, one’s typical self-esteem, and focus by their very nature on stability over time and place. Though stability is characteristic of self-esteem investigated with these types of instruments, slight changes can be found over the life span. Overviews of research in this area (Baltes et al., 1998; Damon and Hart, 1988; Demo, 1992; Harter, 1998) report a decline and a subsequent recovery in global self-worth in middle childhood. The decline probably is due to the emergence of a more realistic view of children after the tendency in early childhood to overrate themselves: ‘‘I’m the biggest and strongest boy of my class’’ (Van der Meulen, 1987). In early adolescence again a slight decline can be noticed, possibly as a result of changes in school settings and other social settings at that age. From then on a steady increase occurs until late adulthood. In the elderly self-esteem may drop somewhat again. Evidence from several studies in the last few years confirms what most people experience in their daily life, namely that one’s current self-esteem is not always the same, but has its ups and downs over the day or week. An explanation for this has been given above. The working self-concept is constituted of a changing array of self-conceptions that are evaluated differently and this has its effect on current, global


27

self-esteem. Current self-esteem is often measured with well-known self-esteem questionnaires, like Harter’s Self-Perception Profile or Rosenberg’s Self-Esteem Scale, with the instruction to answer how one evaluates oneself right now (Kernis and Waschull, 1995; Wells, 1992). The findings of relative stability over several years and short-time situational variability have been integrated into a picture of a baseline level of self-esteem with moderate fluctuations around it (Demo, 1992; Kernis and Waschull, 1995; Rosenberg, 1986; Wells, 1992). This picture, however, cannot be generalized right away, but is probably true for only part of the people. Demo and Savin-Williams (1992) found high stability over a period of three years in adolescents for three dimensions: self-reported self-esteem, self-esteem reported by peers, and self-feelings. Over a period of a week, about one-half of the adolescents could be classified as stable, the other half showing moderate to high instability. This variability may be partly a characteristic of adolescence. It is possible that in older age groups, when relations and situations have become more stabilized, large fluctuations in selfesteem decrease. The interaction between stability and level of self-esteem For a long time people with low self-esteem seemed to form the problematic group in self-esteem literature (Baumeister, 1993). There exists a large body of research about the relationship between psychological and clinical problems or membership of a problematic group and level of self-esteem. The results of this type of research are not unequivocal in the direction of expected low self-esteem. Why some people’s selfesteem seems not to be influenced by adverse circumstances and others’ by seemingly trivial occurrences cannot be satisfactorily explained by only considering level of esteem. Kernis and associates found that psychological processes, serving self-preservation or self-enhancement, appear to be different for people who vary not only in level but also in stability of self-esteem. Level and stability of self-esteem hardly correlate and can be treated as two distinct dimensions. In this, level of self-esteem refers to people’s relative stable baseline self-esteem and (in)stability to the short-term, contextual fluctuations around this baseline (Kernis, 1993; Kernis and Waschull, 1995; Kernis, Grannemann, and Barclay, 1989). The attention of Kernis and associates in the first place focuses on reactions of people with unstable self-esteem. People with high but unstable self-esteem seem to depend on positive evaluations by themselves and others. Furthermore, they may react defensively and emotionally, with anger, hostility, and aggression, when their fragile

28


self-esteem is threatened (see also Baumeister et al., 1996; Rhodewalt and Morf, 1998). In contrast, people with stable high self-esteem need no validation for this by contingent, external feedback. They possess authentic or ‘‘true’’ positive self-regard, which is not easily altered by positive or negative events. The role of instability for low self-esteem is less clear. People with low self-esteem may incidentally experience positive self-esteem, but probably lack the skills to refute negative information about themselves. Low and stable self-esteem seems to be associated with a chronic dislike of oneself and a resistance to use self-protective or self-enhancement strategies. Conclusion: stability and variability as dynamically interrelated states In contemporary viewpoints, both the self-concept and its evaluative equivalent self-esteem can be characterized by stability as well as variability. For the self-concept, stability refers to certain main dimensions or to the entire collection of possible self-conceptions. These are assumed to remain relatively stable during a particular period in life, but may also show changes along the life course. Variability within the self-concept concerns concrete, contextual elements or different sets of elements, resulting from changing roles and contexts. These two characteristics, stability and variability, at first sight seem to exist relatively independently of each other. Yet, the distinction is one of degree rather than of type. What seem to be relatively stable core elements of one’s self-concept at a certain time and place may be irrelevant in another and vice versa. New experiences and insights into oneself may add new elements to the open set of beliefs about oneself, while others disappear. At a particular turning point this may regroup existing dimensions and also influence the structure between dimensions. Instead of being fixed, the self-concept can be imagined as a continuously shrinking, expanding, and reorganizing collection of selfconceptions. With regard to self-esteem, stability concerns the more or less stable baseline of general or typical self-esteem. Overall, this baseline selfesteem may show slight changes – decreases or increases – along the life span. Variability in self-esteem is expressed in fluctuations along the baseline, the current or barometric self-esteem, due to differences in self-evaluation in different relations, contexts, or through particular positive or negative events. Though correlations between baseline and current self-esteem are usually moderate to low, fluctuations may induce change in a relatively


29

stable pattern. This may happen through frequently occurring large fluctuations, but originally small changes may also get a process of change going. On the other hand, a pattern of more or less stable self-esteem may set limits to variability. People with a stable high self-esteem, for instance, have a ceiling for upward peaks and, similarly, people with stable low self-esteem have a bottom for peaks going downward. In sum, stability and variability in the self-concept and self-esteem should be conceptualized rather as dynamically interrelated states of ongoing processes of self-concepting and self-evaluating than as two independent states with their own internal laws. What processes exactly take place, how stability may evolve out of variability, and at what point variability moves towards stability needs closer inspection. Conclusions and implications The traitlike, cognitivistic self-concept of the sixties and seventies has known several revisions and reconceptualizations over the years (e.g. Brinthaupt and Lipka, 1992; Damon and Hart, 1988; Epstein, 1973; Gorden and Gergen, 1968; Greenwald and Pratkanis, 1984; Markus and Wurf, 1987), but so far this has not resulted in a generally accepted circumscription. Regarding the issues discussed in this chapter, the following general conclusions can be drawn. The set of beliefs a person has about him- or herself (a) is not an exclusively cognitive matter, but something that is in several ways closely associated with affect, (b) does not (solely) concern general, decontextualized beliefs, but beliefs that emerge in close interaction with the physical, social, and cultural context, and (c) consists of both more or less enduring, stable beliefs as well as more short-term, variable ones. Several problems and questions, however, remain inconclusive and need further investigation. First, the precise place and role of emotions in the self-concept are still unclear. Are emotions excluded or to be included in the self-concept? Which role do emotions play in the processing of self-relevant information? Which functions do emotions have for the self-concept? Second, the relationship between context and the self-concept should be worked out more thoroughly. What is precisely meant by dynamic interaction between context and the self-concept and how can this be investigated empirically? Third, how exactly should the relationship between stability and variability of the selfconcept and self-esteem be represented? How does stability emerge out of variability and when does variability tip over to stability? So far affect, context, and variability have been treated separately. These issues are, however, related to each other in several ways. An

30


integrated approach could be beneficial in dealing with the above questions. The interplay between affect, context, and variability After considering stability and variability of the self-concept and selfesteem, these three issues turn out to be interrelated. In the variability of the working self-concept and in fluctuations in current self-esteem, context plays a prominent role. What elements of the context become important at a particular moment depends on the interaction between context and individual in the ‘‘activated life space’’ (Sansone and Berg, 1993). One of the functions of emotions is to provide insight into what situations are relevant and what is valenced information (Epstein, 1991a; Fischer and Tangney, 1995; Robinson, 1998; Saarni et al., 1998). In particular, those elements and events come to the fore that potentially facilitate or threaten individuals’ need to preserve or enhance their self-esteem. A further integration of these three issues in theory and research may contribute to the development of self-concept psychology. A shift from content and structural aspects to dynamic processes To understand the interaction of emotion, context, and variability for the self-concept the focus should be on processes rather than on the outcomes of these processes. Processes can be characterized as ‘‘a continuous flow of interrelated, interdependent events’’ (Magnusson and Stattin, 1998: 699). This means that time is a fundamental element in the study of processes. To grasp the complex interactional processes of emotions, context, and behavior (Saarni et al. 1998) or those of person and environment (Magnusson and Stattin, 1998), the key word is ‘‘dynamic,’’ in particular a dynamic systems approach. Several authors argue that this approach should also be applied to the study of the self-concept (e.g. Baltes et al., 1998; Hattie and Marsh, 1996; Hermans and Kempen, 1993). A need for adequate measuring methods In accounting for affect, context, and variability and focusing on processes rather than content and structure, other measuring methods are required than the general self-report questionnaires usually used in self-concept research. Several authors realize this and plead for the use of more idiographic and qualitative methods, like autobiographies and case studies (Demo and Savin-Williams, 1992; Hattie and Marsh, 1996).


31

Epstein (1991a) suggests that we should observe situations in which persons react emotionally. Robinson (1998) and Izard (1992) suggest that facial expressions and physiological indicators should supplement self-report instruments. Markus and Kunda (1986) conclude that the variability of the self-concept should be investigated with methods that clarify the entire behavior repertoire that is involved in selfcategorization, self-defining etc. Reaction times of behavior in response to self-relevant stimuli may be one of these methods. The Experience Sampling Method, devised by Csikszentmihalyi and used in several studies (Jaquish and Savin-Williams, 1981; Kernis and Waschull, 1995; Savin-Williams and Demo, 1983; Wells, 1992) to investigate variability in level of current self-esteem in different contexts and in relation to affect, is an example of a fruitful procedure in this respect. Prospects for the self-concept Judging from recent overviews (e.g. Baltes et al., 1998; Baumeister, 1998; Harter, 1998) and the abundance of articles on empirical research, the self-concept (or ‘‘self’’ as noun or prefix) is alive and well. Two things are important for the prospects of the self-concept: terminological clarity and keeping on track with developments within psychology in general. Self-psychology is notorious for its terminological confusion and diversity. The term ‘‘self-concept’’ has been used throughout this chapter as a comparatively general and neutral term to cover most of the field. New conceptualizations of a construct often lead to the introduction of new terms. Some authors have substituted the term ‘‘selfconcept’’ for terms which take into account the hierarchical complex structure (‘‘self-system’’; e.g. Oosterwegel, 1992) or active, motivational aspects and flexible adaptation to altering information (‘‘self-schema’’ and ‘‘self-theory’’; e.g. Epstein, 1973; Markus and Sentis, 1982). The term ‘‘self’’ is also used quite often to include both objective and subjective aspects of selfhood, product as well as process (Baumeister, 1998; Harre´, 1998). This may lead to confusion about what exactly is meant and to reification. Since it appears difficult to arrive at some sort of agreement in this area, the best solution for the present is, first, to describe carefully the phenomena under study: specific aspects of people’s beliefs about themselves, their current or typical evaluations of themselves, their self-preserving or self-enhancing behavior, or the selection and processing of self-relevant information etc. Second, in referring to such phenomena a relatively neutral term can be applied, such as the ‘‘self-concept’’ or ‘‘self-concepting,’’ or ‘‘self’’ can be used as a prefix, as in ‘‘self-relevant information.’’

32


The connection of self-concept theory and research with current issues in psychology at large may contribute to preserving its role in the understanding of human behavior. A focus on the functions of the self-concept, like the processing of information, the motivation and guidance of behavior, or the integration of developmental processes (Baltes et al., 1998; Harter, 1998; Higgins, 1996), could further that understanding. A dynamic systems approach, which incorporates affect, context, and variability, may provide heuristic models for this as well as theoretical models which should be tested empirically.

COM M E N T A R Y

The self-concept is dead, long live . . . which construct or process? Differentiation and organization of self-related theories Annerieke Oosterwegel

In her chapter Van der Meulen organizes both theory and research on the ‘‘self-concept’’ around three key themes: cognition versus affect, interactions between self and context, and something termed ‘‘variability,’’ which can perhaps be characterized as a function of the former two plus motivation. This structure offers an excellent framework for the chapters that follow, where the specifics of these issues are taken up in some detail. At the same time, it appears that both the structure and the global nature of Van der Meulen’s account illustrate the major tension in contemporary ‘‘self-concept’’ research: important aspects are studied in isolation, and yet few specifics are known. Given these themes, and all the relevant issues which surround them, how do we combine them into an integrated whole, whilst also seeking to specify the precise nature of the mechanisms involved? This is the task which the current volume invites us to embark upon and for which its authors ‘‘set the stage.’’ Van der Meulen provides us with a useful listing of key issues which prevents us from losing sight of the broader picture as we immerse ourselves in detailed discussions of specific aspects. Here, I wish to go one step further and tentatively suggest an initial framework, based on the argument that we need (a) clear definitions that distinguish between the ‘‘self-as-object’’ and the ‘‘self-as-subject’’; (b) scope for several types of interactions between cognition and affect; and (c) clarity concerning what we understand as constituting ‘‘context’’ and where we position its boundaries. The not-so-omnipotent self-concept Van der Meulen makes a plea for clearer definitions of terms. The need for such clarity cannot be stressed enough. I would like to be even more rigorous, and, for the moment, dismiss any pretension of an 33

34


overarching label such as ‘‘The Self,’’ or Van der Meulen’s ‘‘the selfconcept.’’ It would be clearer to refer to each term with the prefix ‘‘self-’’ as a separate construct until we know how they are related. With respect to definition, the term ‘‘Self’’ refers to the person as a whole, taken from that person’s own perspective. In other words, the Self refers to the ‘‘I’’ of the individual, the acting entity of which affective, biological, neurological, and cognitive processes are part. Seen in these terms, however, studying the Self comes close to studying psychology; a statement that may be true, but does not afford much explanatory power. The self-concept, on the other hand, is nothing more than a sample of cognitive self-representations, a cross-section in a process of personal construction, one frame in a movie. Affect and contexts interact with self-evaluation and self-regulation, but only become a part of a self-concept once the individual has reflected on that affective process or interaction with the context, and has found it to be self-descriptive. Between the self and a self-concept lies the whole range of constructs we tend to study. Some of them, like self-regulation or self-evaluation, are conceived more actively and dynamically, others (fortunately fewer and fewer) are understood more statically. These constructs appear to be inextricably intertwined with context and affect, but again (as Frijda, this volume, argues) do not necessarily coincide. Too often and too early, in my opinion, we try to encompass different aspects and mechanisms into one omnipotent construct. One might say that self–other differentiation is the first step in healthy self-research development. It is not only important to be clear about what we study, but also to be aware of the aspects we do not cover in any given piece of research. Differentiation among self-related theoretical constructs becomes especially important if one takes a dynamic systems approach to research – as this volume advocates. More specifically, in order to do justice to this approach a distinction between the self-as-subject and the self-asobject is in order; a distinction between activities themselves and products of self-reflection. Dynamic systems theory aims to study processes. This implies that the focus of dynamic systems theory is on the self-assubject. Ignoring the distinction between the self-as-subject and the self-as-object, between processes and products, would seriously handicap a dynamic systems approach because it would confound its objectives. And so too would sticking to the old ways of talking in terms of products. In other words, if we intend to study dynamics, we should also think in terms of processes and mechanisms instead of products. Moreover, a proper study of dynamic systems implies that the self-assubject cannot simply be taken as one ‘‘Self’’ or some sort of homunculus, but rather should be understood as being differentiated, and

Commentary on Van der Meulen

35

encompassing a number of simultaneous and interactive executive processes and functions. The multiple roles of affect Differentiation is also warranted with regard to the relationship between self-related mechanisms such as self-verification and selfenhancement, on one hand, and affect, on the other. If the relationship between these two fields is not clear, this may be because there are multiple interactions instead of just one causal relationship. Emotional intensity, for instance, functions as an instigator of selfreflection (e.g. Salovey, 1992), or can be used as an indicator for Me’s versus Not-Me’s. One might, then, infer from Frijda’s theory (e.g. this volume) that focusing on the self is the action tendency of intense global arousal. At the same time, this emotional intensity signals what individuals consciously or unconsciously consider to be part of their interests. Hermans’ Self-Confrontation Method (e.g. Hermans and Hermans-Jansen, this volume) uses the latter principle. Positive emotions instigate approaching or repeating, whilst negative emotions motivate avoidant self-regulating processes (cf. Van der Meulen, this volume). Because of these approaching or avoiding tendencies and their interpretative framework, specific emotions may – by means of positive feedback loops – function as attractors and organizers of selfrelated events (cf. Haviland-Jones, Boulifard, and Magai, this volume; Lewis and Ferrari, this volume). Moods (to be distinguished from emotions) are known to color one’s interpretative framework, but emotions also follow from such interpretations. Van der Meulen has already reviewed the vast majority of available options. The examples described above simply serve to underscore that the interactions and their functions can be many – depending on the self-related construct under study. The fact that diverse ones appear throughout the different chapters in this book suggests that we are still learning about each of them and that it may well be too early to integrate them. Contexts: systems of relationships As becomes clear from Van der Meulen’s review, contexts are construed in many ways. Personally, I have difficulty in defining the construct. One may focus on the dyad at the microlevel (cf. Fogel, this volume) and collect a very rich set of data. But the dyad as the unit of study and analysis is comparable to observing one contingency at a neo-Piagetian first stage (e.g. Fischer, 1980). At some point, we will

36


want to study triads, tetrads, and complex social systems, which will then extend beyond direct interaction. Or we may start from the other side and study the role of the social environment (past and present) through the perception of the individual, as more or less internalized ‘‘voices.’’ I speculate that we will soon find (in so far as we have not yet done so) that these voices depend on the . . . context. In sum, is it not the case that context is all that remains as Not-Self? How parsimonious is such a construct, and, if one takes the self-context interaction seriously, where do we set the boundaries between them? Much research concerned with the self in context would be more accurately described as the study of (the development of) subjective theories about the world and one’s stance in it (e.g. Kelly, 1955). These theories, then, have spin-offs for self-related mechanisms. These spinoffs, in turn, will interact with the context. The result is a theoretical perpetuum mobile of interactions between the various aspects of the self-as-subject and the self-as-object and various aspects of the context. In sum, with contexts as well, it is important to define the actual unit and focus of study. Everyone in his or her place All that I have said so far should not be interpreted as implying that any of the issues mentioned here are not worthy of study. On the contrary, I think that they represent vitally important attempts to understand human behavior. Perhaps so important (emotional intensity!) that we cannot be self-critical and precise enough in defining where we stand and which (part of which) mechanism or product we are trying to explain. Such specificity requires an awareness of the general field, where we stand, and, by implication, what we leave out or fail to address. A list such as the one offered by Van der Meulen helps and represents an important initial conceptualization of the relevant issues, but some sort of coherent framework would be even better. Here, I suggest the need for a three-dimensional space, created by shooting three theoretical vectors through the issues in Van der Meulen’s list and the research presented in the remainder of this volume. Perhaps recent research on the individual’s stance in the world could be positioned in a three-dimensional space of self-awareness, contextual complexity, and internalization. The self-awareness axis refers to interactions between cognition and affect in complementary gradations. On one side of the self-awareness axis we find experiential behavior (Epstein, e.g. this volume, commentary on chapter 3). Here, affect seems to predominate. The process may remain in this loop of sensation (global affect) and appraisal without the self becoming the


37

focus of attention. That is, the individual may remain in the organistic state (Rogers, 1951) or flow (Csikszentmihalyi, 1993) without objective self-awareness (Duval and Wicklund, 1972). The focus of study is solely on the self-as-subject. The individual may, however, also be drawn out of this state by the intensity of affect, the completion of a task, or physical (mirrors, cameras) or social triggers (an audience, feedback) into self-awareness (of self-discrepancies) and the inference of emotion. In other words, the flow of action is interrupted and attention shifts from an external orientation to a focus on the self. On this side of the axis we find self-narrating, commitment (identity), and rational, goaldirected behavior. Here, cognition predominates. In short, the axis of self-awareness represents recurrent loops of affect and attention with an increasing amount of cognitive self-reflection (cf. Oosterwegel, 1996). Again, in such a conceptualization, the question is not whether self-processes are cognitive (see Van der Meulen, this volume), but which particular part of a potential process one studies. The focus of this volume is to a large extent on this dimension. The process described above may take place whilst the individual is alone, in a dyad, triad, or complex social environment (the second axis). It may also occur in concrete social interaction where the other is largely extrinsically present or in a situation in which others are increasingly internalized into discrepant, conflicting or integrated internal ‘‘voices’’ (the third axis). The second axis reflects the number of actors in the actual physical social context. This is a truly social dimension, ranging from carer–infant dyads to group processes such as peer interactions. With the exception of work by Fogel and colleagues, this dimension is poorly represented in research on self-related concepts and mechanisms. In most research, the dimension mainly appears to provide a setting for individual functioning, and interactions with the contexts and within the context itself are taken to be relatively static. The third axis adds the personal history and experience each individual brings into the actual context. Albeit not lacking a social component, this dimension is more motivational than social. It refers to the degree to which experiences are internalized. On one side of the axis we find appraisals, beliefs, and standards (self-as-object), tendencies and reactions (self-as-subject) that appear to have their origins in the self. Here, we find drives, intrinsic motivation, and autonomous behavior. Further along this axis we find reflected appraisals and introjected motivation. Here, we find the internal voices of significant others and issues of self–other differentiation. Towards the end of the axis the input comes from outside the individual and motivation becomes extrinsic. Taken together, the three dimensions appear to provide a framework within which cognitive, affective, social, and motivational aspects of

38


the self-as-object can be located. These aspects of the functioning of the self-as-subject can be defined, and perhaps in due course everything can be understood in relation to each other. It would go well beyond the scope of this reply, and well beyond my potential, to map all theories of self-related constructs and mechanisms along these axes. I have, so to speak, merely offered an initial attempt to fill in the corners and edges of the theoretical jigsaw puzzle. My impression is that much research could be located at particular points or along one narrow line in this three-dimensional space. The reader may also find that many of the conceptualizations presented in this volume tend to move through this space more freely, each assuming a slightly different track. And what about the people we try to study? They seem to jump around like fleas. A dynamic systems perspective forces us to be specific about what makes them jump and where, and provides a tool for organizing some of this jumping into a complex integrated process.

COM M E N T A R Y

The self-concept is dead, long live . . . which construct or process? Differentiation and organization of self-related theories Annerieke Oosterwegel

In her chapter Van der Meulen organizes both theory and research on the ‘‘self-concept’’ around three key themes: cognition versus affect, interactions between self and context, and something termed ‘‘variability,’’ which can perhaps be characterized as a function of the former two plus motivation. This structure offers an excellent framework for the chapters that follow, where the specifics of these issues are taken up in some detail. At the same time, it appears that both the structure and the global nature of Van der Meulen’s account illustrate the major tension in contemporary ‘‘self-concept’’ research: important aspects are studied in isolation, and yet few specifics are known. Given these themes, and all the relevant issues which surround them, how do we combine them into an integrated whole, whilst also seeking to specify the precise nature of the mechanisms involved? This is the task which the current volume invites us to embark upon and for which its authors ‘‘set the stage.’’ Van der Meulen provides us with a useful listing of key issues which prevents us from losing sight of the broader picture as we immerse ourselves in detailed discussions of specific aspects. Here, I wish to go one step further and tentatively suggest an initial framework, based on the argument that we need (a) clear definitions that distinguish between the ‘‘self-as-object’’ and the ‘‘self-as-subject’’; (b) scope for several types of interactions between cognition and affect; and (c) clarity concerning what we understand as constituting ‘‘context’’ and where we position its boundaries. The not-so-omnipotent self-concept Van der Meulen makes a plea for clearer definitions of terms. The need for such clarity cannot be stressed enough. I would like to be even more rigorous, and, for the moment, dismiss any pretension of an 33

34


overarching label such as ‘‘The Self,’’ or Van der Meulen’s ‘‘the selfconcept.’’ It would be clearer to refer to each term with the prefix ‘‘self-’’ as a separate construct until we know how they are related. With respect to definition, the term ‘‘Self’’ refers to the person as a whole, taken from that person’s own perspective. In other words, the Self refers to the ‘‘I’’ of the individual, the acting entity of which affective, biological, neurological, and cognitive processes are part. Seen in these terms, however, studying the Self comes close to studying psychology; a statement that may be true, but does not afford much explanatory power. The self-concept, on the other hand, is nothing more than a sample of cognitive self-representations, a cross-section in a process of personal construction, one frame in a movie. Affect and contexts interact with self-evaluation and self-regulation, but only become a part of a self-concept once the individual has reflected on that affective process or interaction with the context, and has found it to be self-descriptive. Between the self and a self-concept lies the whole range of constructs we tend to study. Some of them, like self-regulation or self-evaluation, are conceived more actively and dynamically, others (fortunately fewer and fewer) are understood more statically. These constructs appear to be inextricably intertwined with context and affect, but again (as Frijda, this volume, argues) do not necessarily coincide. Too often and too early, in my opinion, we try to encompass different aspects and mechanisms into one omnipotent construct. One might say that self–other differentiation is the first step in healthy self-research development. It is not only important to be clear about what we study, but also to be aware of the aspects we do not cover in any given piece of research. Differentiation among self-related theoretical constructs becomes especially important if one takes a dynamic systems approach to research – as this volume advocates. More specifically, in order to do justice to this approach a distinction between the self-as-subject and the self-asobject is in order; a distinction between activities themselves and products of self-reflection. Dynamic systems theory aims to study processes. This implies that the focus of dynamic systems theory is on the self-assubject. Ignoring the distinction between the self-as-subject and the self-as-object, between processes and products, would seriously handicap a dynamic systems approach because it would confound its objectives. And so too would sticking to the old ways of talking in terms of products. In other words, if we intend to study dynamics, we should also think in terms of processes and mechanisms instead of products. Moreover, a proper study of dynamic systems implies that the self-assubject cannot simply be taken as one ‘‘Self’’ or some sort of homunculus, but rather should be understood as being differentiated, and


35

encompassing a number of simultaneous and interactive executive processes and functions. The multiple roles of affect Differentiation is also warranted with regard to the relationship between self-related mechanisms such as self-verification and selfenhancement, on one hand, and affect, on the other. If the relationship between these two fields is not clear, this may be because there are multiple interactions instead of just one causal relationship. Emotional intensity, for instance, functions as an instigator of selfreflection (e.g. Salovey, 1992), or can be used as an indicator for Me’s versus Not-Me’s. One might, then, infer from Frijda’s theory (e.g. this volume) that focusing on the self is the action tendency of intense global arousal. At the same time, this emotional intensity signals what individuals consciously or unconsciously consider to be part of their interests. Hermans’ Self-Confrontation Method (e.g. Hermans and Hermans-Jansen, this volume) uses the latter principle. Positive emotions instigate approaching or repeating, whilst negative emotions motivate avoidant self-regulating processes (cf. Van der Meulen, this volume). Because of these approaching or avoiding tendencies and their interpretative framework, specific emotions may – by means of positive feedback loops – function as attractors and organizers of selfrelated events (cf. Haviland-Jones, Boulifard, and Magai, this volume; Lewis and Ferrari, this volume). Moods (to be distinguished from emotions) are known to color one’s interpretative framework, but emotions also follow from such interpretations. Van der Meulen has already reviewed the vast majority of available options. The examples described above simply serve to underscore that the interactions and their functions can be many – depending on the self-related construct under study. The fact that diverse ones appear throughout the different chapters in this book suggests that we are still learning about each of them and that it may well be too early to integrate them. Contexts: systems of relationships As becomes clear from Van der Meulen’s review, contexts are construed in many ways. Personally, I have difficulty in defining the construct. One may focus on the dyad at the microlevel (cf. Fogel, this volume) and collect a very rich set of data. But the dyad as the unit of study and analysis is comparable to observing one contingency at a neo-Piagetian first stage (e.g. Fischer, 1980). At some point, we will

36


want to study triads, tetrads, and complex social systems, which will then extend beyond direct interaction. Or we may start from the other side and study the role of the social environment (past and present) through the perception of the individual, as more or less internalized ‘‘voices.’’ I speculate that we will soon find (in so far as we have not yet done so) that these voices depend on the . . . context. In sum, is it not the case that context is all that remains as Not-Self? How parsimonious is such a construct, and, if one takes the self-context interaction seriously, where do we set the boundaries between them? Much research concerned with the self in context would be more accurately described as the study of (the development of) subjective theories about the world and one’s stance in it (e.g. Kelly, 1955). These theories, then, have spin-offs for self-related mechanisms. These spinoffs, in turn, will interact with the context. The result is a theoretical perpetuum mobile of interactions between the various aspects of the self-as-subject and the self-as-object and various aspects of the context. In sum, with contexts as well, it is important to define the actual unit and focus of study. Everyone in his or her place All that I have said so far should not be interpreted as implying that any of the issues mentioned here are not worthy of study. On the contrary, I think that they represent vitally important attempts to understand human behavior. Perhaps so important (emotional intensity!) that we cannot be self-critical and precise enough in defining where we stand and which (part of which) mechanism or product we are trying to explain. Such specificity requires an awareness of the general field, where we stand, and, by implication, what we leave out or fail to address. A list such as the one offered by Van der Meulen helps and represents an important initial conceptualization of the relevant issues, but some sort of coherent framework would be even better. Here, I suggest the need for a three-dimensional space, created by shooting three theoretical vectors through the issues in Van der Meulen’s list and the research presented in the remainder of this volume. Perhaps recent research on the individual’s stance in the world could be positioned in a three-dimensional space of self-awareness, contextual complexity, and internalization. The self-awareness axis refers to interactions between cognition and affect in complementary gradations. On one side of the self-awareness axis we find experiential behavior (Epstein, e.g. this volume, commentary on chapter 3). Here, affect seems to predominate. The process may remain in this loop of sensation (global affect) and appraisal without the self becoming the


37

focus of attention. That is, the individual may remain in the organistic state (Rogers, 1951) or flow (Csikszentmihalyi, 1993) without objective self-awareness (Duval and Wicklund, 1972). The focus of study is solely on the self-as-subject. The individual may, however, also be drawn out of this state by the intensity of affect, the completion of a task, or physical (mirrors, cameras) or social triggers (an audience, feedback) into self-awareness (of self-discrepancies) and the inference of emotion. In other words, the flow of action is interrupted and attention shifts from an external orientation to a focus on the self. On this side of the axis we find self-narrating, commitment (identity), and rational, goaldirected behavior. Here, cognition predominates. In short, the axis of self-awareness represents recurrent loops of affect and attention with an increasing amount of cognitive self-reflection (cf. Oosterwegel, 1996). Again, in such a conceptualization, the question is not whether self-processes are cognitive (see Van der Meulen, this volume), but which particular part of a potential process one studies. The focus of this volume is to a large extent on this dimension. The process described above may take place whilst the individual is alone, in a dyad, triad, or complex social environment (the second axis). It may also occur in concrete social interaction where the other is largely extrinsically present or in a situation in which others are increasingly internalized into discrepant, conflicting or integrated internal ‘‘voices’’ (the third axis). The second axis reflects the number of actors in the actual physical social context. This is a truly social dimension, ranging from carer–infant dyads to group processes such as peer interactions. With the exception of work by Fogel and colleagues, this dimension is poorly represented in research on self-related concepts and mechanisms. In most research, the dimension mainly appears to provide a setting for individual functioning, and interactions with the contexts and within the context itself are taken to be relatively static. The third axis adds the personal history and experience each individual brings into the actual context. Albeit not lacking a social component, this dimension is more motivational than social. It refers to the degree to which experiences are internalized. On one side of the axis we find appraisals, beliefs, and standards (self-as-object), tendencies and reactions (self-as-subject) that appear to have their origins in the self. Here, we find drives, intrinsic motivation, and autonomous behavior. Further along this axis we find reflected appraisals and introjected motivation. Here, we find the internal voices of significant others and issues of self–other differentiation. Towards the end of the axis the input comes from outside the individual and motivation becomes extrinsic. Taken together, the three dimensions appear to provide a framework within which cognitive, affective, social, and motivational aspects of

38


the self-as-object can be located. These aspects of the functioning of the self-as-subject can be defined, and perhaps in due course everything can be understood in relation to each other. It would go well beyond the scope of this reply, and well beyond my potential, to map all theories of self-related constructs and mechanisms along these axes. I have, so to speak, merely offered an initial attempt to fill in the corners and edges of the theoretical jigsaw puzzle. My impression is that much research could be located at particular points or along one narrow line in this three-dimensional space. The reader may also find that many of the conceptualizations presented in this volume tend to move through this space more freely, each assuming a slightly different track. And what about the people we try to study? They seem to jump around like fleas. A dynamic systems perspective forces us to be specific about what makes them jump and where, and provides a tool for organizing some of this jumping into a complex integrated process.

CH A P T E R 3

The self and emotions Nico H. Frijda

What is the role of the self in emotions? Is it essential for emotions? Emotions are subjective; therefore, what place has subjectivity in emotions, and what place has the subject’s representation of this subjectivity? It appears to be legimate to question the extent to which a notion of ‘‘self’’ is indispensable to understanding emotions. I find it difficult to discuss this issue in a coherent manner. One reason is my lack of familiarity with discussions of ‘‘the self.’’ Another is that the word ‘‘self’’ seems to possess many different meanings that may or may not have much to do with one another. Most certainly, each individual is in some sense a ‘‘self,’’ by definition. But does he or she necessarily ‘‘have’’ a self? For myself, I am not sure. The best way to approach these issues, therefore, seems for me to briefly review my analysis of emotions, and to examine whether and where a notion like ‘‘self’’ appears needed, and in what function. Such an approach may be useful because by concentrating upon the emotional phenomena one may perhaps steer clear of certain conceptual confusions. The literature on emotional development suffers from them here and there. For instance, it sometimes worries about the age at which a given emotion, say shame (Lewis, 1992), appears. This given emotion is thereby identified by a name that itself is undefined. At what age shame appears is, however, a meaningless question when it is not independently made clear what ‘‘shame’’ refers to, and which of the defining features appear at what age; they usually will not all make their appearance at the same time. Involvement of some notion of the self may be one, but some other role for that notion may be another, independent one. When, for instance, a certain development of selfawareness is considered a necessary defining condition for shame it cannot be an empirical finding that appearance of shame depends upon a certain development of self-awareness. The perspective of my analysis will, therefore, be to examine to what extent a reference to self is needed for describing or understanding the phenomena, and what that reference then is needed for. 39

40

Nico H. Frijda

Emotions and subjectivity Emotions, admittedly, are subjective. They depend upon the subject’s viewpoints, appraisals, and personal attributions of meaning. They are subjective also in that each subject may have different emotions in response to the same event. The subject (the ‘‘transcendental subject’’) is a logical condition for explaining emotions. Is then some awareness of the self not a precondition for emotions? I think this conclusion is a misconception. It confuses the standpoints of the experiencer of the emotion with those of an outside observer or explainer. It confuses the phenomenological and the causal-analytical perspectives. When trying to understand how emotional phenomena come about, an observer or a theoretician observes that two entities are involved, the subject and an object. He knows that the emotional phenomena depend upon both, and that these often allow more inferences regarding the state of the subject than the nature of the object. A dog is not by its very nature fearsome or endearing; the subject sees it as such. Phenomenologically, however, this subjectivity need not be present. To the subject, the object’s emotional nature may be the truth. The dog is fearsome, period, just as the Kosovars simply are evil to the Serbs, and the Serbs to the Kosovars. Emotions are perceptions in the first place. Phenomenologically, emotions are out there, just as the red and green of an apple are out there. One perceives an object or a scene that reveals its meaning. And it is not subjective that the object may make one shrink. Neither is it subjective that a tall tree makes one bend one’s neck backwards. In this perceptual experience the subject is not present. There is an object and no self. The subject him- or herself may not be part of the experience. He or she may be – and I will come back to this – but this is not necessary in order for an emotion to occur. A representation by the subject of him- or herself may or may not be a condition for that experience; we cannot know in advance. But it is not so by necessity. Appearance of an emotional reaction, be it an experience or some overt behavior or behavioral impulse, does not allow a compelling inference regarding the representations that play a role. It is necessary to separate experiences of ‘‘the self,’’ or those of the subject of him- or herself, from the subject as a logical condition for the explanation of emotions. And thus: what the role of the self is in the constitution of emotions is an empirical matter.

The self and emotions

41

Analysis of emotions Determining empirically where ‘‘the self’’ plays a role in emotions, and how, necessitates an analysis of what emotions are, that is, of the phenomena that the term ‘‘emotion’’ refers to, and how these are best understood. ‘‘Emotion’’ does not necessarily refer to subjective experience. Emotions, in fact, are multicomponential (Frijda, 1986; Scherer, 1984); by this we mean that the human and animal reactions that are designated ‘‘emotions’’ or ‘‘emotional’’ consist of a number of components. The major components are appraisal (that is, the way an event is appraised), state of action readiness, physiological response, and one’s feelings stemming from these components. Several of these components tend to occur together, and there rarely is only one of them. Emotions can thus be described in terms other than only those of experience, and emotional experience is usually part of a larger whole. More importantly, the reactions that are called ‘‘emotions’’ can be described in functional terms that apply to each of the components as well as to the reaction as a whole. Such functional terms characterize emotions as involuntary changes in an individual’s relationship with an object – most often an object in the environment, often the subject him- or herself, sometimes an object in thought or a conceptual object such as an ideal. From a functional point of view, emotions (the reactions or states that we call emotions) are interactive states rather than inner, felt states. They operate between an individual and his or her environment. They are ways for the person to position him- or herself towards the object and ways of being inclined to deal with it correspondingly. This is, I think, the most profitable perspective to take on the phenomena concerned, such as experience, motivation, behavior, and bodily changes. Emotions are interactive events. They have to do with the relationship between two entities, a subject and an object. That of course is the case with any psychological event, even a simple perception, in that it somehow affirms the independent existence of some entity out there. What makes an emotion different from a perception is that a particular, specific relationship between subject and object is affirmed or constructed. Emotional behavior and emotional feeling fit this description. The behavior regarded by bystanders or by the subject as emotional by and large represents activities that change the relationship with an object or the environment as a whole. Emotional behavior can be interpreted as behavior that manifests changes in action readiness (Frijda, 1986). It is relational behavior, implemented to establish, maintain, or disrupt such a relationship, or it may be behavior that shows that no

42

Nico H. Frijda

relationship is being sought after. Manifesting a particular form of action readiness is the common denominator of the various behaviors shown under particular emotional conditions; states of action readiness are in fact inferences from what those behaviors have in common. Behaviors called ‘‘angry’’ are largely behaviors meant to block the progress of an antagonist or to hurt or damage an antagonist; behaviors called ‘‘joyful’’ are largely exuberant behaviors, that is, interactive behaviors in excess of what might have been needed under the circumstances, and so forth. A systematic description can be given of the modes of action readiness that can meaningfully be distinguished in humans and higher animals. It is indeed likely that they form a limited collection, each element of which represents the activation of some distinct neural disposition. Modes of action readiness can be divided into modes of activation (activation increase and decrease, and inhibition, including tenseness) and action tendencies. The action tendencies each seek to implement a particular type of relationship or relationship change. Major modes that are usually distinguished are desire or seeking to obtain or handle, attending, proximity-seeking, exuberance or playing, avoidance, antagonism, rejection, dominance, and submission. Each of these modes of action readiness can be identified in self-reports, and tends to appear from factor analyses based upon them (Davitz, 1969; Frijda, Kuipers, and Terschure, 1989) as well as in behavior (e.g. Van Hooff, 1972, for chimpanzees). They thus represent data of subjective experience, of motivational states, and of behavior. Each of the distinct modes of action readiness tends to correspond to one or several of the major common emotion categories, as illustrated in table 3.1. All modes of action readiness have to do with dealing with the environment in a particular fashion, or with not dealing with it. ‘‘Deactivation’’ means not simply doing nothing, but also denotes lack of striving and lack of interest, as in sad apathy. The behaviors suggesting these states of action readiness tend to be involuntary. The suggested states share the characteristic of ‘‘control precedence,’’ that is, they tend to interfere with voluntary behaviors or other ongoing goals, and regulation processes appear needed to prevent this. Behaviors resulting from states of action readiness may take a form determined by cultural habit and individual preference. However, the behavioral repertoire contains core features that appear to belong to the endowment of the species. These are called expressive behaviors. The behaviors are ‘‘expressive’’ in the sense that onlookers can and do make inferences concerning the states of readiness involved (and thus concerning the emotions). However, they are not expressive in the sense that they are meant to communicate emotions to others. In part, they are


43

Table 3.1. Modes of action readiness and corresponding illustrative emotions Modes of action readiness

Emotion names

desire, seeking attending proximity seeking exuberance or playing avoidance antagonism rejection dominance submission interrupting depending

desire interest affection joy fear anger disgust pride shame surprise sadness

activation deactivation inhibition

elements of relational behavior repertoires. The standard surprise expression, for instance, is part of an orienting response, and the standard fear expression (Ekman, 1982) a self-protective one. In part they are indeed communicative signals to the social environment, not so much to make one’s emotions known, but to influence the behavior of onlookers so as to favor the prevailing action readiness; for example, crying incites help-giving and exists for that purpose (Fridlund, 1994; Frijda and Tcherkassof, 1997). The point of the preceding analysis was to show that emotions include more than experiences or cognitive representations of one’s emotion. The self, a representation of the person by the person, does not need to be part of them. Events bring an individual into a state of self-protection, or of aimless agitation, or of defensive action impulse, or perhaps they only diffusely increase the state of activation, with no particular object at all. Physiological changes usually accompany this, as the ‘‘logistic support’’ of action, in case action is produced. Maybe this description is adequate only for primitive or elementary emotions, such as reflex-like instances of fear and anger; but these are among the emotions nonetheless. Even emotional experience may on occasion be entirely nonrepresentational. There exist object-less moods (Morris, 1989). Furthermore, a major aspect of emotional experience is experience of affect, that is, of feelings of pleasure and pain. Pleasure and pain are

44

Nico H. Frijda

non-cognitive states that can be felt, and that can also be experienced without, or regardless of, feeling (Zajonc, 1980). They are best understood as states of acceptance or non-acceptance of stimuli or of the individual’s own condition (Frijda, 2000). The individual may just suffer them, or, in the case of negative affect, be driven to action to escape from them. The latter, for instance, holds good for battery-farm animals – chickens, pigs, cows – that increase their endorphin-release via self-mutilation. Cognitive representations will help to focus or instigate actions, but the urges to action may be there in their absence. All this is of course a description of a crying newborn baby in which the action urge avails itself of the innate crying repertoire. Furthermore, there exist experiences of anger, fear, panic, depression, or joy reflecting the awareness of states of action readiness that may have no representational content. Emotional experience Changes in action readiness do not usually come about spontaneously, by physiological process (they may do so, in exceptional and notably pathological cases; Izard, 1993). They are instigated by stimuli or events, as appraised by the subject. Emotions, in the typical case, are the products of appraisal. Let us define emotions by including this restriction: emotions are response structures centering upon changes in action readiness, called forth by events as appraised. ‘‘Events’’ include recollections and thoughts, as well as perceived bodily changes. ‘‘Appraisal’’ means that some event or thought obtains affective value, and usually further aspects of meaning, by virtue of certain processes in the individual. These processes may be simple. This is the case when certain sensory stimuli and bodily states impinge upon elementary affective sensitivities like pain sensitivity and those producing bodily discomfort, or pleasure given by sweet taste (Blass and Shah, 1995). Almost equally elementary is elementary matching with stored information, as with the effects of mere exposure (Zajonc, 1980). More often it is more cognitive, calling upon learned expectancies and cognitive schemas (Lazarus, 1991a), and sometimes conscious inferences. Appraisal includes two rather distinct sets of processes. One of these includes the processes by which a stimulus event obtains affective value; it is sometimes referred to as ‘‘primary appraisal.’’ The second consists in picking up or generating information that gives clues for the kind of action readiness that might suit the context at hand; it is called ‘‘secondary appraisal’’ (Lazarus, 1991a). Secondary appraisal may be absent when no response-relevant cues are there. A crying newborn baby is simply distressed, and its helplessness response is the only


45

available one. A little later it may become aware of the presence and relevance of the mother, and modify its emotional urge accordingly. Clearly, the nature of emotions and of the emotion repertoire changes when the repertoire of cognitive cues changes or is extended. Different emotions differ in appraisals, and they may also differ as to state of action readiness. Appraisals can be described at two levels. They can be described globally as ‘‘themes’’ (Lazarus, 1991a) such as ‘‘loss,’’ ‘‘success,’’ ‘‘threat,’’ or ‘‘frustration’’; or more analytically in terms of the variables of which the themes are composed, such as present or expected, favorable or harmful, increase or decrease, controllable or uncontrollable, or being due to some other person’s agency or one’s own agency (as these are often prominent in anger and guilt feeling, respectively). Different proposals for appropriate sets of variables have been made (see Scherer, 1999, for a review of current appraisal theory). The point of all this in the present context is again that the subject is not a necessary element of the cognitive domain that constitutes appraisal. This is consistent with what was said earlier about the nature of emotional feelings. Emotional experience or feeling consists in the first place in experiencing the emotional event as appraised. It is, as I said, a form of perceptual experience; it is out there. One sees an object or scene that is disgusting, fearful, attractive, or desirable. The epithets ‘‘disgusting,’’ ‘‘fearful’’ or ‘‘desirable’’ translate a perceived quality of what is out there. They indicate what the object or event is in the process of doing, withholding or offering, or the potential contained in them for further effects that might ensue with further contact or longer confrontation. ‘‘Disgusting’’ is what is felt to be unbearable upon close contact, to possibly soil indelibly, to contaminate (Rozin and Fallon, 1987; Miller, 1997); ‘‘fearful’’ is something or someone that spells danger or spells something unknown; and so on. Emotional experience is not in the first place ‘‘in here.’’ It is not felt as my state; it does not necessarily have the component ‘‘I feel.’’ This, of course, is not always so. We do come to realize that our feelings are in our minds, just as we come to realize that our perceptions are in our heads, and that the origin of their meaning is in here. Of course, people can, when reflecting, realize that emotional experience is a subjective experience. They know they can have such an experience without anybody else having an inkling of it. They can be vividly aware of the fact that they have a certain feeling. But it does not need to be the case, and experiencing an event-as-appraised can exist without it perfectly well. Emotional experience does not presuppose a sense of self. It presupposes a sense of the world.

46

Nico H. Frijda

This applies to all facets of emotional experience. Emotional experience is a combination of experience of the world with a particular situational meaning structure (appraisal awareness as discussed), experience of affect, experience of a state of action readiness, and experience of arousal. Experience of affect is most often a form of experience of the world. Affect is primarily experienced not as a subjective state, but rather as an encounter with a repulsive or attractive object. The attractiveness, aversiveness, acceptability, or unacceptability are all out there, if not in the lightness and ease or heaviness and difficulty of one’s actions. Action readiness, to the extent that it is experienced and not just ‘‘had,’’ is experienced as a reachable, to-be-escaped-from, to-beexamined, to-be-approached event in the world or goal to be achieved. In joy the world is felt as an accessible world, in desire the world is felt as a world that has to be accessed. And the experience of one’s body? Is it not one’s own? Well, in some sense it is (in some sense ‘‘own’’ has experiential content), and I will come back to this. But primarily, in emotional experience, engagement of the body is experienced as the glow of reality of the emotional event, the fact that its meaning is to be taken seriously, imminent, there (this formulation is taken from Sartre, 1939). All this implies that the subject does not necessarily figure in emotional experience. The subject is not necessarily part of the emotional object. It – the subject – is of course a condition of potential emotional experience, as it is for any other experience. We have to suppose a ‘‘transcendental subject,’’ that reflects at Dennett’s (1978) ‘‘intentional level of analysis’’ the sum-total of the information-processing processes at Dennett’s ‘‘functional’’ level. We do not, however, have to suppose, nor do we find, a ‘‘self’’ in the sense of a representation of an entity with properties on a level with the perceived objects. This analysis, as already indicated, does not apply to all emotions. Not all emotional experiences conform to it, nor do some emotion categories as defined or as used in labeling emotion instances. ‘‘Conceit’’ implies explicit reference to the self, and so does ‘‘remorse’’ or ‘‘guilt feeling.’’ But there is a large range of emotions for which it does not apply. These emotions, by consequence, are within reach of both animals and small children at an age at which no representation of the self can be assumed to be established. The emotional experience does not imply more than the presence of a valenced perception, plus the calls for action that may or may not be part of the state of awareness. The calls for action presumably require no awareness to push for simple actions like crying or defensive movements; nor does the activation of what are best considered support mechanisms for coping actions, the changes in autonomic functioning. Phenomenally, one may


47

be drowned in one’s perceived environment. This is the case for an animal that, presumably, has no awareness of self, or at least most of the time does not have one. It is the case for a baby. It is the case for adults in rapture or rage, where they ‘‘forget themselves.’’ There is no self of which awareness has to have grown before emotions are possible. There need not be a sense of the subjectivity of emotions for emotions and emotional experiences to be possible, because no subjectivity whatsoever is required in emotional experience. There need not be any awareness of ‘‘I feel sad,’’ and most emphatically not of ‘‘the sadness is mine.’’ No. The sadness is there, in the event, in the world. The self as subject Analytically, of course, there is a transcendental self, the logical reference point for any experience and striving, for any experience of ‘‘being-in-the-world.’’ The transcendental self alone ‘‘has’’ no self. It has no properties. It is just there as the condition for the possibility of experience. It is not usually an object of awareness, but simply a logical condition for such awareness. Yet, I think there is something special in awareness and in the conditions of behavior that would seem an empirical content of the transcendental self. I think there is something in emotional experience that, from a particular moment onward, allows the use of the first-person pronoun in verbal expressions, but that was there in experience before such use and irrespective of such use. Probably, the awareness has little or no cognitive content. It is a form of knowledge-by-acquaintance rather than knowledge-by-description. The knowledge-by-acquaintance, it would seem, is built upon experiences of motor action, motor preparation, and motor planning, that is, experiences of familiarity and of action initiation or spontaneity. Familiarity. One of the sources for this supposition is the experience of familiarity of objects, events, and one’s body and actions. Objects, surroundings, and oneself are experienced as familiar or, at least, as real; I will use ‘‘familiarity’’ and ‘‘reality’’ interchangeably because both are meant here in an emotional sense, referring to impressions rather than to justifiable knowledge. Usually, even unfamiliar objects do not entirely lack this characteristic because they do not necessarily appear ‘‘strange.’’ The ‘‘awareness’’ of familiarity, strange as it may seem, is not something one is usually aware of. It is a bit like the awareness of water for a fish. It becomes articulate only after it disappears, in feelings of strangeness and unreality. Such feelings are common during fever, under the influence of drugs or brain damage

48

Nico H. Frijda

(sensory neglect), in response to sudden threatening events such as when one’s car skids or upon being notified that a close person has died or desires to break up one’s relationship. They are also common under torture and other traumata, and after traumata, where they have been called states of numbness or ‘‘denial states’’ (Horowitz, 1992). One may roam the ruins of one’s bombed city in a daze, as if none of it is true and as if none of it has happened to oneself. These experiences are thus almost ubiquitous during and after threatening events. By inference, under normal conditions the opposite feelings of familiarity, reality, or of being in contact with the events are always there, as a background. How the sense of familiarity develops I do not know, nor do I know at what age it develops, or what processes are involved. Mere previous experience is clearly neither sufficient nor necessary. The ubiquitousness of feelings of estrangement and their short latency after the onset of the eliciting event (to judge from descriptions of experiences during skidding and under torture) suggest that endorphin mechanisms are involved. The feelings of estrangement and numbness themselves suggest that they stem from a disturbance in the automatic involuntary motor responses to sensory stimuli which represent their ‘‘knowledge by acquaintance,’’ such as implicit orienting, readying for grasping, reaching, or using; or perhaps the disturbance is in the awareness of those responses. Responses of this nature emerge from the beginning of life onward, when the coordination of eye movements and perceived stability of the environment develops. It is well known that the visual world is perceived as stationary under eye movement as long as movements across the retina are compensated for by self-initiated eye muscle innervations, that is, by movement intentions. Involvement of incipient or actual motor actions has of course been emphasized and was studied at the time in the so-called ‘‘sensory-tonic field theory’’ by Werner and Wapner (1953), which obtained a solid extension in the work on adjustment to distorted visual fields (Held, 1965). Spontaneity. There is not only coordination between planned and executed movements (as evident from the stationary visual field with eye movement), but also between the planning of actions and their perceived outcomes. As Piaget (1935) argued at the time, the beginnings of intentional action are seen in the secondary circular reaction, that is the child noticing, for instance, the coincidence of its movements with the sound of a rattle or the rocking of its crib, and seeking to reproduce the latter. Anyway, the cycle of intent-operate-test-exit is one of the most fundamental building blocks of human functioning. So, it would seem, is some awareness of it by the individual. Monitoring the completion of those cycles is fundamental because the purposive nature of planned movement depends upon it, whether that of crawling


49

towards a goal or anything more complicated. Planned movement can only be executed when a movement outcome can be observed to coincide with what the movement set out for. Self-initiated movement, spontaneity, intentional action, and initiative: all four are expressions to denote this process structure. It should not surprise us that spontaneity appears to have a special status in experience, to judge from the emotions connected with its vicissitudes. To begin with, movement interference is one of the earliest frustrations to be observed. Watson regarded it as the unconditional stimulus for anger. It leads to protest by animals and babies alike. Movement restraint (rolling a rat in a towel) was the first experimental stressor examined by Selye (1956). Interruption of voluntary action quite generally generates distress or aggression. This would seem to be a direct consequence of the fact that monitoring task completion is at the heart of voluntary behavior. Indeed, a strong case can be made for the hypothesis that monitoring the effective functioning of an individual’s processes is the major function of pleasure and displeasure experience (Frijda, 2000); and executing intentions is one of the very basic functions of those processes. Non-intentional processes, by contrast but probably by virtue of these same facts, have a special status in experience. This presumably is because they impose themselves upon intended ones. The special status I mean is that of emotional experience. Ancient philosophy labeled what we now call ‘‘emotions’’ as ‘‘passions’’ – that which the soul receives passively – or ‘‘affections’’ – affects, that which affects the soul – as distinct from the soul’s actions. Emotional experience, as I noted above, has control precedence. It is felt to overrule ongoing actions, or take over control of action. The self does not figure in emotional experience, but infringement upon the self by emotional events certainly does. Spontaneity and self-initiation are emotionally valued, from an early age onward. They are desired and appraised positively. Even before the use of verbal self-reference, whether in the form of ‘‘I’’ or by the child’s use of its own name, a child makes efforts at independent actions. It may violently resent when a helpful parent takes over. I once observed a one-year-old who wanted his toy to be put back where it was when he began to reach for it, but his father picked it up for him, so that his own reaching could be completed. Evidently, it was the game that counted rather than the toy. My understanding of all this is that, at this stage, there is no self that obtains satisfaction. Rather, there are planned processes that are recognized and monitored as such, actions in the philosophical sense, the realization of which gives satisfaction. It may well be the recognition of these very processes that verbal and conceptual development takes

50

Nico H. Frijda

hold of in acquiring the proper name and the personal pronoun, which then is used as the object of self-reference. ‘‘I’’ is a way to talk with a pronoun about a way of existing or functioning that contrasts with all other observable ways of functioning. The contrast is based upon one’s knowing what is going to happen the next instant and upon having emotions when the next instant yields what one is prepared for. James (1890) indeed designated ‘‘I as actor’’ as one of the modes of selfawareness. ‘‘I as actor’’ thus exists, not only transcendentally but also functionally, as a process well before it becomes a conceptual entity. It eventually becomes a conceptual entity, of course, but when it does, this is still in a more or less formal role. ‘‘I’’ as actor has no other properties than that of being a causal agent. It refers to the causal agent who topples the tower, and has a continuous existence beyond that event. I can cause harm and take the blame for it, in the sense of being aware of having caused it. The constellation may give rise to protoguilt. That is, it may motivate actions directed at atonement and the regaining of love, to my mind long before there is any notion of responsibility, of obligation, or of morality. Social interactions. There are more emotional constellations in which ‘‘I’’ is a formal background or condition, rather than an object in the field of awareness. That is, when the person enters the field as an actual or possible target of actions by others. To be a target of others as such may cause emotions, without any realization that they are caused by being a target. Maltreatment hurts or leaves the individual in an emotional cold, in a very elementary fashion. Things change when comparisons with others or with oneself at other times start to play a role. Counterfactual comparisons have, I think, to be assumed in understanding why the good fortune of others may cause distress, as it does in envy and jealousy. The fact that I get upset about what another gets must be because I do not get it. Cognitively, it must be primitive because even small infants may become excited when other infants receive food or attention; yet a frustration is felt that is not directly taking place, because nothing is actually being withheld. But it is not that primitive, after all, since at least an understanding is involved that eating or receiving in others can be recognized as similar to actions of one’s own; there is an elementary empathy involved, perhaps related to the possibilities of imitation identified by Meltzoff and Moore (1979). No counterfactual comparisons are necessary to recognize that one can be a target of others’ actions. The glow of anticipating approval for what one did also centers around oneself as an actor. It may even be a first step towards becoming an object of one’s own emotions. Expecting praise contingent upon what I do or will do would seem to hover


51

midway between expecting a pleasurable event with oneself as actor and having a representation of oneself as both the actor and the recipient of the praise. Perhaps negative emotions show this more clearly. There of course exists distress felt and shown by being rejected from participating in a group. In a variant of such distress the rejection is attributed to what one did or who one is (‘‘We don’t want you! Red hair!’’). It may give rise to non-moral shame (Ausubel, 1955), that is, to distress linked to identifying oneself as the cause of the rejection, and leading to withdrawal and desire to hide, or change one’s hair-color. No self-schema (Markus, 1977) would seem to be involved, let alone a falling short of some self-ideal; but something akin to oneself-as-actor would appear to play a role. One is the causal agent of rejection, with the shame reaction (distress, submission to the judgment of others as well as submission in a posture with bent head) the appropriate one. Shame is close to being humiliated, which can also emerge, I think, with an elementary cognitive structure. Humiliation is here defined as being pushed to an inferior role in a relationship. It is the structure of frustrating expectations and the construction of a role around them that carries the emotion beyond mere distress. Yet it does not necessarily include a conceptualization of one’s role. One may live a life of humiliation without conceptualizing oneself accordingly. Conceptually pride, shame, and humiliation all have to do with the self. They all conceptually imply counterfactuals, and they may also do so psychologically (Niedenthal, Tangney, and Gavanski, 1994). But psychologically they may exist without such a cognitive structure and I think, therefore, that they do not need the corresponding cognitive development. The concepts should not lead the analysis of the emotions themselves astray. The self as object There are of course many emotions in which the person him- or herself figures as an object. There are two forms of this. First, there is consciousness of oneself as an experiencing subject. Traditionally, emotional feelings have been regarded as subjective experiences in their phenomenology as well as their genesis. They are my experiences, my inner experiences. Introspective reflectiveness makes them so, and attention to body feelings adds to this. There is in fact a doubling of awareness: there is awareness of some meaningful object or event, and then this awareness is the object of reflective awareness. ‘‘There is a frightening event’’ changes into ‘‘I feel fear.’’ And there is awareness of the relationship between the two, in that there is an ‘‘I’’ who has, feels, or perhaps thinks he feels, the emotion. Adult

52

Nico H. Frijda

emotionality is hardly ever free of reflective self-observation. We almost always know what we feel and that, in fact, we feel that we feel. Only when self-forgetful, in ecstasy, under extreme suffering, or when blinded by anger or desire, is this different. Second, however, one can oneself be an object of appraisal and thus of an emotion: one of one’s actions, thoughts or emotions, or oneself as a person. Anger towards oneself, self-hatred, pride, shame, and guilt are obvious examples. At least, they are in their more prototypical forms, which are close to the definitions of the concepts and to the legal meaning of the word ‘‘guilt.’’ There are intricacies here. I just mentioned shame and guilt, these words being used to denote states giving rise to certain response patterns. Both words are indeed used in this way in daily conversation. In those usages they likewise fit the eliciting circumstances as appraised by the subject reasonably well. ‘‘Shame’’ mostly follows devaluation by others, which the subject accepts, and ‘‘guilt’’ follows having caused harm to others, even if inadvertently. Neither devaluation nor having caused harm need imply any moral overtone. One can be ashamed of having red hair, and feel guilty about having hurt someone who behaved unpredictably and imprudently (Frijda, 1993b). By consequence, emotions given those names, and involving the action tendencies of desiring to hide from view and to atone, respectively, can occur without the cognitive complexities of comparisons with self-ideals or moral standards. But that does not do away with the fact that the more prototypical shame and guilt emotions also exist, and that in them a representation of the self and a self-schema are elements. In most situations of shame, as defined by appraised evaluative rejection by others and the action tendency to withdraw or hide, there also is an image of one’s person as being of lesser value. In most situations of guilt feeling, as defined by feeling responsible for harm to others and an impulse to atone, there is an explicit notion of oneself as a blameful agent. Pride often is not just joy at achievement or position, but also a high evaluation of one’s person and a high view of one’s capacities for achievement. And in many emotions that have the appearance of simple joy, fear, or sadness, the value of oneself as a person or as an actor plays a role in the background. Often, indeed, it plays a role in the foreground. The emotions may include a representation of oneself as valuable or worthless, leading to actions that communicate that evaluation in social interaction (in timidity or conceit), and in expectations to be treated accordingly by others or to fill a particular social-status slot. Still, one has to be careful in assessing precisely what that role is. Is the self-reference or self-evaluation an elicitor of the emotion, or merely


53

something that gave form to an emotion that would have been there anyway? Or is it merely something that served social integration in the wake of a more elementary emotional response? These possibilities became apparent in self-report studies of shame and guilt (reported in Frijda, 1993b). The self-evaluations in shame and guilt, at least sometimes, appeared to have come after an emotion had been aroused, rather than being one of its antecedents. Sometimes they appeared to give structure to the diffuse distresses of rejection and empathic distress, respectively. The negative self-evaluation in shame also functions to bridge the gap between the view of cherished others and one’s own view (coming to dislike one’s red hair at least removes one difference); taking the blame in guilt feeling at least goes some way in equalizing suffering (Baumeister, Stillwell, and Heatherton, 1994). To repeat: the conceptual analysis of the emotion words, and even the subject’s selfreports, provide no convincing evidence of what was going on, psychologically. Emotion significance Not only one’s acts, properties, and person can be the object of emotional appraisal. So, too, can one’s emotions. Emotions are appraised; many and perhaps most emotions are. This applies to individual emotion instances as well as to emotion categories (‘‘anger,’’ ‘‘fear,’’ ‘‘fear of spiders’’). One may feel anger to be a blameworthy passion, one may dislike having fallen out with someone on a given day, one may be proud of having stood up for oneself when scolded, one may despise oneself for being afraid to speak in public. Emotions are appraised because of the implications connected to having, showing, or feeling them. I will refer to knowledge (conscious or non-conscious) of these implications as the emotion’s ‘‘significance’’ (Frijda, 1986). Emotion significance is what motivates emotion regulation; one regulates one’s emotions, or one’s emotions are regulated, because of their possible consequences. Emotions may show various forms of significance. Their most immediate consequences are practical. Other people may retaliate to one’s burst of anger; nervousness tends to decrease motor efficiency; examination fear blocks the availability of knowledge; some desires bring one into trouble. Emotions may also conflict with how one wants to be or to appear to others. They may, on the contrary, also improve one’s image in the eyes of others or of oneself. A careless and carefree attitude in an amorous encounter may be thought to work better and, anyway, one might wish to possess daring. There also exist moral rules and social norms to which one would want to conform and which, if one does not, might generate shame.

54

Nico H. Frijda

The various forms of significance may affect the actual occurrence of emotion or whether or not one feels it, or these forms may merely affect one or some of its manifestations. In the latter case significance tends to go by the name of display rules (Ekman, 1972). Not all manifestations of significance involve representations of the self, as these distinctions indicate. Some of them are outright and basically emotional; the fear evoked by sexual temptation when the consequences are unknown (the seducer may wish to take advantage) is an example. But some forms most distinctly do involve representations of the self, because they involve who one knows one is, who one wants to be, and how one wants to be and to be seen. Concerns From a dynamic, that is, motivational point of view the transcendental self or, rather, what and how the subject ‘‘is,’’ plays a much more important role in emotions than what individuals think they are or do. Many of the properties that characterize an individual as a particular individual are decisive in emotions. They underlie emotions whether one knows it or not; and whether one knows it or not also makes little difference. The most relevant properties are the individual’s concerns. ‘‘Concerns’’ is the concept used for an individual’s motives, major goals, interests, attachments, values, ideals, sensitivities, and aversions and likings. It covers the dispositions that give events their emotional meaning. In the analysis of emotions, concerns are pivotal. Emotions result from the interplay of an individual’s concerns and events confronting him or her. Emotions result from events appraised as relevant – as harmful or beneficial – to one’s concerns (Frijda, 1986; Oatley, 1992; Stein and Trabasso, 1992). No concern, no emotion. When someone dies whom one does not care about, the reaction is not very emotional. Grief comes when someone dies whom one loved. Properties of the concerns to which an event is relevant (that is, is appraised as relevant) determine properties of the emotion. The strength or centrality of the concern determines emotional intensity. So does the number of concerns to which a given event is appraised as relevant. Both variables have been demonstrated to be correlated to rated emotional intensity (Sonnemans and Frijda, 1995). Note that most events are relevant to more than one concern; most events have multiple meanings. Loss of a love partner is emotionally so serious because it represents loss of company, loss of a place for confidence, loss of sex, and loss of familiar social surroundings. The point is of importance because so many events have relevance for concerns less


55

obvious than the central one. An event conducive to jealousy not only means a threat to one’s relationship, but, in addition, to self-esteem and to a desire to be in control. Often those secondary meanings are what gives the event its sting. An individual’s concerns change and grow in number over a lifetime. Their range increases, and so does the number of objects that appeal to a given concern (people whom one likes, ideals that one strives for, the ‘‘surface concerns’’). The set of concerns can be said to constitute the individual’s ‘‘self’’; the term ‘‘self’’ is indeed often used in that sense. The word ‘‘self’’ in this sense means something quite different from one’s self-schema or self-conception; the former can exist without the latter. One is this self; one does not have such a self. The fact that the same word is used for both is confusing. Emotions always reflect the self in the former sense (except perhaps when the concerns are only focused at their most impersonal, general, and elementary level). They need not involve ‘‘self’’ in the second sense. Nor do emotions necessarily presuppose the cognitive complexities that are involved in self-awareness, let alone the self-awareness that goes beyond the formal one that underlies recognizing another human individual as similar. One can of course take stock of one’s concerns, and start to act accordingly. The self as the sum-total of one’s concerns, as one knows them, can become an object of thought, and it can be used to guide behavior. In line with this, people also have concerns that have one’s own person or one or more of its properties as their target. The most obvious examples are strivings for self-esteem and for social regard, and for maintaining and enhancing identity. Identity striving is one of those concerns that operate at the background of encounters with different superficial meanings, such as the defense of a power position, reaching a given goal, or undergoing humiliating treatment (Frijda, 1994). Still, the role of the self in self-regard in a strict psychological sense is, however, not easy to assess. Self-enhancing self-presentation in erotic contexts need not have anything to do with it. Even the male blue lizard self-presents to females in a manner that is not entirely unlike selfpresentation in the human male under similar circumstances (Maclean, 1990). Social-status strivings are prominent among male baboons and among chimpanzees, but only the latter show evidence of selfawareness in mirror experiments and self-decoration (De Waal, 1996). Efforts at status achievement and maintenance, whether in lizards, monkeys, or humans may well be driven merely by momentary desires for social-dominance achievements. On the other hand, the distress at being dethroned as an alpha male chimpanzee is so poignant that damage to self-conception is hard to deny (De Waal, 1982). The lines are

56

Nico H. Frijda

difficult to draw, and research on children as well as animals would appear to be needed. In any case, concerns for one’s personal value and valuation do emerge as results of development. They are of obvious importance as guiding powers in individual as well as in social life (Carver and Scheier, 1990), where they give rise to the whole gamut of the more complete forms of the emotions of pride, contempt, shame, and guilt. Competence All this is relevant to yet another aspect of the self as person and his or her properties that is an essential factor in shaping emotions, that is, coping competence and trust in one’s coping resources, or self-efficacy (Bandura, 1986). Emotions are a function of three classes of variables: an eliciting event as appraised, the individual’s concerns that give that event its emotional impact, and the individual’s repertoire of action for dealing with the event. This is one of the most general statements about emotions that can be made. Imagine an event such as losing one’s physical balance. It threatens one’s concern for absence of pain, as well as that for knowing what is going to happen next and for being able to deal with that. Being able to deal with what is going to happen next partly depends on being able to deal with uncertainty. When the latter ability is weak, loss of balance is a much more upsetting event. When it is strong, it may be a suspenseful event, a challenge. You straighten up with a laugh. When in addition one knows perfectly well that it will end innocently, it may just be felt as a loss of time. This, of course, describes the development of how infants between three and seven months of age respond to being lifted, thrown up, and caught again. First they cry or look anxiously; then they laugh; then they just want to be put down again (Sroufe and Waters, 1976). This threefold relationship holds for all emotions. Satisfactions that come as a matter of course usually fail to evoke joy or happiness. Satisfactions give positive emotions when they come after uncertainty about whether they will come or whether one can indeed avail oneself of them, and their intensity is proportional to the previous uncertainty (Ortony, Clore, and Collins, 1988). Trust in one’s competence is in principle entirely implicit in expectations concerning the outcome of an encounter. There need be no ‘‘selfimage,’’ no comparison between what one would want to do and what one thinks one can accomplish. The object one confronts simply looks weak or powerful, dangerous or challenging, within reach or out of reach, in the projective way described earlier. Self-confidence in this sense is open to animals and human infants as well as to human adults.


57

Mice that have been victorious in previous antagonistic encounters are more aggressive than those who have lost a fight. There also are innate differences between daring and fearful mice and rats. Assertive behavior in chimpanzees goes up and down with position in the social hierarchy, which itself is dependent upon earlier success in power contests (De Waal, 1982). Conclusions In this chapter I examined the role of the self in emotions. We encountered reference to the person who experiences emotions in different roles. It appeared useful or necessary to distinguish: (a) the transcendental self, as a logical premise of the causal as well as the phenomenological analysis of emotions; (b) the self as an equivalent of the person with given properties, notably competences and concerns in terms of which events can obtain emotional meanings; (c) ‘‘I as actor,’’ a functional formulation for an individual’s spontaneity and intentional activity, and the basis for the emotions that emerge when events interfere with that spontaneity; and (d) a notion of ‘‘self’’ experienced as experiencer and as an object of certain emotions. A transcendental self and oneself as a person with concerns and other properties are logical prerequisites of emotions but need not be elements of experience. They even need not have informational content, because emotions may be caused by strivings and sensitivities that are simply there, as the mechanisms that cause goal-directed behavior. ‘‘I as actor’’ is at first implicit in experience, mainly through the pleasures of spontaneous action and the pains of interferences. It becomes a content of awareness only gradually, presumably when language provides a reference in the form of a name or a pronoun. Oneself as a person, and thus a ‘‘self’’ with properties, would, from the analysis of emotions, only appear much later, and much later than certain emotion concepts like shame and pride would seem to suggest. It would be helpful if separate concepts were developed for the various notions that the word ‘‘self’’ points to, at some stage.

COMMENTARY

The self and emotions Seymour Epstein

In order to properly examine the relation between emotions and the self-concept, Frijda believes it is necessary to distinguish between the self as object of knowledge and the transcendental self. He makes a compelling case for the importance of the self as object of knowledge, but I have trouble understanding exactly what he means by the transcendental self. At one point he defines it as ‘‘the sum total of the information processes at a functional level’’ and at another as a person’s ‘‘set of concerns.’’ These are both vague constructs that require clarification, and, unless I am missing something, they are not even the same. Others, as well as I, have proposed more clearly articulated distinctions between two somewhat similar constructs to the ones proposed by Frijda. For example, William James (1910) initially, and Gordon Allport (1955) later, expressed the difference between the two selves in terms of the self as an object of knowledge and the self as an executive self that is the source of behavior. Both, however, eventually disowned the executive self as scientifically indefensible, likening it to a homunculus residing inside a person’s head and directing the person’s behavior. The problem with the executive self and its homunculus analogy is that understanding their behavior is no simpler than understanding the behavior of the whole person and therefore does nothing to further our understanding. I will discuss a more meaningful representation of the executive self later, when I present my own views. Although I am not enamored with Frijda’s transcendental self, I am in agreement with almost everything he says about emotions, including the important role he assigns to appraisal in the instigation of almost all emotions, the possibility of the occurrence of emotions in the absence of representations of the self, and the view that the most central aspect of all emotions consists of an action tendency, such as flight for fear, withdrawal for depression, and attack for anger. However, I do not agree with him and many others, including Lazarus (1991a) and Beck (1976), that the specific instigator of an emotion is the appraisal of a stimulus situation. Rather, I believe it is the appraisal of a desirable action. For example, the appraisal of a situation as threatening does not 58

Commentary on Frijda

59

necessarily produce fear. If the perceived threat is viewed as unjustified and the perpetrator as vulnerable to attack, and this results in an appraisal of attack as the most desirable response option, the emotion will be anger rather than fear. Moreover, to be logically consistent, once it is assumed that the essence of an emotion is an action tendency, then the implicit appraisal of an action tendency should be the critical instigator of a specific emotion. Of course, certain stimulus situations are so closely associated with specific action tendencies that the appraisals of these situations are likely to be predictive of the appraisals of the preferred action. This near correspondence between the two kinds of appraisal, no doubt, is the source of the conceptual confusion that is prevalent on this issue. Frijda emphasizes the importance of ‘‘concerns’’ in producing emotions. I agree with him that, in the absence of concerns, there is no emotion. However, unlike Frijda, I distinguish between concerns (which I refer to as motives), in two different information-processing systems that I refer to as the experiential and rational systems. It is only the motives in the experiential system that are associated with emotions. Intellectual concerns that reside in the rational system, unless they are also associated with experiential concerns, do not elicit emotions. In fact, one way of distinguishing between motives in the two systems is by observing whether or not they are accompanied by emotions. I will have more to say about this shortly. It is time now to turn to a consideration of my own views on the executive self. In response to the editors’ request for commentators to compare their views to those expressed in the chapters they reviewed, I will briefly describe the thoughts that I had about the self-concept many years ago that led to the introduction of a new global theory of personality (Epstein, 1973) that I have since labeled cognitive–experiential selftheory (CEST). At the time, I gave the matter of an executive self considerable thought because I intuitively felt there had to be something like it but that it needed to be better articulated. I posed the following riddle for myself: ‘‘What is it that has neither material substance nor fixed content, yet can interpret and organize experience, direct behavior, interact with the environment, and grow in the process, all properties that have been attributed to an executive self?’’ When the answer came to me, it not only solved the problem of what the executive self is, but it became the nucleus of a highly integrative, psychodynamic theory of personality, CEST, which, unlike other selftheories that emphasized a single cognitive system and ignored emotions, proposed two cognitive systems, one that operates consciously and analytically, and is relatively affect-free, and the other that operates preconsciously and intuitively, and is intimately associated with affect.

60

Seymour Epstein

The solution to the riddle of the executive self lies in the recognition that it corresponds to an implicit personal theory of reality that, like a scientific theory, organizes experience, directs behavior, and grows through its interaction with the data of experience. An implicit theory of reality, according to CEST, resides in the experiential system and includes a self-theory, a world-theory, and their interaction. As an implicit theory that operates automatically and preconsciously, the self-theory in a person’s experiential system must be distinguished from a person’s conscious explicit self-theory in the rational system. Thus, according to CEST, a person has two self-theories, and therefore two self-concepts, one in the experiential system and one in the rational system. The two may correspond or diverge to different degrees, with the degree of divergence being an important source of stress and maladjustment. I believe that this conceptualization of the self-concept along with other aspects of CEST go a long way toward resolving the problems concerning the self that the editors listed in the introduction to this volume requiring a solution. Following are the questions presented by the editors and the solutions proposed by CEST: 1. What is the role of emotions in the self-concept? According to CEST, emotions are intimately associated with the operation of the experiential system (in which the experiential self-concept resides). First they are considered to act as a barometer of the significance of events in a person’s self-concept. Accordingly, by noting the events that trigger emotions, the significant schemas in a person’s selfconcept can be inferred. For example, if I react with greater emotion to an assault on my appearance than to an assault on my intelligence, it can be inferred that my appearance is more important to me (in my experiential system) than my intelligence, no matter what I may consciously believe (in my rational system). Second, since particular emotions are instigated by corresponding appraisals, it is possible to infer the schemas in a person’s self-system from a person’s characteristic emotions. Thus, knowing that a person is characteristically angry (over a representative sample of life events) suggests that the person (in his or her experiential system) tends to view himself or herself as good and just and other people as bad and unjust. Third, positive and negative affect, which are most often determined by automatic appraisals in the experiential system, bias conscious thinking in a person’s rational system. Through this mechanism the experiential self-system routinely influences the rational self-concept in the absence of awareness. The implications of this for the widespread presence of human irrationality and for the meaning of ‘‘knowing onself’’ are evident. Fourth, emotions reinforce schemas in both the experiential and rational self-concepts.


61

2. How do emotions and cognitions influence each other? This question has been answered in my response to the first question. 3. What direction does this influence take? As indicated in my response to the first question, the influence is reciprocal. That is, emotions determine which cognitions are reinforced, and preconscious cognitions determine which emotions are experienced. Also, conscious, reflective cognitions are often used to correct preconscious, automatic cognitions that are the source of maladaptive emotions and behavioral tendencies. Further, the relatively primitive emotions available in early development serve as nuclei around which broader and more differentiated cognitive-emotional networks develop. 4. What kinds of events are relevant to an individual’s sense of self and why do the same events influence different individuals differently? I already indicated that the degree of relevance of an event for a person’s experiential self-concept is indicated by the degree of emotion it arouses. In turn, the emotion-eliciting properties of events are determined by certain inherent properties (e.g. a universal fear of heights) and by their acquired significance as a result of previous experiences. As experiences differ for different individuals, people acquire different emotional sensitivities to the same events. Of additional relevance to this question is the assumption in CEST that there are four basic beliefs that develop as a result of their association with four basic needs. The four basic needs include the need to maximize pleasure and minimize pain, the need to assimilate the data of reality (by maintaining and extending one’s implicit theory of reality), the need for relatedness, and the need to enhance one’s self-esteem. The four related beliefs, respectively, include the belief in the degree to which the world is benign relative to malevolent; the belief in the degree to which the world is organized, predictable, and controllable relative to chaotic, unpredictable, and uncontrollable; the belief in the degree to which people are supportive, trustworthy, and loving relative to dangerous, untrustworthy, and rejecting, and the belief in the degree to which one is worthy, lovable, and competent relative to unworthy, unlovable, and incompetent. Any event that is appraised as relevant to these needs or beliefs is experienced as significant to the self and therefore produces an emotional response. 5. What processes can account for stability and change in selfperception and self-evaluation over different time spans? The selfsystem is hierarchically organized. Central concepts are highly stable, whereas peripheral ones readily change. Holding centrality constant, although momentary changes occur frequently, they tend to occur around a relatively stable mean. Thus, greater change is observed in perceptual, evaluative, behavioral, and any other kind of response

62

Seymour Epstein

when measures are aggregated (intuitively or objectively) over shorter than over longer time spans. 6. What processes are responsible for long-term developmental changes from infancy to adulthood? Change occurs as a result of the interaction of maturation and emotionally significant experiences. There are critical periods in life in infancy, childhood, adolescence, young adulthood, middle age, and old age, each with its characteristic adaptive demands, as well as unique, individual experiences (e.g. traumas; transforming love relationships), that, depending on their strength, duration, and repetition, have the capacity to produce longterm changes in a person’s conception of self and world, and therefore in a person’s personality. 7. Is fundamental and enduring change possible in adulthood, and, if so, what mechanisms are responsible for such change? Such changes can occur when there are emotionally significant experiences of sufficient intensity, duration, or repetition to change fundamental schemas in a person’s experiential self-concept and/or world-concept. The editors concluded that these questions have not yet been adequately addressed because of a lack of an integrative framework for considering them in combination. I submit that CEST provides just such a framework. Moreover, not only have these issues been addressed in CEST at a theoretical level, but many of the assumptions in CEST have been supported by an extensive body of research conducted over the past twenty-six years (see review in Epstein and Pacini, 1999). In addition to providing answers to the above questions, CEST identifies and provides a solution to a fundamental issue that the editors have overlooked, and which, until it is resolved, makes a resolution of the other problems impossible. The issue concerns the difference between implicit versus explicit cognitions, or, if you prefer, conscious versus preconscious or unconscious processing. According to CEST, until it is recognized that there are two self-systems, preconscious experiential and conscious rational, that operate by different rules and contain different schemas (sometimes coinciding and sometimes discrepant), it is impossible to answer questions about the self because the answers differ for the two selves. It is beyond the scope of this brief comment to discuss these systems or the responses to the above questions in greater detail. The interested reader can obtain further information in a number of publications that are readily available in the literature and can also be obtained by writing to the author (e.g. Epstein, 1983, 1984, 1985, 1989, 1990, 1991a, 1991b, 1992b, 1993a, 1993b, 1994, 1998a, 1998b, 1998c; Epstein and Pacini, 1999). In summary, Frijda has provided several important insights about the relation of the self-concept and emotions, which is all he was


63

requested to do. In contrast, my efforts have been devoted to constructing a global theory of personality in which the self-concept, emotions, motivation, levels of processing, and their interactions are central constructs.

COMMENTARY

The self and emotions Seymour Epstein

In order to properly examine the relation between emotions and the self-concept, Frijda believes it is necessary to distinguish between the self as object of knowledge and the transcendental self. He makes a compelling case for the importance of the self as object of knowledge, but I have trouble understanding exactly what he means by the transcendental self. At one point he defines it as ‘‘the sum total of the information processes at a functional level’’ and at another as a person’s ‘‘set of concerns.’’ These are both vague constructs that require clarification, and, unless I am missing something, they are not even the same. Others, as well as I, have proposed more clearly articulated distinctions between two somewhat similar constructs to the ones proposed by Frijda. For example, William James (1910) initially, and Gordon Allport (1955) later, expressed the difference between the two selves in terms of the self as an object of knowledge and the self as an executive self that is the source of behavior. Both, however, eventually disowned the executive self as scientifically indefensible, likening it to a homunculus residing inside a person’s head and directing the person’s behavior. The problem with the executive self and its homunculus analogy is that understanding their behavior is no simpler than understanding the behavior of the whole person and therefore does nothing to further our understanding. I will discuss a more meaningful representation of the executive self later, when I present my own views. Although I am not enamored with Frijda’s transcendental self, I am in agreement with almost everything he says about emotions, including the important role he assigns to appraisal in the instigation of almost all emotions, the possibility of the occurrence of emotions in the absence of representations of the self, and the view that the most central aspect of all emotions consists of an action tendency, such as flight for fear, withdrawal for depression, and attack for anger. However, I do not agree with him and many others, including Lazarus (1991a) and Beck (1976), that the specific instigator of an emotion is the appraisal of a stimulus situation. Rather, I believe it is the appraisal of a desirable action. For example, the appraisal of a situation as threatening does not 58


59

necessarily produce fear. If the perceived threat is viewed as unjustified and the perpetrator as vulnerable to attack, and this results in an appraisal of attack as the most desirable response option, the emotion will be anger rather than fear. Moreover, to be logically consistent, once it is assumed that the essence of an emotion is an action tendency, then the implicit appraisal of an action tendency should be the critical instigator of a specific emotion. Of course, certain stimulus situations are so closely associated with specific action tendencies that the appraisals of these situations are likely to be predictive of the appraisals of the preferred action. This near correspondence between the two kinds of appraisal, no doubt, is the source of the conceptual confusion that is prevalent on this issue. Frijda emphasizes the importance of ‘‘concerns’’ in producing emotions. I agree with him that, in the absence of concerns, there is no emotion. However, unlike Frijda, I distinguish between concerns (which I refer to as motives), in two different information-processing systems that I refer to as the experiential and rational systems. It is only the motives in the experiential system that are associated with emotions. Intellectual concerns that reside in the rational system, unless they are also associated with experiential concerns, do not elicit emotions. In fact, one way of distinguishing between motives in the two systems is by observing whether or not they are accompanied by emotions. I will have more to say about this shortly. It is time now to turn to a consideration of my own views on the executive self. In response to the editors’ request for commentators to compare their views to those expressed in the chapters they reviewed, I will briefly describe the thoughts that I had about the self-concept many years ago that led to the introduction of a new global theory of personality (Epstein, 1973) that I have since labeled cognitive–experiential selftheory (CEST). At the time, I gave the matter of an executive self considerable thought because I intuitively felt there had to be something like it but that it needed to be better articulated. I posed the following riddle for myself: ‘‘What is it that has neither material substance nor fixed content, yet can interpret and organize experience, direct behavior, interact with the environment, and grow in the process, all properties that have been attributed to an executive self?’’ When the answer came to me, it not only solved the problem of what the executive self is, but it became the nucleus of a highly integrative, psychodynamic theory of personality, CEST, which, unlike other selftheories that emphasized a single cognitive system and ignored emotions, proposed two cognitive systems, one that operates consciously and analytically, and is relatively affect-free, and the other that operates preconsciously and intuitively, and is intimately associated with affect.

60

Seymour Epstein

The solution to the riddle of the executive self lies in the recognition that it corresponds to an implicit personal theory of reality that, like a scientific theory, organizes experience, directs behavior, and grows through its interaction with the data of experience. An implicit theory of reality, according to CEST, resides in the experiential system and includes a self-theory, a world-theory, and their interaction. As an implicit theory that operates automatically and preconsciously, the self-theory in a person’s experiential system must be distinguished from a person’s conscious explicit self-theory in the rational system. Thus, according to CEST, a person has two self-theories, and therefore two self-concepts, one in the experiential system and one in the rational system. The two may correspond or diverge to different degrees, with the degree of divergence being an important source of stress and maladjustment. I believe that this conceptualization of the self-concept along with other aspects of CEST go a long way toward resolving the problems concerning the self that the editors listed in the introduction to this volume requiring a solution. Following are the questions presented by the editors and the solutions proposed by CEST: 1. What is the role of emotions in the self-concept? According to CEST, emotions are intimately associated with the operation of the experiential system (in which the experiential self-concept resides). First they are considered to act as a barometer of the significance of events in a person’s self-concept. Accordingly, by noting the events that trigger emotions, the significant schemas in a person’s selfconcept can be inferred. For example, if I react with greater emotion to an assault on my appearance than to an assault on my intelligence, it can be inferred that my appearance is more important to me (in my experiential system) than my intelligence, no matter what I may consciously believe (in my rational system). Second, since particular emotions are instigated by corresponding appraisals, it is possible to infer the schemas in a person’s self-system from a person’s characteristic emotions. Thus, knowing that a person is characteristically angry (over a representative sample of life events) suggests that the person (in his or her experiential system) tends to view himself or herself as good and just and other people as bad and unjust. Third, positive and negative affect, which are most often determined by automatic appraisals in the experiential system, bias conscious thinking in a person’s rational system. Through this mechanism the experiential self-system routinely influences the rational self-concept in the absence of awareness. The implications of this for the widespread presence of human irrationality and for the meaning of ‘‘knowing onself’’ are evident. Fourth, emotions reinforce schemas in both the experiential and rational self-concepts.


61

2. How do emotions and cognitions influence each other? This question has been answered in my response to the first question. 3. What direction does this influence take? As indicated in my response to the first question, the influence is reciprocal. That is, emotions determine which cognitions are reinforced, and preconscious cognitions determine which emotions are experienced. Also, conscious, reflective cognitions are often used to correct preconscious, automatic cognitions that are the source of maladaptive emotions and behavioral tendencies. Further, the relatively primitive emotions available in early development serve as nuclei around which broader and more differentiated cognitive-emotional networks develop. 4. What kinds of events are relevant to an individual’s sense of self and why do the same events influence different individuals differently? I already indicated that the degree of relevance of an event for a person’s experiential self-concept is indicated by the degree of emotion it arouses. In turn, the emotion-eliciting properties of events are determined by certain inherent properties (e.g. a universal fear of heights) and by their acquired significance as a result of previous experiences. As experiences differ for different individuals, people acquire different emotional sensitivities to the same events. Of additional relevance to this question is the assumption in CEST that there are four basic beliefs that develop as a result of their association with four basic needs. The four basic needs include the need to maximize pleasure and minimize pain, the need to assimilate the data of reality (by maintaining and extending one’s implicit theory of reality), the need for relatedness, and the need to enhance one’s self-esteem. The four related beliefs, respectively, include the belief in the degree to which the world is benign relative to malevolent; the belief in the degree to which the world is organized, predictable, and controllable relative to chaotic, unpredictable, and uncontrollable; the belief in the degree to which people are supportive, trustworthy, and loving relative to dangerous, untrustworthy, and rejecting, and the belief in the degree to which one is worthy, lovable, and competent relative to unworthy, unlovable, and incompetent. Any event that is appraised as relevant to these needs or beliefs is experienced as significant to the self and therefore produces an emotional response. 5. What processes can account for stability and change in selfperception and self-evaluation over different time spans? The selfsystem is hierarchically organized. Central concepts are highly stable, whereas peripheral ones readily change. Holding centrality constant, although momentary changes occur frequently, they tend to occur around a relatively stable mean. Thus, greater change is observed in perceptual, evaluative, behavioral, and any other kind of response

62

Seymour Epstein

when measures are aggregated (intuitively or objectively) over shorter than over longer time spans. 6. What processes are responsible for long-term developmental changes from infancy to adulthood? Change occurs as a result of the interaction of maturation and emotionally significant experiences. There are critical periods in life in infancy, childhood, adolescence, young adulthood, middle age, and old age, each with its characteristic adaptive demands, as well as unique, individual experiences (e.g. traumas; transforming love relationships), that, depending on their strength, duration, and repetition, have the capacity to produce longterm changes in a person’s conception of self and world, and therefore in a person’s personality. 7. Is fundamental and enduring change possible in adulthood, and, if so, what mechanisms are responsible for such change? Such changes can occur when there are emotionally significant experiences of sufficient intensity, duration, or repetition to change fundamental schemas in a person’s experiential self-concept and/or world-concept. The editors concluded that these questions have not yet been adequately addressed because of a lack of an integrative framework for considering them in combination. I submit that CEST provides just such a framework. Moreover, not only have these issues been addressed in CEST at a theoretical level, but many of the assumptions in CEST have been supported by an extensive body of research conducted over the past twenty-six years (see review in Epstein and Pacini, 1999). In addition to providing answers to the above questions, CEST identifies and provides a solution to a fundamental issue that the editors have overlooked, and which, until it is resolved, makes a resolution of the other problems impossible. The issue concerns the difference between implicit versus explicit cognitions, or, if you prefer, conscious versus preconscious or unconscious processing. According to CEST, until it is recognized that there are two self-systems, preconscious experiential and conscious rational, that operate by different rules and contain different schemas (sometimes coinciding and sometimes discrepant), it is impossible to answer questions about the self because the answers differ for the two selves. It is beyond the scope of this brief comment to discuss these systems or the responses to the above questions in greater detail. The interested reader can obtain further information in a number of publications that are readily available in the literature and can also be obtained by writing to the author (e.g. Epstein, 1983, 1984, 1985, 1989, 1990, 1991a, 1991b, 1992b, 1993a, 1993b, 1994, 1998a, 1998b, 1998c; Epstein and Pacini, 1999). In summary, Frijda has provided several important insights about the relation of the self-concept and emotions, which is all he was


63

requested to do. In contrast, my efforts have been devoted to constructing a global theory of personality in which the self-concept, emotions, motivation, levels of processing, and their interactions are central constructs.

C HA P T E R 4

Fish, foxes, and talking in the classroom: introducing dynamic systems concepts and approaches Paul L. C. Van Geert Dynamic system: a general definition The phrase dynamic system contains the words dynamic – referring to the Greek ‘‘dynamikos,’’ which means ‘‘powerful’’ – and system – a word that stems from a Greek verb that means ‘‘to combine.’’ Thus, a dynamic system is a combination of things to which certain powers or forces apply. If a force is applied to something, it moves or changes (unless the force is counteracted). Defined in this way, the term dynamic system has an extremely broad meaning. For instance, a bunch of dry leaves blown by the autumn wind can already be considered a dynamic system. It consists of a collection of leaves all subject to the same external force, the wind. In order to avoid such trivial applications, we should confine the term to something that is more conceptually appealing. Let us begin by confining the notion of system to collections of things that are related to one another in a way that corresponds with the notion of dynamic, that is force- or power-related. We shall call something a dynamic system if it consists of elements that exert specific influences or forces upon one another and, by doing so, change each other’s and their own properties (for general, technical introductions to dynamic systems, see, among others, Beltrami, 1987; Jackson, 1991a, b). In this chapter I shall adopt examples from different fields: physics, biology, and, of course, psychology. My point is to show that dynamic principles apply to systems, irrespective of those systems’ actual form or nature. What matters are the relationships, not the content matter. Dynamic systems’ thinking is basically a way of thinking about systems, not about psychology per se. Its importance for the field of psychology is that psychology, on the one hand, abounds with interesting opportunities for applying systems principles, while, on the other hand, systems thinking has not yet fully permeated this field of inquiry (for applications of dynamic systems thinking to psychology, see, among others, Abraham, 1990; Levine and Fitzgerald, 1992a, b; Molenaar and Newell, 1998; Port and Van Gelder, 1995; Smith and Thelen, 1993; Thelen and Smith, 1994; Van Geert, 1999). 64

Introducing dynamic systems concepts and approaches

65

Because we shall wander through different universes, let me begin with a little Star Trek. Assume we join Captain Picard on one of his dramatic missions. One day we arrive at that part of Deep Space known as the Newtonian Universe. There we spot two freely moving bodies, exerting gravitational forces upon one another and moving in an endless complicated swirl. Since this is the Newtonian Universe – says Captain Picard – the two bodies are completely isolated from anything else in the Cosmos. The two bodies surely constitute a dynamic system. They affect one another through gravitational forces and by doing so change their position in space. The result, it turns out, is a simple, cyclical motion. But then our Captain breaks in and introduces a third celestial body and immediately closes the Newtonian Universe again, leaving the three bodies on their own. But now we observe something quite different from the simple cycle: the three bodies move around each other in a complicated swirl that never seems to repeat itself. The three-body problem has been notoriously difficult to understand, in spite of the fact that the two-body problem – which is only one less – was the basic Newtonian idealization of a gravitational system. It was the French mathematician Poincare´ who presented a formal approach to this and similar problems and by doing so paved the way for dynamic systems theory (Jackson, 1991a; Stewart, 1989). Dynamic systems: general properties The three-body problem illustrates one of the most appealing features of dynamic systems theory: you don’t need complicated systems to arrive at complicated behavior. It illustrates several other properties that I shall explain further. It shows, among other things, that complicated patterns such as the motion of the three celestial bodies can be produced by a few interacting elements. An external guidance plan, leading the paths in their various directions, is not necessary. By watching other simple systems evolve over time we observe that the nature of the systems’ behavior varies widely. The variations are caused by the way the elements interact, they are not caused by varying external forces (this is not to deny the possibility, of course, that external forces can eventually cause similar effects). However much the patterns may differ, they are not just chaos. Most of the patterns produced by interacting elements that exchange forces are not just mumbo-jumbo variation. They display both complexity and regularity that intuitively appeals to the human eye. The ideal self-contained Newtonian Universe visited by our Star Trek crew lies of course outside the realm of the real. There is no part of the

66


existing world that is completely isolated from all the rest. Even the three celestial bodies deep in outer space will be subject to gravitational and other forces from outside. The issue is, however, whether or not it makes sense to treat the three objects as a semi-isolated system where the most important forces are those that the objects exert on one another. It makes sense to do so if the introduction of the dynamic systems notion allows one to come up with a better explanation of the behavior of the system than some other approach would allow. The question is, however, what kind of explanation is this? Before introducing applications to psychology, let us proceed to some biological examples (see, among others, Burghess and Wood, 1985; Hofbauer and Sigmund, 1988; Kingsland, 1995; Murray, 1989). In 1926, an Italian biologist by the name of Humberto d’Ancona made a study of certain fish populations in the Adriatic Sea and observed a pattern of increases and decreases in populations of fish. How could this pattern be explained? Did populations of fish have some inbuilt, genetic tendency to rise and fall in an endless cycle? Was there some unknown disease that killed fish in a cyclical pattern? D’Ancona turned to his father-in-law, Vito Volterra, who was a famous mathematician. Volterra transformed the problem into a dynamic systems issue by treating the rise and fall of the populations as the result of an interaction between predator and prey fish (which his son-in-law had already observed from the biological point of view). Dynamic systems: an example from biology One can conceptually isolate the system by assuming that prey fish thrive on some unspecified food resource that, for the sake of simplicity, is viewed as something that comes in more or less constant quantities. Prey fish die because they are eaten by predator fish (mackerel for instance). Predator fish, in turn, thrive on the prey fish and the more prey there is to catch the more predators will live. Predators die because of some external cause, the nature of which need not be taken into account for the moment. For the sake of simplicity, we can treat the causes of the predators’ deaths as a constant factor. Thus Volterra and d’Ancona’s fish system is a semi-closed system. It has a constant inflow of food for the prey fish and a constant outflow of dead predator fish. How could one ever expect to find a pattern of oscillations if both the inflow and outflow are a simple linear, constant flux (see figure 4.1)? Here’s where the dynamic nature of the interaction between predator and prey fish comes in. Volterra showed that the interaction that I specified above in verbal terms (prey eating food, dying because they


67

Figure 4.1. Volterra and d’Ancona’s model of population changes in predator and prey fish; in order to make the populations graphically comparable, relative population sizes have been used

were eaten by predators, predators thriving on prey and dying from some external cause) could be described in simple mathematical terms. The solution took the form of two coupled differential equations. One equation described how predators were affected by prey, the other specified how prey was affected by predators. Dynamic systems and geometric representation A system like this one – and any other dynamic system for that matter – can easily be represented in a geometric form. Consider the number of predators – in a specific region of the Adriatic Sea – as one dimension and the number of prey fish as another dimension. What we get then is a simple plane with two coordinates, one for the prey and one for the predators. Each point in the plane represents a specific population size of prey and a population size of predator fish. The differential equations mentioned above are used to calculate how the two populations will evolve over a specific amount of time. The series of changing populations sizes – which are points in the plane – specify trajectories. The characteristic trajectory of a predator–prey dynamics is a somewhat deformed cycle (see figure 4.2). This example demonstrates various aspects of what a dynamic systems approach to a phenomenon usually entails. Let us now take a closer look at those aspects.

68


Figure 4.2. Geometric representation of the predator and prey populations over time

Dynamic systems as simplifying, conceptually semi-closed structures First, the example makes clear that the system is in fact a conceptual construction. More precisely, we conceptually isolate some part of reality (say the predator and the prey fish). By doing so, we create a system (the relationship between the predator and prey fish) and its environment. The environment is basically anything else in the universe, except the system. The relationship between the system and its environment is then simplified as much as possible. The easiest simplification is to assume that the exchange between the system and its environment is a constant flow. For instance, we assumed that the prey fish feed on a constant supply of food that comes from a source outside the predator–prey system (let us assume the prey fish feed on algae). We also assumed that the predator fish die at a constant rate (e.g. one fish out of ten dies per week). It is important to note that the assumptions about the system– environment exchange need not be empirically correct, in the sense that they do not need to refer to an actual empirical system. It suffices that the assumed exchange is empirically possible. Once we understand the system, we can apply actual empirical conditions to the system and see whether the system helps us understand the actual state of affairs (e.g. the food supply is variable, the death rate varies stochastically as a consequence of diseases or the presence of predators that feed on predators). To put it differently, the system is in fact a model that shows


69

some deep similarities with the reality it models but that otherwise simplifies and reduces that reality as much as possible. Being a conceptual construction, a dynamic system is basically a semi-closed (or semi-open, depending on one’s perspective) system. By treating the predator–prey relationship as an isolated event, we try to protect it as much as possible from interference from external sources, whatever their nature. However, given that the system refers to some part of the natural universe, it must have a connection with that universe. We have seen that this connection is conceptually reduced to the simplest form possible. Dynamic systems are characterized by a structure of interactions Once the system has been conceptually isolated, it must be specified by characterizing the time-dependent interaction of its components. The first step is to provide a conceptual description of the interaction. That description must confine itself to the most elementary and essential aspect of the interaction. For instance, we observe that, first, the prey fish population grows at some constant rate that directly reflects the constant external food supply; second, that the prey fish die at some constant rate that directly reflects the species’ natural longevity; third, that the prey fish die at some variable rate that reflects the number of predators that feeds on them; fourth, that the predator fish population grows as a function of the food supply, i.e. the prey fish; and, fifth, that the predator fish die at a rate that reflects the species’ longevity. This elementary conceptual description is of course based on empirical facts or at least on empirically justifiable assumptions. However, there are a lot of empirically established facts that can and in fact should be left out of the conceptual specification because they are not essential. For instance, prey fish not only die of old age or because a predator catches them, they also die from diseases. Diseases, however, are not explicitly mentioned in the model and are in fact treated as a constant. The art of building a dynamic systems model is to reduce the conceptual description to its bare minimum. What this bare minimum is depends on whether the minimal description of the relationships between the components suffices to make the system do what it is supposed to do. In the case of the predator and prey fish example, the system should produce the empirically observed cyclic, directed pattern of oscillations in the predator and prey population. As far as the minimum is concerned, it turns out that we can actually leave out the second assumption (the constant death rate of prey assumption): the remaining four suffice to produce the required pattern. Fish – whether or not swimming in the warm waters of the Adriatic

70


Sea are not exactly the psychologist’s staple. However, principles that govern dynamic systems do not depend on the actual content matter to which they apply. System properties apply to specific types of interaction and it does not really matter whether such interactions occur with biological organisms or psychological variables. Let me try to explain how the dynamics that characterized the nature of the interaction between prey and predator fish also apply to human behaviors and psychological variables. Some actions that people undertake lead to negative consequences. For instance, if a child talks too much or too loudly in the classroom, it is likely that the teacher will show her discontent, for instance by punishing the talking by a negative remark spoken in a loud voice. Assuming that such remarks are so-called aversive stimuli for the child in question and that the child experiences them as negative, for instance because they cause negative emotions in the child, we know that the punishment will contribute to the child’s avoidance of the punished behavior. However, the unwanted behavior – the talking in the classroom – has its own rewards and reinforcements, for instance because exchanging feelings and ideas with other children brings about positive emotions in the child and thus increases the frequency with which such behavior will occur in the future. All this amounts to the principles of simple, general learning theory (see, for instance, Novak, 1996). We also know that the so-called aversive stimuli, the teacher’s reprimands and angry looks, are related to the cause of the stimulus, which is the child’s talking. Moreover, it is reasonable to assume that the gravity of the aversive stimulus, let us say the seriousness of the punishment, is related to the frequency with which the unwanted behaviors actually occur. This relationship could be a simple linear one – the gravity of the punishment is proportional to the frequency of the unwanted talking – but it is unlikely that this would be so. Teachers have an image of their pupils and that image changes as a consequence of their experiences with the pupils (and their eventual biases or prejudices with regard to those pupils). The relationship between a pupil’s unwanted-talking frequency and the gravity of the punishment – let us say the intensity of negative emotions that accompany the teacher’s reprimands – is more probably one of change or growth over time. As the pupil’s frequency of unwanted behavior increases, the average seriousness of the teacher’s reprimands also grows bigger. On the other hand, we may also assume that the reprimands are subject to some form of habituation or, in learning-theoretical terms, extinction. The reprimands of the teacher do not always have the desired effect, in that the pupil still shows the unwanted behavior (its frequency is less than the frequency


71

it would have without punishment, but there is no possibility for the teacher to know what this frequency would have been and to reckon with it; that is, the unknown frequency cannot play a reinforcing or rewarding role with regard to increasing the frequency of the punishment behaviors). Thus we may assume that some amount of extinction will apply to the frequency of the punishment or to the gravity of each punishment event. What is the relationship between the fish, the talking in the classroom, and the reprimands? It is easy to see that the unwanted talking in the classroom is formally equivalent to the prey fish. The talking increases proportionally as a consequence of being rewarded and reinforced through social processes of the classmates, just as the prey population grows proportionally as a consequence of the availability of a food resource. On the other hand, the talking decreases as a consequence of the punishment (the reprimands) in line with the principle of avoidance learning. It decreases, first, as a consequence of the frequency of the punishment, which is linearly related to the frequency of the unwanted talking, and, second, as a consequence of the gravity of the punishment (more serious punishment causes more avoidance learning effect). The prey population, in turn, decreases because individual fish are being eaten by predators. The decrease depends, first, on how many fish are available as food for the predators (the prey population) and, second, on how many predators there are (the predator population). The variable seriousness-of-the-reprimands is formally similar to the variable size-of-the-predator-population. The seriousness or gravity of the reprimands increases proportionally as a consequence of the frequency of the unwanted talking in the classroom, similar to the predator population that increases proportionally to the available food supply, the prey fishes. The gravity, however, also decreases as a consequence of extinction or habituation, which directly acts upon the frequency of the punishments, the gravity, or both. This system of relationships can be compared to that applying to the predator fish, Their population increases proportionally as a consequence of the available food supply (the prey) but meanwhile decreases as a consequence of natural deaths of predator fish, which is of course also proportional to the predator population. In summary, a dynamic system model applies to a set of relationships between components of a system, not to the actual physical instantiation of those components. From that point of view, the relationships between predator and prey fish, on the one hand, and unwanted talking and reprimands, on the other hand, can in principle be described by a similar dynamic model.

72


From a structure of interactions to a mathematical model Now that we have obtained some minimal conceptual specification of the time-dependent interactions in the system under scrutiny, we are ready to specify those interactions in a mathematical form. I take the term ‘‘mathematical’’ in its broadest possible form, i.e. as any algorithmic procedure that can be used to generate certain results or outcomes. In general, the mathematical expression of the conceptual relationships is written in the language of differential calculus. I shall take the unwanted-talking–reprimands situation as the domain of application of the mathematical model, although it should be noted that the resulting model applies to any situation that is compliant with the system of interaction described in this particular situation and in the predator– prey case dB = B − BP dt dP = PB − P dt

equation 1

In spite of its – probably – somewhat unusual appearance (most readers will not be used to reading equations with Greek characters), the equation is in fact astonishingly simple. It says that any change in the frequency of unwanted talking over time (represented by B for ‘‘behavior’’; dB/dt means ‘‘the difference in the frequency of the behavior over some change in time’’) is equal to the increase in the behavioral frequency thanks to some factor alpha (which depends on the rewarding or reinforcing social effects of the talking to other children in the classroom and which can be represented by any reasonable constant) minus the decrease in the frequency of talking (a decrease that depends on the actual frequency of the reprimands, which is a linear function of the frequency of the unwanted talking B itself, the gravity of the punishment, P, and some constant that specifies the magnitude of the effect of punishment on the decrease in talking frequency; note that this constant, lambda, is the product of an effect size constant and additional constants, such as the punishment ratio schedule). The second part of the equation does a similar thing with the gravity of the punishment. It says that the change in the gravity of the punishment depends, first, on the automatic decrease in the gravity of the punishment due to extinction or habituation effects (specified by a constant beta; note that beta may affect both the gravity of the punishment itself and the frequency of the punishment: if the frequency decreases, the resulting average gravity of the punishment per unit time also diminishes); and, second, on the proportional increase in the gravity P of the punishment


73

due to the level of unwanted behaviors B on which it ‘‘feeds,’’ so to speak, and on a constant mu that moderates the effect size of P and B. The remarkable thing is that the complicated interaction between unwanted behavior and punishment or other aversive stimuli – or the natural interaction between predators and their prey, for that matter – can in fact be simplified to this extremely frugal mathematical specification. In spite of its incredible simplicity, it does capture the essence of the real dynamics between behaviors or predators and prey. From the mathematical model to the geometric relationship Given the equations that formally specify our model, we can proceed to the next step in exploring the dynamic model. It is easy to see that the model and the equation are related to a simple geometric structure, namely a two-dimensional plane. One dimension of the plane is the possible frequency of the unwanted talking behavior, B, the other is the possible gravity of the punishment, P. The beauty of combining the mathematical equation with the representation in the two-dimensional plane is that we can literally see how the talking and the punishment, or the predator and prey populations, for that matter, evolve across time. Since frequencies of behavior and level of gravity of punishments are not easy to imagine, I shall revert to the original biological predator– prey example (recall that the dynamics as such remain the same, irrespective of the domain of application; what applies to this domain are specific assumptions regarding characteristic levels and time units; such assumptions may differ from domain to domain). For the sake of simplicity, I shall use a classical predator–prey example, which consists of rabbits, represented by r and foxes, represented by f. Assume that, given the time interval is a year, we can make some reasonable estimations of the values of the alpha, beta, lambda, and mu parameters. Assuming that we know at least something about rabbit and fox populations, or about predator and prey fish in the Adriatic Sea, it shouldn’t be too difficult to arrive at an initial guess of the values of those parameters. The initial guess need not be empirically correct. It suffices that it is empirically possible, given our knowledge of the animals involved. Armed with these values, we can take any number of foxes and rabbits that we want and calculate how many rabbits and foxes – or predator and prey fish – we shall find after a year (dt). Again, we use our common-sense knowledge of those animals in order to set a reasonable initial guess. For instance, it is reasonable to assume that the number of rabbits is considerably bigger than the number of foxes (and not the other way round). For the sake of illustration, I shall assume that the parameter values are as follows: alpha .11, beta .2, lambda .01 and

74


Figure 4.3. Numbers of predators and prey in the predator–prey space over three years

mu .0002. The starting values are 1000 for the rabbit population and 13 for the fox population. With these values, it is easy to calculate the rabbit population next year (which is the actual population – 1000 animals – plus the change, which is given by .11 * 1000 – .01 * 13 * 1000), which is 980. The fox population next year will again be 13 (the actual population, 13, plus the change, which is –.2 * 13 + .0002 * 1000 * 13). Our equations tell us that two years later, the rabbit population will be 960.4 and the fox population will be 12.948 . . . Confronted with these numbers, I take the opportunity to make two important remarks. The first remark addresses the issue of geometric representation. If we temporarily forbear the ghostly .4 rabbit and .948 fox and take the numbers as they are, we can easily see why the dynamic model produces a geometric representation of the change in predator and prey populations. Since both the number of predators (foxes) and the number of prey (rabbits) are considered as dimensions of a two-dimensional space, the combinations of a rabbit and a fox population (1000 and 13, 980 and 13, 960.4 and 12.984) correspond with three separate points in the predator–prey plane. This is a simple application of the principle of the Cartesian plane (see figure 4.3). Since the points are connected by the predator–prey dynamic principle (the first pair of values produces the second, the second the third, etc.) we may connect the consecutive points by a line. By doing so, we produce a line in the predator–prey space. If we go on calculating the


75

Figure 4.4. Numbers of predators and prey in the predator–prey space over 100 years

predator and prey populations for, say, 100 consecutive years, we obtain a cyclical pattern (which, in the case of a model that operates on fixed steps of one year looks more or less like a menhir fallen on its side; see figure 4.4). This cyclical pattern corresponds with sinusoidal patterns over time of the population sizes of predators and prey respectively (see figure 4.1). The cyclical pattern in the predator– prey space is the typical geometric format of a predator–prey dynamics. If we experiment with various parameter values, we find in fact a structure of nested cycles, reminiscent of the structure of an onion (see figure 4.5). Note that a qualitatively similar pattern will apply to the unwantedtalking–punishment example, because the behavioral situation is based on a similar dynamics. Quantitative differences may apply to the size and form of the cycles, but not to the principle of cyclical evolution, which is a direct consequence of the current dynamics and of nothing else. Mathematical analysis versus modeling The second remark I want to make concerns the strange .984 fox and the .4 rabbit. Presumably .4 rabbit is what we may find on the plate of a carnivorous gourmet, but it is certainly not something that runs around in the wild. At this point we hit the distinction between mathematical

76


Figure 4.5. Geometric representation of the predator and prey populations over time; different cycles result from applying different starting values

analysis and modeling. From the point of view of mathematical analysis of a dynamic model, the dimensions are just numbers, not representations of rabbit and fox populations. Moreover, since the dynamic model is based on differential equations, they should be solved as differential equations. More precisely, the dr/dt specification is a description of continuous change, which can be found if dt (the time interval) approaches the zero limit. Approaching the predator–prey dynamics from the mathematical side, the American mathematician Lotka, together with Volterra, whom we have already mentioned, succeeded in solving the differential equations that form the backbone of any predator–prey relationship. The typical solution of a Lotka–Volterra system looks a bit like the pattern of layers of an onion, or more precisely, like those of a clove of garlic (see figure 4.5). From a biological point of view, however, a population consists of whole numbers of animals. This means that if we want to reckon with the fact that the Lotka–Volterra equations refer to numbers of animals instead of mere numbers, we must change the equations in order to account for this whole-number restriction. Put differently, if we want to model the fate of two populations we must take into account all factors that we consider relevant from the viewpoint of describing the aspect of reality in question. It goes without saying that a similar adaptation has to be made with models that specify discrete behavioral interactions, for instance, discrete behavioral events or discrete units of knowledge.


77

Accounting for the fact that populations – or discrete behavioral events, for that matter – consist of whole numbers of animals is just a first step. We might also wish to consider the fact that real dynamics involve a host of additional factors. For instance, with regard to the relationship between whiting (prey fish) and mackerel (predator), biologists discovered that there is a third party that complicates the predator–prey cycle. They found that the squid Illex feeds upon young mackerel, whereas the adult mackerel feed upon young squid. In addition, squid also are cannibalistic, which implies that they feed on themselves (speaking population-wise). Put differently, the predator–prey cycle consists of a complicated set of cycles and subcycles, each of which complies with the basic predator– prey structure described earlier. The behavioral case poses similar difficulties. For instance, neither the frequency nor the gravity of the punishments is really fixed. Although the model itself specifies only a single value for both variables, the actual values will fluctuate considerably. If the model provides a reasonable image of the process, the fluctuations should center on the computed values. One way to bridge the gap between a mathematical model – for instance the predator–prey model – and a process model – that also accounts for additional factors – is to add random factors to the mathematical model, in an attempt to delineate the potential effects of random or accidental variation. By randomizing the parameters and variables, one can carry out a sensitivity analysis of the model, to determine how sensitively the model behaves with regard to empirically acceptable variations. Models that are too sensitive, i.e. that produce strongly varying outcomes as a result of minor random variations, should in general not be considered empirically adequate. Note, though, that this rule does not hold if the variations occur in the vicinity of bifurcation points, i.e. places in the space of parameters that lie at the boundary of different attractor regions (different equilibria). In spite of these complications, however, two conclusions remain valid. The first is that complicated dynamics may be modeled on the basis of the relationships exemplified in the simple dynamics. The second is that the simplifications, for instance of the most elementary predator–prey dynamics, nevertheless cover a considerable part of the actual population phenomena. In summary, we have now seen that a dynamic system is in fact a conceptual structure that specifies a semi-closed system. It isolates a particular interaction that takes place between particular parts or aspects of the world and specifies the relationship with the rest of the world in terms of the simplest possible flow. In the case of the unwanted behavior-aversive stimuli model (or the predator–prey model

78


for that matter), the isolated property concerns the interaction between an unwanted behavior and the gravity of an aversive stimulus (punishment) and describes how both the unwanted behavior and the aversive stimulus change as a consequence of this interaction. The relationship with the rest of the world concerns everything in the model that is not specified by the interaction term. In the behavioral case of the predator– prey equations described earlier, this amounts, for instance, to all conditions that make the unwanted behavior change in frequency (e.g. the social rewards the child gets by talking to his or her classmates). Dynamic systems and self-organization Given the distinction between the internal interaction component and the external–internal flow component (food, deaths, . . .), we can now proceed to an important, if not central property of dynamic systems, namely that of emergent properties and self-organization. Self-organization is sometimes treated as a somewhat mystical property. In this overview I shall try to explain it in its simplest possible form, although it remains true that many forms of self-organization are amazing in their complexity (Holland, 1998; Kauffman, 1993; Prigogine and Stengers, 1984; van Geert, 1989). I shall begin with the first and second laws of thermodynamics (Atkins, 1984). Whereas the first law concerns the overall conservation of energy, the second law states that physical systems move spontaneously and irreversibly toward a state of disorder. For instance, when I get myself a cup of freshly made, hot tea and leave the cup on the table, physically speaking I start with a state of increased order. There exists a concentration of high temperature in the cup, which differs from the ambient temperature in my study by about 60°C. When I come back after an hour I shall discover, probably to my regret, that my tea has cooled off and that the heat from the tea has been transferred from the cup to the room, resulting in a – barely noticeable – increase in the room’s temperature. We refer to this state as a state of increased disorder, in that the concentrated heat has now been distributed more or less evenly across the room. This process of increasing disorder is also known as entropy. The Russian-born, Belgian physical chemist Prigogine showed that complex systems that receive and transmit a flow of energy and matter from and to their environment eventually show a process that runs counter to the second law. In these systems order increases rather than decreases. Another way to put this is to say that the system selforganizes into a state of increased order. There are several concepts that need further explanation here. Let me


79

begin with the term ‘‘self-organization.’’ Self-organization means that the order comes from the system itself, not from the environment (in the latter case it would be organization, not self-organization). The second important concept, order, implies a degree of patterning or similarity. The predator–prey model nicely illustrates the principles implicit in both self-organization and order (the behavioral example does that too, of course, since it is based on the same dynamics; the explanation is probably a bit easier to follow if I use the concrete example of populations of animals as from the predator–prey model). Let us first assume that the foxes and the rabbits are not coupled in a predator–prey relationship, the rabbit–fox system (which would just be a collection of rabbits and foxes in some bounded region). In that case, the system would be characterized by a proportional inflow of energy, which is the food consumed by the ever-increasing population of rabbits (the net result of rabbit births minus rabbit deaths). There would also be a constant outflow of energy, which consists of dead foxes (in this simplified world, the foxes would have nothing to eat – they are not coupled to the rabbits – and hence only the population decrease factor would apply). Formally speaking, the ratio between any value of r (the rabbit population) and any preceding value of r is a constant (namely 1 + r; the same principle holds for the fox population f ). Thus the flow exhibits a simple order, expressed by the constants 1 + r and 1 + f. However, once we couple the rabbits and foxes in the form of a predator–prey dynamics an entirely different pattern results which consists of a regular and coherent repetition of rises and falls in the populations. This pattern shows a considerably higher level of complexity than the simple constant from the original system–environment flow. In order to visualize this pattern I shall use a simple trick. Recall that the ratio between any level of f (or r for that matter) and any preceding level equals a constant, namely 1 + f. I can plot the value of that ratio (in fact the first derivative of the growth function) in a two-dimensional space. One dimension is the value of the ratio at time t, the other dimension is the value of that ratio at some later time, t + n (n could be 10, for instance). Since the ratio is a constant in the case of the flow, the values at time t and t + n are identical. They specify a point in the two-dimensional space with coordinates (1 + f, 1 + f ). I now compute the ratio between any fox population level and any preceding level in the predator–prey case. When I now plot the ratio at time t against that of time t + n, the resulting image takes the form of a rounded triangular shape (figure 4.6). The simple point – the simplest possible form of order in the two-dimensional plane – has now been replaced by a regular but

80


Figure 4.6. The ratio of the present population level (time t) and the preceding level (time t − 10) for 1000 consecutive years.

considerably more complicated form of the rounded triangle. This new form must be the product of the predator–prey relationship and of nothing else but the predator–prey relationship. It cannot come from the environment, since the interaction with the environment has the geometric form of a simple point, which refers to a considerably lower order of organization. Likewise, it has also not been imported or smuggled into the form of the dynamics, since the only things we have added to the flow part of the equations are the components that specify the coupling (which is BP and BP in the behavioral example and rf and rf in the rabbit–fox predator–prey system; see equation 1). Consequently, the higher order is the sole product of the interaction dynamics. In that sense, we say that the system shows self-organization, meaning that it produces a higher level of order (organization) by itself. Since the form of the interaction, represented by the rounded triangular shape, is in no way prefigured or pre-represented in the form or parameters of the dynamics, we can also say this form is a property that emerges from the dynamics. Assume that, instead of running into the oscillatory pattern, the predator–prey dynamics produced a completely random pattern of population sizes. This means that the ratio between any current population level and a preceding level would be random as well (again, the example directly generalizes to the unwanted behavior-aversive stimulus case). Plotting those ratios in our t and t + n space would finally


81

result in covering all points in the t and t + n space. This result (if it applied to the predator–prey dynamics, but recall that it does not) would actually amount to a complete loss of order. Instead of the highly constricted order of the flow (it is specified by one and only one point in the space), we would now have a completely unlimited range of possibilities. This case would be similar to maximal entropy on the thermodynamics level. However, the order produced by the predator–prey dynamics is of a completely different kind. In contrast with the completely random case, it is highly specific: only a very limited and highly specific set of points are allowed (namely the points that form the rounded triangular shape). On the other hand, in contrast with the single-point case, it contains much more information, that is, a higher degree of order. In summary, we can define a dynamic system as a semi-isolated system, characterized by a specific internal structure and a specific flow from the environment. If it is a non-linear dynamic system, it will increase the amount of order in comparison with the amount of order present in the environment flow. In principle, it will increase the amount of order up to a level that is characteristic of the system (which, in the case of the predator–prey or of the unwanted-behavior– punishment relationship is the oscillatory pattern of the populations and of the frequency–punishment relationship, respectively, specified by the rounded triangular and egg-shaped forms). Dynamic systems and time An important, though almost tautological, aspect of dynamic systems is the aspect of time. The components or variables out of which the system is composed affect one another across time. The patterns that emerge are temporal patterns, characteristic changes in the values of the variables involved or characteristic replacements and transformations of the system’s physical elements. The time-governed aspect of dynamic systems is what distinguishes them from the structures of relationships between variables described in terms of regression models or comparable models of association. An association model describes the way a variable is linked with another variable, for instance how level of schooling is associated with a person’s income, to mention a standard example from introductory courses in statistics. Association does not imply that one variable causes the other to change (or the other way round), as would be implied in a dynamic system. Most often, association measures such as regression describe the coherence of variables across a sample at a specific moment in time. For instance, in cases of relationships between unwanted behavior, such as talking in the

82


classroom, and the punishment that follows such behavior, the association could easily be summarized in the form of a simple linear relationship, a correlation measure. However, because of the oscillatory pattern characteristic of this particular dynamics, the Pearson product moment correlation between the values of the coupled variables is about zero. This finding from the regression model implies that – in the model – unwanted behavior and punishment are not related at all, whereas we know that there exists a strong, deterministic, and mutual relationship between the two. Dynamic systems describe changes in variables across time for a specific case (e.g. a single individual). The dynamic relationship does not always show in the form of a linear association between the variables (which is not to say, of course, that all cases where the correlation coefficient is small or almost zero hide some complicated dynamics; the relationship between linear and dynamic associations is clearly not symmetrical). Dynamic systems and attractors Although the properties described so far constitute the basic features of dynamic systems – the semi-closedness, the geometric nature of the representation, the creation of order – there is of course far more to say about them. There is a host of concepts and notions that are regularly used in the context of dynamic systems modeling and application. An important concept concerns the notion of attractor. An attractor can be defined as the region of state space that captures the long-term behavior of the system. By state space I mean the geometric space formed by the components or variables that constitute the system. In the unwantedbehavior–punishment case, the state space consisted of the variable unwanted-behavior-frequency and the variable gravity-of-the-punishment, respectively. We have seen that any combination of a behavior frequency and a specific punishment level defines one point in the behavioral state space. The dynamics explains how, for any given starting level of unwanted behavior and of punishment and any given set of parameter values, the variables will evolve either towards some fixed set of values or towards a repeated pattern of changes. The change in the variables is described by the geometric pattern from figures 4.5 and 4.6 (note that similar patterns apply to the fate of predator–prey populations in the biological example). This pattern occupies a specified region of the state space, hence the definition of attractor. We distinguish between four types of attractors. The first is a single point in the state space and corresponds with a so-called steady state (for instance, a fixed level of unwanted behavior and a fixed level of punishment for which the system is in equilibrium). The second is a closed


83

loop or periodic cycle, as in the current behavioral or predator–prey example. The third is a torus, which is in fact a combination of cycles. The fourth is a relatively new discovery, dating back to the 1960s, and is called a strange attractor. A strange attractor is like a bounded region of the state space (for instance the space circumscribed by a cycle in the state space). In that region, the changes in the system are chaotic. Chaos is another highly appealing concept from dynamic systems theory. In classical mechanics, randomness and unpredictability of processes were considered apparent and were attributed to the fact that the conditions under which the processes occurred, including their causes, were not sufficiently known. We now know that there exist very simple and completely deterministic systems that nevertheless display behavior that is – seemingly – completely random and unpredictable. A good example of such a system is the logistic map. The logistic map is the discrete form of the logistic growth equation (by discrete is meant that the time steps occur in discrete units and not continuously). The equation of the logistic map with an equilibrium level equal to 1 is Lt+1 = Lt + Lt (1 − Lt)

equation 2

If the parameter is between 2.54 and 3 the consecutive values of Lt+1 show a chaotic oscillation. An important feature of deterministic chaos is its extreme sensitivity to initial conditions. Any infinitesimally small difference between the starting values of an initial Lt and another initial Lt will after a while lead to an arbitrarily large deviation between their chaotic series. The practical importance of chaos is that it provides a potentially internal source of arbitrarily large or small variation to dynamic processes. In the standard view, variation or more precisely fluctuation over time is viewed as the result of error terms, that is, terms resulting from causal processes that are not being controlled for. It follows then that the more variables are being controlled by a process operator (e.g. an experimental researcher), the smaller the contribution of random factors to the process (because the number of random factors is reduced by the extensive control). The existence of deterministic chaos, however, opens the possibility for randomization based on the process itself. Although the actual fluctuations in the process will be the result of both the intrinsic fluctuations based on the internal chaotic process and the extrinsic fluctuations based on uncontrolled external factors, reduction of the latter will not considerably reduce the degree of randomness produced. The reason is that chaotic processes can fill up any arbitrarily large region of the state space. As a result, externally added randomness will not change the nature of the process fluctuations.

84


Dynamic systems and bifurcations The concept of bifurcation refers to the occurrence of sudden shifts in the behavior of a system as a result of a gradual upgrading of the value of one of the process parameters. In a linear process we expect that the increase in the value of a parameter leads to an increase in a dependent variable. In principle, it does not matter whether the increase is from some very low value to a somewhat higher value or from some high value to an even higher value. The effects will be quantitatively different, but qualitatively similar (an increase, say). In non-linear systems, that is, systems that are capable of producing or increasing order, quantitative changes in parameter values sometimes lead to sudden changes in the quality of the system’s output. Some systems – including predator–prey systems – home in onto a point attractor for any value of a parameter under a critical value pc. If the parameter crosses the critical value, the system no longer settles into a point attractor but moves towards a cyclical attractor. This transition is known as a Hopf bifurcation. Another, very simple example of bifurcation is provided by the logistic map. The growth rate parameter, in the equation, leads to a point attractor for values smaller than 2; for values between 2 and 2.54 it leads to cyclical attractors with a doubling period (from 2 points to 4, from 4 to 8, etc.); above 2.54 the attractor becomes chaotic. Bifurcations are interesting phenomena because they demonstrate that discontinuity in the system’s output (the sudden change in the qualitative behavior) can be based on continuity in the system’s input (for example, a flow that causes the parameter at issue to increase in a continuous way). Or, to put it differently, discontinuity can be the intrinsic product of the system dynamics and does not necessarily depend on a discontinuous external input (Van der Maas and Molenaar, 1992). Using dynamic systems thinking in understanding psychological phenomena After giving this overview of general properties and principles of dynamic systems, we may now ask ourselves how all this could eventually be put to use in psychological theory building and research. Our aim is to understand psychological phenomena, not the fate of rabbits and foxes or mackerel and whiting. One of the major advantages of dynamic systems models is their extreme generality. They are almost literally stripped of anything that reminds us of the complexity and specificity of real natural phenomena. In the preceding sections, we have seen that the dynamics that applied to fish, or to rabbits and foxes, also applied to unwanted behaviors, such as talking in the classroom


85

and the resulting punishment. Dependent on the nature of the phenomena under scrutiny or the models available for such phenomena, we can apply either a quantitative or a qualitative approach to dynamic systems modeling. A quantitative approach We accepted that the equations of the predator–prey model apply to biological populations because the components of the equations were seen as highly generalized descriptions of biological phenomena such as ‘‘animals die, hence the population decreases’’ and ‘‘animals feed on prey, hence the population increases relative to the available prey.’’ The equations themselves, however, referred to additions and subtractions, to multiplication and comparable operations. We have seen that the logic of predator–prey interactions generalizes to interactions between behaviors or actions in different persons. The meaning of this – probable – similarity is not that the relationship between correctional and antisocial activities is a form of predator–prey interaction (although it is not wrong to state the relationship in this way). It is probably more correct to say that there exists a general asymmetrical model, describable by equation 1, that describes a wide variety of phenomena, including biological predator–prey relationships and relationships between correctional and antisocial or unwanted behaviors. According to Thatcher (1994) the model also specifies the changes in post-natal intra-cortical connections, with different regions of the brain corresponding to the ‘‘predator’’ and the ‘‘prey’’ parts. Comparable models can be built for many other phenomena in psychological change, growth, and development. For instance, it can be argued that the so-called elementary or primitive emotions are not like pre-wired emotional schemes but rather self-organize rapidly as a result of simple competitive and supportive relationships between constituents of various kinds. Examples of such constituents are motor patterns of facial muscles, internal hormonal secretions, perceptual patterns, and so forth (see De Weerth and Van Geert, 2000; Camras, Lambrecht, and Michel, 1996). At this point in the discussion, we should mention two kinds of difficulties associated with this quantitative approach to dynamic model building in psychology. The first difficulty concerns the possibility of representing the phenomena at issue by a simple mathematical model. It is often assumed that psychological phenomena are far more complex than biological interactions between foxes and rabbits, for instance, and that, consequently, simple models that may be applicable to such biological interactions are not suited for the complex

86


psychological realm. There is no reason, however, to believe that biological phenomena should be simpler than psychological phenomena. The point of building dynamic systems is to reduce the complexity inherent in the phenomena under scrutiny and to distill the simplest possible dynamic relationship that nevertheless specifies the essential aspect of the ongoing process. Such reduction is difficult to accomplish, but it is no more difficult with psychological than with biological processes. However, if quantitative models are just too far out of reach, it is always possible to confine oneself to applying the general principles of dynamic systems thinking in a more qualitative way. I shall discuss the qualitative approach in the next section. The second difficulty concerns the relationship between the mathematical model and the empirical data it attempts to model. If one models population sizes, the variable and its empirical correlate have a transparent relationship. A population is a countable set and 200 rabbits is twice as much as 100 rabbits, for instance. Things become different with psychological variables where the numerical representation of an underlying variable – for example the ‘‘level’’ of antisocial behavior or the coherence between specific facial motor patterns – amounts to a highly idealized representation that bears only a relatively vague relationship to its observable counterpart. If that is indeed the case, there remains only an indirect way of fitting the model to the data. For instance, if the model specifies a regression or a cycle, we can check if the observable data, whatever their exact nature, show qualitative signs of a regression or cycle respectively. A comparable problem with quantitative models is that they can easily be viewed as mere metaphors, because there is no direct mapping between the components of the model and certain aspects of the modeled reality. For instance, although a growth model applies transparently to the growth of a population or the growth of crops, it allegedly applies only metaphorically to the growth of syntactic understanding or moral reasoning, to name just two examples (Van Geert, 1991). The point is that a problem like this – metaphor or model – cannot be answered in general. By their very nature, dynamic models are very abstract and general specifications of relationships. The question of whether or not they adequately apply to a specific empirical phenomenon should be answered by taking all relevant aspects into account and not by drawing a line between a domain where such models apply and where they do not. A qualitative approach As said before, it often makes no sense to try to represent interesting aspects of reality by one-dimensional numerical variables that form the


87

centerpieces of dynamic mathematical models. The reason could be that such parts of reality are simply too complex or involve relationships that are too numerous. Another reason could be that we have no idea whatsoever as to what kind of dynamic relationships hold between the aspects or constituents of interest. If there are theoretical and empirical reasons to believe that the phenomenon under study can fruitfully be conceptualized as a dynamic system, it becomes interesting to try to check whether that phenomenon displays the properties that are often identified with non-linear dynamic systems in general. Such properties are, among other things, self-organization in the first place, but also the existence of attractors, threshold phenomena, bifurcations, chaos, discontinuities, self-similarity, and so forth (see for instance Fogel, Nwokah, Dedo, and Messinger, 1992; Fogel, 1993; Lewis, 1995, 1996; Lewis and Granic, 1999b). They are properties that make particular sense in the context of processes of change and development (they are less interesting from a purely differential psychological point of view, for instance). The concept of self-organization provides an interesting illustration. In principle, processes of whatever kind obey the general entropic principle. That is, the transmission of information or the creation of order always leads to a decrease of order or to a loss of information. Many models in psychology are implicitly or explicitly built upon this – generally correct – assumption. Examples are the general theory of information, but also theories like the Chomskyan view of language development. The basic argument of the latter runs as follows. The grammatical structure of language is – obviously – something that a child acquires by being confronted with a specific language, the mother tongue. That is, the structure of language comes from the ‘‘outside’’ and is transferred to the ‘‘inside.’’ It can be shown, however, that the observable language input is technically insufficient to get the structure of grammar across the outside/inside border, so to speak. That is, language input under-determines language structure. Thus, if grammatical structure cannot be transferred from the outside to the inside, where does it come from? There is, according to the Chomskyan view, only one logical possibility left, namely that it is already present inside. That is, it must be an innate property. The input is needed to set certain general innate parameters to the values adopted by the child’s mother tongue. From a dynamic systems point of view, the problem lies with the phrase ‘‘only one logical possibility left.’’ It is true only under the entropic assumption, which states that a process of information transmission can at most retain the order or information given and can, in principle, by no means increase the order or information contained in the signal (the input, say). However, non-linear dynamic systems

88


are characterized by the fact that they are order- and informationincreasing systems. Hence, the statement that there is only one logical possibility left must be rejected. More precisely, there are two logical possibilities left, namely, first, that grammatical structure is already internally present and, second, that grammatical structure arises by self-organization, given the properties of the learner and the properties of the ambient language. Both are mere statements-of-possibility. At present, we have no idea of how a structure such as grammar could become posted inside the brain preceding any actual confrontation with language. Similarly, we have no idea of how grammatical structure can emerge by self-organization, given the properties of the learner and the ambient language. From the viewpoint of conceptual likelihood, however, the self-organizational option constitutes the safest bet. The reason is that considerably more complicated forms of selforganization than the genesis of a grammar – for instance morphogenesis – are known to occur in living systems. Another reason is that the innateness assumption transfers the problem of the genesis of language structure from the ontogenetic to the phylogenetic domain. It remains to be explained how the structure of grammar came to be established in the human genome by some process that, by definition, must have been a process of self-organization, since there was no grammar around before it got materialized in the human brain along the path of human evolution. In summary, the grammar example shows how the application of a general property of dynamic systems – the likelihood of selforganization – throws a new and interesting light on old problems. It goes without saying that dynamic systems thinking has considerably more to offer than that. The present book provides a number of examples of how this particular approach may be fruitfully applied to a host of problems concerning the self, emotions, and human relationships.

COM M E N T A R Y

Fish, foxes, identity, and emotion Linda A. Camras and George F. Michel

Paul Van Geert’s chapter provides a lucid explanation of the dynamic systems approach that will be invaluable in orienting the reader to the other contributions in this volume. Dynamic systems theory was developed to explain the organization and maintenance of physical and biological systems that appear to defy the second law of thermodynamics (i.e. the law of entropy). The editors of this volume hope to demonstrate that dynamic systems theory can provide a framework for understanding one’s ‘‘identity’’ or ‘‘self’’ as a stable system that may yet change over time in a lawful manner. They propose that this system emerges from the interaction of component elements, including cognition, emotion, and contextual experience. If the dynamic systems approach is to be a guiding framework for studying the development of self and identity, then it should offer novel solutions to old problems or redefine what constitutes a problem. Only the experts who study these phenomena can determine whether it does so. We would like to highlight certain issues raised by Van Geert’s description of the characteristics of dynamic systems that should be kept in mind during such evaluation. Van Geert begins by emphasizing that dynamic systems involve a set of component elements. This seemingly obvious fact is actually quite important because several features of dynamic systems might also seem applicable to phenomena conceived as unanalyzable wholes. Thus, one who views identity as a unitary phenomenon might observe that it develops in a discontinuous fashion and assumes a limited range of values within a particular time period. For example, categories of identity status (e.g. Marcia, 1980) might appear to be sets of ‘‘attractor states’’ through which one moves as the result of a ‘‘bifurcation’’ process. Van Geert believes that making such analogies has value and can lead us to a fruitful reconceptualization of a phenomenon. In fact, such analogies are the cornerstone of the qualitative approach to using dynamic systems theory described at the end of his chapter. However, by preceding this discussion with a clear explanation that the characteristic features of dynamic systems are the result of lawful interactions

90


among component elements, Van Geert implicitly reminds us that eventually the set of component elements in any dynamic system must be accurately determined. Defining identity clearly (as urged by the editors of this volume) and determining whether it is an emergent phenomenon whose components are emotion, cognition, and experience or whether identity is itself a component of some broader system of individual functioning that includes also cognition, emotion, and experience is an important challenge. If we view the identity as emerging from the pattern of interaction among emotion, cognition, and experience, then this emergent phenomenon cannot itself affect the states of its constituent components. Rather it is the interactions among components (i.e. emotion, cognition, and experience) that enhance and constrain specific aspects of each other and, in doing so, give rise to the emergent phenomenon (i.e. ‘‘identity’’). At the same time, most appraisal theories of emotion would consider one’s identity to be an important influence on one’s emotional response to a stimulus event. One possible solution to this apparent theoretical incompatibility might be to make a conceptual distinction between identity and self-concept. That is, selfconcept (i.e. one’s cognitive self-representation) might be considered one component of a dynamical system of identity that also includes emotion, experience, and other aspects of cognition as additional components. One feature of dynamic systems that Van Geert strongly emphasizes is self-organization. As illustrated by the differential equations describing predator–prey interactions, the non-linear relationship among elements is produced by forces within the system, i.e. the coupled interaction of components. Thus a dynamic system is not just any system that changes over time. Rather it is an internally coherent system whose changes are primarily determined by internal (rather than external) forces. This feature might make the dynamic systems approach particularly appropriate to the study of identity since identity is indeed considered by most psychologists to be an internally coherent phenomenon. However, if identity-dependent behavior varies with the situational context, then selected aspects of environment and experience must now be incorporated into the system. This may require extending the concept of identity beyond its usual boundaries. As a result, the complex phenomenon of identity-functioning might be successfully modeled as a self-organized system requiring no external guidance plan controlling identity-related behavior. Currently, psychology is not experiencing a headlong rush toward the adoption of dynamic systems approaches. One reason may be that many researchers are simply unacquainted with the concept of dynamic systems. Thus, the current volume can play a seminal role in

Commentary on Van Geert

91

encouraging more widespread adoption of this perspective. Van Geert explicitly addresses two other reasons why dynamic systems perspectives have not made much headway within the psychology community. First, many researchers feel that certain psychological phenomena are too complex to be adequately represented by a simple mathematical model. Van Geert rightly asserts that biological phenomena are as complex as psychological phenomena. Nevertheless, with appropriate simplifying assumptions, complex biological (and physical) phenomena can be productively modeled as dynamic systems. This is the ‘‘art’’ of model building. Second, and less easy to dismiss, is the issue of measurement. As Van Geert points out, many psychological phenomena (e.g. self-esteem) are measured ‘‘very indirectly’’ with only a vague relation to an observable counterpart. Therefore, it may be difficult to assign numerical values to states of emotion, cognition, and contextual experience that are robust enough to permit the empirical testing needed for the development and refinement of dynamic models of self and identity. For example, there is no universal consensus regarding the measurement of emotion. Although systems for identifying and quantifying the production of facial expression are well developed, the relationship between facial expression and emotion itself is subject to debate. Until such thorny problems of numerical representation can be resolved, Van Geert suggests taking a qualitative approach rather than seeking to establish precise parameters for one’s model. However, this solution may be unsatisfactory to those for whom the very appeal of the dynamic systems approach is its potential for mathematical representation. Possibly more common than the dissatisfaction engendered by under-mathematizing is the dissatisfaction engendered by overmathematizing. Many psychological theories are primarily verbal narratives rather than formal mathematical models. In terms of Aristotle’s doctrine of causation, such narratives traditionally have presented efficient (mechanistic) causes or final (teleological) causes for a phenomenon rather than a formal cause (Bambrough, 1963). Formal equations may describe a phenomenon but, for many psychologists, they will not be able to explain it. Using Van Geert’s classroom-talking example, some constant value (alpha) may be found to accurately represent the magnitude of increase in classroom talking, but most psychologists will be more concerned with confirming or disconfirming the proposal that social reinforcement is the (mechanistic) cause of increased talking. Thus it may be important to emphasize that dynamic systems modeling does not exclude the possibility of explaining a phenomenon on a ‘‘different level of analysis.’’ Within the field of biology, researchers may describe phenomena such as morphogenesis as a self-organizing

92


dynamic system but they further seek to understand the mechanistic forces that produce the mutual influences among the elements of the system. Perhaps for psychologists, one important virtue of the dynamic systems approach will be the guidance that such formal (or informal) modeling can provide in our search for non-obvious mechanistic or developmental causes. In our own work (Camras, 1992, 2000; Camras, Lambrecht, and Michel, 1996; Michel, Camras, and Sullivan, 1992), we are attempting to use the dynamic systems perspective to achieve a greater understanding of the relationship between facial expression and emotion. We came to this perspective out of a sense of dissatisfaction with the approaches that have dominated the field in recent years. The story of our work bears a resemblance to Van Geert’s example of the potential value that the dynamic systems perspective provides for the study of language development. Many emotion theorists have viewed their target phenomenon as the manifestation of an internal controller (i.e. a central emotion ‘‘program,’’ analogous to Chomsky’s genetically provided Language Acquisition Device). Such theories predict a strong linear coherence among the components of emotion (e.g. feelings, facial expressions, neurophysiological substrates). In contrast, we were struck by the fact that stereotypic emotional facial expressions do not always seem to occur during episodes of emotion and, in fact, they may sometimes occur when the predicted corresponding emotion is unlikely to take place. For example, focusing narrowly on the phenomenon of infant facial expression, we have empirically demonstrated that non-emotion factors (i.e. head and gaze movements) may systematically recruit specific facial movements resulting in the production of an ‘‘emotional’’ expression. Yet such ‘‘inappropriate’’ responses do not produce confusion in the emotion communication process and, in fact, go largely unnoticed during social interactions. We have sought to account for this by reconceptualizing emotion as a ‘‘softly assembled’’ system in which relationships among components are orderly but non-linear. Currently, we are attempting to delineate more specifically those circumstances under which the various components of emotion are manifested. This step must precede any attempt to produce a precise mathematical model of emotion responding. Because emotion is indeed a complex psychobiological phenomenon, we are far from achieving our ultimate goals. Nonetheless, we feel that adopting a dynamic systems perspective has offered us a means of escape from earlier paradigms that do not adequately capture the real phenomenon of emotion.

COM M E N T A R Y

Fish, foxes, identity, and emotion Linda A. Camras and George F. Michel

Paul Van Geert’s chapter provides a lucid explanation of the dynamic systems approach that will be invaluable in orienting the reader to the other contributions in this volume. Dynamic systems theory was developed to explain the organization and maintenance of physical and biological systems that appear to defy the second law of thermodynamics (i.e. the law of entropy). The editors of this volume hope to demonstrate that dynamic systems theory can provide a framework for understanding one’s ‘‘identity’’ or ‘‘self’’ as a stable system that may yet change over time in a lawful manner. They propose that this system emerges from the interaction of component elements, including cognition, emotion, and contextual experience. If the dynamic systems approach is to be a guiding framework for studying the development of self and identity, then it should offer novel solutions to old problems or redefine what constitutes a problem. Only the experts who study these phenomena can determine whether it does so. We would like to highlight certain issues raised by Van Geert’s description of the characteristics of dynamic systems that should be kept in mind during such evaluation. Van Geert begins by emphasizing that dynamic systems involve a set of component elements. This seemingly obvious fact is actually quite important because several features of dynamic systems might also seem applicable to phenomena conceived as unanalyzable wholes. Thus, one who views identity as a unitary phenomenon might observe that it develops in a discontinuous fashion and assumes a limited range of values within a particular time period. For example, categories of identity status (e.g. Marcia, 1980) might appear to be sets of ‘‘attractor states’’ through which one moves as the result of a ‘‘bifurcation’’ process. Van Geert believes that making such analogies has value and can lead us to a fruitful reconceptualization of a phenomenon. In fact, such analogies are the cornerstone of the qualitative approach to using dynamic systems theory described at the end of his chapter. However, by preceding this discussion with a clear explanation that the characteristic features of dynamic systems are the result of lawful interactions

90


among component elements, Van Geert implicitly reminds us that eventually the set of component elements in any dynamic system must be accurately determined. Defining identity clearly (as urged by the editors of this volume) and determining whether it is an emergent phenomenon whose components are emotion, cognition, and experience or whether identity is itself a component of some broader system of individual functioning that includes also cognition, emotion, and experience is an important challenge. If we view the identity as emerging from the pattern of interaction among emotion, cognition, and experience, then this emergent phenomenon cannot itself affect the states of its constituent components. Rather it is the interactions among components (i.e. emotion, cognition, and experience) that enhance and constrain specific aspects of each other and, in doing so, give rise to the emergent phenomenon (i.e. ‘‘identity’’). At the same time, most appraisal theories of emotion would consider one’s identity to be an important influence on one’s emotional response to a stimulus event. One possible solution to this apparent theoretical incompatibility might be to make a conceptual distinction between identity and self-concept. That is, selfconcept (i.e. one’s cognitive self-representation) might be considered one component of a dynamical system of identity that also includes emotion, experience, and other aspects of cognition as additional components. One feature of dynamic systems that Van Geert strongly emphasizes is self-organization. As illustrated by the differential equations describing predator–prey interactions, the non-linear relationship among elements is produced by forces within the system, i.e. the coupled interaction of components. Thus a dynamic system is not just any system that changes over time. Rather it is an internally coherent system whose changes are primarily determined by internal (rather than external) forces. This feature might make the dynamic systems approach particularly appropriate to the study of identity since identity is indeed considered by most psychologists to be an internally coherent phenomenon. However, if identity-dependent behavior varies with the situational context, then selected aspects of environment and experience must now be incorporated into the system. This may require extending the concept of identity beyond its usual boundaries. As a result, the complex phenomenon of identity-functioning might be successfully modeled as a self-organized system requiring no external guidance plan controlling identity-related behavior. Currently, psychology is not experiencing a headlong rush toward the adoption of dynamic systems approaches. One reason may be that many researchers are simply unacquainted with the concept of dynamic systems. Thus, the current volume can play a seminal role in

Commentary on Van Geert

91

encouraging more widespread adoption of this perspective. Van Geert explicitly addresses two other reasons why dynamic systems perspectives have not made much headway within the psychology community. First, many researchers feel that certain psychological phenomena are too complex to be adequately represented by a simple mathematical model. Van Geert rightly asserts that biological phenomena are as complex as psychological phenomena. Nevertheless, with appropriate simplifying assumptions, complex biological (and physical) phenomena can be productively modeled as dynamic systems. This is the ‘‘art’’ of model building. Second, and less easy to dismiss, is the issue of measurement. As Van Geert points out, many psychological phenomena (e.g. self-esteem) are measured ‘‘very indirectly’’ with only a vague relation to an observable counterpart. Therefore, it may be difficult to assign numerical values to states of emotion, cognition, and contextual experience that are robust enough to permit the empirical testing needed for the development and refinement of dynamic models of self and identity. For example, there is no universal consensus regarding the measurement of emotion. Although systems for identifying and quantifying the production of facial expression are well developed, the relationship between facial expression and emotion itself is subject to debate. Until such thorny problems of numerical representation can be resolved, Van Geert suggests taking a qualitative approach rather than seeking to establish precise parameters for one’s model. However, this solution may be unsatisfactory to those for whom the very appeal of the dynamic systems approach is its potential for mathematical representation. Possibly more common than the dissatisfaction engendered by under-mathematizing is the dissatisfaction engendered by overmathematizing. Many psychological theories are primarily verbal narratives rather than formal mathematical models. In terms of Aristotle’s doctrine of causation, such narratives traditionally have presented efficient (mechanistic) causes or final (teleological) causes for a phenomenon rather than a formal cause (Bambrough, 1963). Formal equations may describe a phenomenon but, for many psychologists, they will not be able to explain it. Using Van Geert’s classroom-talking example, some constant value (alpha) may be found to accurately represent the magnitude of increase in classroom talking, but most psychologists will be more concerned with confirming or disconfirming the proposal that social reinforcement is the (mechanistic) cause of increased talking. Thus it may be important to emphasize that dynamic systems modeling does not exclude the possibility of explaining a phenomenon on a ‘‘different level of analysis.’’ Within the field of biology, researchers may describe phenomena such as morphogenesis as a self-organizing

92


dynamic system but they further seek to understand the mechanistic forces that produce the mutual influences among the elements of the system. Perhaps for psychologists, one important virtue of the dynamic systems approach will be the guidance that such formal (or informal) modeling can provide in our search for non-obvious mechanistic or developmental causes. In our own work (Camras, 1992, 2000; Camras, Lambrecht, and Michel, 1996; Michel, Camras, and Sullivan, 1992), we are attempting to use the dynamic systems perspective to achieve a greater understanding of the relationship between facial expression and emotion. We came to this perspective out of a sense of dissatisfaction with the approaches that have dominated the field in recent years. The story of our work bears a resemblance to Van Geert’s example of the potential value that the dynamic systems perspective provides for the study of language development. Many emotion theorists have viewed their target phenomenon as the manifestation of an internal controller (i.e. a central emotion ‘‘program,’’ analogous to Chomsky’s genetically provided Language Acquisition Device). Such theories predict a strong linear coherence among the components of emotion (e.g. feelings, facial expressions, neurophysiological substrates). In contrast, we were struck by the fact that stereotypic emotional facial expressions do not always seem to occur during episodes of emotion and, in fact, they may sometimes occur when the predicted corresponding emotion is unlikely to take place. For example, focusing narrowly on the phenomenon of infant facial expression, we have empirically demonstrated that non-emotion factors (i.e. head and gaze movements) may systematically recruit specific facial movements resulting in the production of an ‘‘emotional’’ expression. Yet such ‘‘inappropriate’’ responses do not produce confusion in the emotion communication process and, in fact, go largely unnoticed during social interactions. We have sought to account for this by reconceptualizing emotion as a ‘‘softly assembled’’ system in which relationships among components are orderly but non-linear. Currently, we are attempting to delineate more specifically those circumstances under which the various components of emotion are manifested. This step must precede any attempt to produce a precise mathematical model of emotion responding. Because emotion is indeed a complex psychobiological phenomenon, we are far from achieving our ultimate goals. Nonetheless, we feel that adopting a dynamic systems perspective has offered us a means of escape from earlier paradigms that do not adequately capture the real phenomenon of emotion.

CH A P T E R 5

A relational perspective on the development of self and emotion Alan Fogel1

Begin with two premises. First, psychological experience always implies a connection, a relationship: with another person, with cultural tools or language, or with the natural environment. Life is a network of relationships. Second, psychological experience is always dynamic and changing. The simplest visual perception requires a change, either in a movement of the object or a movement of the eyes, head, or body. Thoughts and feelings fluctuate in a continuous pattern of change. These patterns of change themselves change as people develop. Life is a series of changes. On the other hand, part of psychological experience is a sense of one’s uniqueness (the self) and a sense of one’s permanence through time (identity). How can this occur? How can people have a sense of themselves and their stability over time if psychological experience is fundamentally relational and dynamic? The answer proposed in this chapter is that people experience the changes in their relationships according to different types of emotion and that emotions provide information about the self. Consistent with the two premises, emotions are conceptualized as dynamic experiences of harm or benefit, perceived as personally meaningful with respect to the individual’s changing relationship with the environment (Barrett, 1993; Barrett and Campos, 1987; Fischer, Shaver, and Carnochan, 1990; Frijda, 1986; Lazarus, 1991b). According to Frijda, emotional experience ‘‘is glued, as it were, to its object, coinciding entirely with apprehending that object’s nature and significance . . . [negative] emotional experience is perception of horrible objects, insupportable people, oppressive events’’ (Frijda, 1986: 188). And De Rivera suggests that ‘‘emotions may be conceived as existing between people, as various sorts of attractions and repulsions . . . which transform their bodies and perceptions’’ (De Rivera, 1992: 200). As we shall see, this perspective suggests that there are an unlimited variety of emotional experiences corresponding to the limitless variety of relationship dynamics. From the perspective of this chapter, emotion is one way of 93

94

Alan Fogel

discovering the meaning of a relationship for the self and, hence, the unique position of the self in the relationship. Over long periods of time, the perception of consistency in one’s emotional experience begins to yield a sense of permanence of the self through time. In this sense, perception and emotion are two aspects of the same process of discovering the way in which the self is related to the world (Dewey, 1934; Stern, 1985). The relational perspective on self and emotion Before developing this perspective, a few words may be said about why it is different from traditional models of self and emotion. If one begins with a different premise, that each person is a unique individual from the outset, then the existence of a sense of self does not need to be explained. The self-contained individual, existing as a totality independent of the surround, is one of the main features of Western culture inherited from Greek philosophy (Levinas, 1969). In contrast to the view presented here, of emotion as the perceived meaning of relationships, emotions in the individual perspective are conceived as basic ‘‘internal’’ states or motivations, generated from the neurophysiology in response to an ‘‘external’’ cause. In this perspective, individuals ‘‘have’’ an emotion that requires expression, via action, with respect to the environment. The individual perspective predicts that there are a limited number of discretely different emotions that are genetically programmed into the neural system and become available to define harms and benefits for the individual. The relational perspective, by contrast, sees individuals as open and changing components in systems of relationships. Cooperative communication, for example, brings people together to create an outcome that no single person could achieve. Consider choral singing. Individual singers have different parts and vocal ranges that, when mixed, form a coherent aural aesthetic. The concept of coregulation describes this type of social process that is jointly created (Fogel, 1993). When coregulating, individuals’ behavior in the group is not theirs alone, but the manifestation of the group’s dynamics in each of their bodies. In the same way, emotion is experienced as harms and benefits within individual bodies but it reflects the body’s experience of a relational dynamic. According to Gergen (1991: 157), for example, If it is not individual ‘‘I’’s who create relationships, but relationships that create the sense of ‘‘I,’’ then ‘‘I’’ ceases to be the center of success or failure, the one who is evaluated well or poorly, and so on. Rather, ‘‘I’’ am just an I by virtue of playing a particular part in a relationship.

The relational perspective does not deny that people often perceive

A relational perspective on self and emotion

95

Table 5.1. Time scales and self experience. It takes only seconds to become aware of some aspects of the self and years to become aware of other aspects of the self. Topic

Event

Frame

Development

Time scale Sense of self Type of experience

Seconds Orientation Orientational emotions

Minutes, hours Authorship/agency Narrative emotions

Years Identity Reference emotions

their part in a communicative process as ‘‘their own’’ contribution or ‘‘their own’’ failure to make an effective contribution. On the contrary, these attributions are one of the undeniable manifestations of the self, the sense of one’s own uniqueness. Each singer in the chorus manifests the group dynamics in a unique way, experiencing his or her own part in the larger musical creation. The relational perspective described in this chapter does, however, seek a principled explanation of the sense of self. How, in fact, does the self arise in spite of the obvious fact of ecology: that everyone and everything is merely an incomplete location in a network of relationships comprising the cultural and physical world? There are different ways in which the sense of uniqueness can be experienced and they differ according to the time scale in which the individual is a participant in a relationship. At each time scale, a different aspect of self arises with respect to a different type of emotional process. In this chapter, three time scales are discussed. They are summarized in table 5.1 and introduced briefly here, followed by a more detailed description in the remainder of the chapter. The first time scale is microseconds and seconds. Even during these very brief periods of time, individuals can have a sense of their orientation with respect to others corresponding to emotions related to approach or avoidance. The second time scale is minutes, hours, or longer. This is the time it takes for orientations to form into a sequential pattern of communication. During this time scale, individuals can have a sense of their unique role in the authorship of the pattern. Authorship is a sense of one’s agency and occurs with regard to emotions such as security versus insecurity or togetherness versus loneliness. The third time scale occurs over years. This is the period of time it takes for individuals to have a sense of the uniquely enduring aspects of themselves, their identity. These experiences typically occur with respect to emotions such as harmony or conflict, satisfaction or dissatisfaction with oneself over time. At any given moment in time during the life course, the individual may experience the self and its related emotions

96

Alan Fogel

at all three time scales. These ideas are developed in the remainder of this chapter. Orientation This section discusses the emergence of the most basic form of self, the sense of a unique orientation with respect to a relationship. In order to coregulate, people orient their actions toward the other within a period of microseconds and seconds. Orientation is not a fully formed action system with a predefined goal but orientation may become more goallike during the process of coregulation. Orientation is similar to the concept of intention defined as ‘‘behavioral object directedness’’ (Merleau-Ponty, 1962; Vedeler, 1991). In this case, an object can be a physical object or a person. Intention implies an orientation movement of the body relative to someone or something rather than toward some fully formed goal. Because there is always an orientation, it means that there is always an intention. Trevarthen’s (1993: 123) concept of ‘‘motive . . . a readiness for perceiving information needed for acting’’ is similar. Orientation is also similar to the concept of ‘‘dialogical position’’ of actual or imagined selves (Hermans and Kempen, 1993), one’s stance vis-a`-vis another. Another related concept is ‘‘action readiness,’’ a propensity to move toward or away accompanied by an emotional experience of approach or avoidance toward a particular relationship (Frijda, 1986). Frijda’s position is the only one that makes clear that orientations are connected to particular forms of emotional experience. Since there is always an orientation, there is always some kind of emotional experience. Orientations can be relatively open or relatively closed. Postural orientations, for example, may differ with respect to whether stance, gaze direction, and limb position is open or closed vis-a`-vis a partner. Mental orientations differ with respect to the willingness to embrace or reject the consideration of concepts or ideas. More open orientations allow a freedom in how the process unfolds and permit creativity within the system. More closed orientations reflect degrees of control over how the process unfolds. The dimension of open–closed highlights the emotional aspect of orientation. Open orientations correspond to emotional attraction or a feeling of wanting to approach the other. Closed orientations correspond to feelings of repulsion or withdrawal. The sense of uniqueness, the self, arises in part because orientations imply the location of the self with respect to the environment. Orientations may be directed from the individual toward the environment or they may be receptive, from the environment toward the individual (Vedeler, 1994). When orientations are more directed, individuals have


97

the experience of acting on the environment with varying degrees of control. When orientations are more receptive, one has the experience of being influenced by the environment or of ‘‘abandoning ourselves to the world’’ (Vedeler, 1994: 346). There is typically an alternation between being directed or receptive. Hardness and softness, roughness and smoothness, moonlight and sunlight, present themselves in our recollection not preeminently as sensory contexts but as certain kinds of symbioses, certain ways the outside world has of invading us and certain ways we have of meeting that invasion. (Merleau-Ponty, 1968: 194)

Thus, although orientations are inherently relational, implying a relationship between self and other, individuals can perceive their position vis-a`-vis others. Since the individual occupies a location in physical and psychological space, the body is perceived as fundamentally in an orientation because all movement is from one location and toward another, either directed away from the self or toward the self. All psychological processes, including thought and narration, have an orientation because of their fundamental embodiment (Hermans and Kempen, 1993; Lakoff and Johnson, 1999; Yasuo, 1987). The self’s uniqueness, then, is experienced from its particular perspective at one pole of a relational orientation (Gibson, 1966; 1979; Michaels and Carello, 1981; Reed, 1987). Recent discoveries suggest that this type of self-awareness can be directly perceived at an early age. Human infants, for example, can identify the movements of their own bodies with respect to the environment and other persons, the so-called ‘‘ecological self’’ that differs from the mirror recognition of self at 18 months (Butterworth, 1995; Fogel, 1995; Legerstee, Anderson, and Schaffer, 1998; Rochat, 1995; Rochat and Hespos, 1997; Stern, 1985). The dynamics of coregulation and the continual shifts of orientation also play a role in establishing a particular emotional quality to the sense of a unique self. Because to coregulate one must continually readjust one’s actions based on the continuously changing actions of the partner, individuals can experience their relative degree of creativity as a participant in a relationship (Fogel, 1993; Ganguly, 1976; Pickering, 1999; Whitehead, 1978). Because the self is inherently creative, the experience of uniqueness is not an experience of being but an experience of becoming, a process of improvisation during communication (Bakhtin, 1981; Barclay, 1994; Boesch, 1991; Bosma, 1995; Hermans and Kempen, 1993; Jansz, 1995; Josephs, 1998; Shotter, 1981). It is thought that the earliest awareness of self in human infancy, called the emergent self, is just this experience of being a unique participant in the creative process (Stern, 1985).

98

Alan Fogel

A similar perspective is offered by philosophies of intersubjectivity (Buber, 1958; Jopling, 1993; Levinas, 1969; De Quincey, 1998) in which self and other arise as a result of communication and dialogue. During the give and take of ordinary discourse, each person’s statements are countered by the other so that mutual reformulation (coregulation) of meaning is required. As speakers make clearer their point of view to listeners, that point of view becomes clearer to the speakers themselves (Hermans and Kempen, 1993). The sense of self that arises from these creative processes, however, is most likely to be experienced just following the creative act or creative moment. It is necessary for the experience to achieve a resolution. After having peak experiences of creative flow, as in mountain climbing or musical performance, adults report feeling a sense of uniqueness that did not occur during the experience itself (Csikszentmihalyi, 1990; Csikszentmihalyi and Rathunde, 1998). Dewey (1934) has pointed out that there is a crucial difference between experiencing (being in the flow of activity) and having an experience. The latter implies a sense of both participation and completeness from which a characteristic emotion may be felt as well as a sense of self. This leads us to consider the issue of the completion of orientations into events. Events At particular times during the process of communication, the creative flow becomes punctuated. There is a change from experiencing as flow to having an experience. An experience is a sense that the flow has coalesced, a noticeable pattern has formed, a pattern that will be called an event. The sense of one’s uniqueness comes into being as a concrete event that emerges from the orientational flow (Fogel and Branco, 1997; Kegan, 1982; Pedrosa, Carvalho, and Imperio-Hamburger, 1997; Sampson, 1989; Whitehead, 1978). Events are the psychological experiences of real or actual entities such as objects, actions, feelings, or thoughts. When we experience something as real, we are participating in a process by which orientations have coalesced into an awareness of events, a process that may take only microseconds (Whitehead, 1978). The emergence of events is the perception of an island of stability against the background of the dynamic flux of change. From a dynamic systems theory perspective, events are self-organizing processes that owe their stability to the mutual interactions among the elements in the relationship and not to the existence of stored, fully formed representations inside the individual (Fogel and Thelen, 1987; Thelen and Smith, 1994). As orientations coalesce into events, there is the simultaneous emergence


99

of a stable emotional experience of value vis-a`-vis the relationship between self and environment (Hermans and Kempen, 1993). All events are experienced as inherently emotional (Dewey, 1934). These orientational emotions yield the experience of values such as good or bad, approach or avoid, accept or reject, tasteful or distasteful, pleasure or pain. Value is the most basic form of awareness, occurring on a time scale of seconds and microseconds. Value is also the earliest form of awareness in human development. Newborn infants have sophisticated approach and withdrawal reflexes by which they can evaluate a wide range of sensory experiences. Newborns suck sweet liquids and spit out bitter ones; they turn toward voices with particular pitch and intensity levels and not others; they follow the movements of high-contrast visual objects and not others (Fogel, 2001). These experiences of value are centers of psychological consistency, called the emergent self (Stern, 1985), by which the self’s unique relationship to the environment can be established. To the extent that orientations are open, rather than closed, they are perceived with some sense of uncertainty. When engaged in an open conversation, for example, there are flow periods in which the outcomes are unknown and participants are willing to remain in a relatively uncertain cocreative process. Events, although they appear when uncertainty in orientations is reduced over time, always have some orientational aspect that leads them from present to future. Although it is theoretically useful to distinguish orientations from events, in practice there are neither pure orientations nor pure events: orientations coalesce into events but events-as-orientations set the stage for the next events (James, 1976; Merleau-Ponty, 1962; Whitehead, 1978). Events, therefore, can be more or less orientational, referring to the possibility for the event to become transformed in the process of coregulation. In science, for example, the ideal situation is that concepts are working models. The scientist is supposed to maintain an orientational stance toward the data using those concepts but the concepts are always open to modification. The values that are perceived as part of an event are rarely finalized. They are always open to a process of revaluation. Becoming and being: balancing orientation and event Orientations and events are two related aspects of awareness of self. The former is connected with the awareness of becoming and the latter with the awareness of being, the former with change and the latter with stability. Within individual experience, there are moments when being and becoming are present in a fruitful and self-sustaining balance. One remains open to change while at the same time one has a sense of

100

Alan Fogel

Table 5.2. The spectrum of self-awareness, from being to becoming Primarily being

Balance of being and becoming

Primarily becoming

Events are stable Orientations are minimal The self is isolated

Events are orientational Orientations flow into events The self is co-regulated

Events do not coalesce Orientations are salient The self is lost or merged

stability and uniqueness. In an open conversation, for example, the participants begin with a frame of mind, a loosely defined set of orientations that partake of their individuality. Via coregulation, each person opens to change but that change is integrated into their prior orientations, thus preserving their individuality and at the same time changing it. In this case, there is an intimate and directly perceived connection between self and other that has the special quality of being cocreative. One is aware of the self, one is aware that the other is a self, and one is aware that the emergence of those selves (as events) depends upon the cocreative process (as mutual orientations) and the self-events are always orientational. Cobeing arises from and flows into cobecoming. Buddhist thinking refers to the concept of dependent coarising, which refers to the awareness of the creative process by which cobecoming is balanced with cobeing. In this process, the individuals become aware that their attempts to ground themselves in the permanence of being are set against the background of the continuous living flow of becoming (Pickering, 1999; Varela, Thompson, and Rosch, 1991; Wilber, 1979). When individuals become fully open to each other during interpersonal encounters, the experience of orientation toward each other is both indeterministic and endless, the experience of ‘‘infinity’’ (Levinas, 1969). In these meetings, the other person becomes the self and viceversa, there is no sense of being directed or receptive, only a feeling of selfhood dissolving into union. One sees the same humanity in the other and meets the other without reserve and in the fullness of the other’s vulnerability to be changed. These have been called ‘‘I–Thou’’ as opposed to ‘‘I–It’’ relationships (Buber, 1958). This balance between being and becoming is only one part of the spectrum of self-awareness. One can shift the balance either more toward becoming or more toward being (see table 5.2). One pole of self-awareness is that of pure becoming, when experience is completely orientational. One is oriented toward another person or thing without an end or resolution. There is a merger with that thing, the experience ‘‘of eternity, the complete absorption in being’’ (Loewald, 1980: 141). Because the sense of self arises only as orienta-


101

tions coalesce into events, purely orientational awareness cannot involve a sense of self. In Frijda’s (1985) view, for example, emotional experience has this character. When you attack someone or something in rage, for example, there is a complete absorption in that thing or person (as when a person says, ‘‘I completely lost myself,’’ or ‘‘I didn’t know what I was doing’’). At the time of making such statements, one is reflecting on having had an experience of anger. During the experience, however, there is no sense of I or you. In relationships, partners fail to provide each other with closure and confirmation for their actions. When infant self-initiated acts are not facilitated by the family system to help the infant appreciate what acts lead to what consequences, for example, then the inner experience will not be felt as the infant’s own (Sander, 1962). The other pole of self-awareness is pure being, when events lose their orientational character. There is no change because no matter what the orientation is, the result is the same event. These are ‘‘experiences of fragmentation’’ such that ‘‘each instant loses its relation to any other instant and stands by itself’’ (Loewald, 1980: 143). One feels estrangement, depression, helplessness, depersonalization, and alienation when faced with unchanging circumstances, stubborn people, irreconcilable differences, chronic tensions or anxieties, or persistent negative emotions. In these cases, the self fills one’s awareness with a kind of heaviness. Thoughts and emotions are self-focused and attempts to orient away from the self typically flounder in a sense of self-negation or self-inflation. Being lost or losing all of one’s freedoms may lead to this experience of an absence of self. Periods of being or becoming may last only seconds or minutes, as when the resolution of a communication process takes longer than expected, or much longer periods as in chronic depression. The orientational experience of self, therefore, is not a static, stored, or structural thing. It is an appearance and disappearance, a coming and going. In conversation, for example, one goes between a selfless merger with the other, an awareness of a cocreated self experienced in the effort to dialogue, and a heightened focus on one’s own needs at the expense of the other. In athletics and other movement activities one alternates from concentration to flow, from an awareness of the self–other relationship and its implications for action and emotion, to selfless immersion in the moment, or to excessive pride or shame at winning or losing. Ideally, then, a sense of a unique self occurs when dynamically unfolding orientations coalesce into stable events, each having a particular emotional quality and a particular connection between self and other. In the same way that orientations self-organize dynamically into events during microseconds and seconds, observation suggests that events are

102

Alan Fogel

perceived to self-organize into sequential patterns of events over minutes, hours, or days. These sequential patterns of events in communication systems are called frames. The sequential coherence of events and its connection to the self is the topic of the next section. Authorship The psychological coherence of patterns of event sequences is captured by the concept of frame. Frames are sequences of events that have a coherent theme, that take place in a specific location, and that involve particular forms of mutual coorientation between participants. Frames have a temporal organization – a beginning, middle, and end – in which the events in the frame cohere (Bateson, 1955; Fogel, 1993; Goffman, 1974; Jones, 1990; Kendon, 1985). Examples of frames are greetings, topics of conversation, conflicts, or children’s social games. Frames are coherent patterns that result from self-organizing processes. Because events are partly orientational; the events in a sequence can change in order to establish a psychological relationship with each other. In this way, all the events in the frame are coconstructed with all the others, an example of self-organization. In the same way that events coalesce and become stable from the dynamics of orientations, frames coalesce as coherent patterns out of the dynamics of event sequences. Because of these psychological stabilities in the flow of action over time, individuals have the opportunity to perceive a stable sense of self in relation to the frame. Frames are similar to narratives, the latter referring primarily to coherent sequences of verbal communication. Narratives have a stable theme that emerges from the self-organization of the events; they have an orientational direction of flow over time that motivates the movements of the actors and events (Ginsburg, 1985; Hermans and Kempen, 1993; Haviland and Kahlbaugh, 1993; Jones, 1990; Ricoeur, 1983). Single events are created in a period of microseconds and seconds, enough time for one to experience an approach or withdrawal orientation. Narratives, on the other hand, are assembled over longer periods of time, over minutes or hours or even longer periods. At this time scale the individual can perceive new stable aspects of the self, beyond simple orientation and emotional value. As the individual becomes aware of the narrative/frame as a whole unit (e.g. a particular form of play or conflict in a dyad or group), the narrative is perceived as an event in itself. The same considerations that apply to the prior discussion of orientations and events can apply to narratives. The perception of the narrative as whole event can occur in microseconds or seconds, while at the same time the individual can also


103

experience the actual unfolding of the narrative over minutes, hours or days. In this way, the entire history and future of frames that extend for long periods of time can come into existence as an event in the present (Fogel, 1993; Merleau-Ponty, 1962). Some theorists (Stern, 1985; Tomkins, 1962) suggest that a new type of emotional experience, different from that of orientational emotions, arises over the time scale of narratives and that these narrative emotions partly account for the perceived coherence of narratives. In the creation of a narrative emotion, the specific values of orientational emotions from each event in the frame do not simply sum together. Rather, the emerging coherence of the events in the frame is partly accounted for by the emerging coherence of a narrative emotional process that is perceived to link events meaningfully together through time. Observational research suggests that each narrative emotion forms into smoothly unfolding temporal contours, patterns of feeling experience (Tomkins, 1962) that have been called ‘‘protonarrative envelopes’’ (Stern, 1985; 1993). These contours create ‘‘feeling qualities best captured by such kinetic terms as ‘crescendo,’ ‘decrescendo,’ ‘fading,’ ‘exploding,’ ‘bursting,’ ‘elongated,’ ‘fleeting,’ ‘pulsing,’ ‘wavering,’ ‘effortful,’ ‘easy,’ and so on’’ (Stern, 1993: 206). The temporal contour of a crescendo, for example, is a relatively slow build-up of emotion, reaching a peak and then declining rapidly. Each narrative emotion also has a meaning for the self, the personal harms or benefits of the relational connections in the frame (Frijda, 1986; Stern, 1993; Tomkins, 1962). These are experiences that connect events over time in some meaningful way for the self. Narrative emotions include, therefore, experiences that take some time to unfold and that establish a personal relational meaning such as love versus hate, respect versus disrespect, security versus insecurity, togetherness versus loneliness, safety versus threat, pride versus shame. Security can only be felt in reference to the consistency of interpersonal contact over the course of a frame. Security has a different temporal contour than other narrative emotions, a steadiness that accompanies feelings of enjoyment or comfort. Narrative emotions self-organize over time through a process that magnifies or minifies the psychological meaning of particular events by focusing or diffusing the experience of initial relational orientations over a sequence of events (Demos, 1982; Fogel et al., 1997; Lewis, 1995; Lewis and Ferrari, this volume; Sarbin, 1986a; Tomkins, 1962; 1978). Consider the following example. Walking down a lonely street at night one hears, or thinks one hears, a sound coming from behind. One’s orientation (attention and posture) toward the direction of movement becomes changed to having more urgency and one’s orientation toward what is

104

Alan Fogel

behind comes into existence for the first time as wariness. What happens next depends upon the specific way in which emotion magnifies or minifies this orientation. Perhaps one turns around and notices a cat, which minifies the potential fear and magnifies the orientation to continue at the same pace, lowering the potential to perceive threat. Or perhaps one turns to see a person. The subsequent process depends upon whether that person is a man or woman, whether he or she is following or merely walking in the same direction, whether there are people ahead, or street lights. The experience of participation in frames, therefore, inevitably highlights the self in a particular manner. Is there anything the person can do to protect him- or herself from harm? To calm him- or herself? To take charge of the situation? The time scale of narratives allows the self to be aware of a sense of its own agency. I will follow Day and Tappan’s (1996) use of Bakhtin’s (1981) concept of authorship to describe that sense of agency. Authorship refers specifically to the sense of self that can only be experienced as a coagent in the construction of a social narrative, a story about the self meant to be communicated to another person. Authorship, however, can vary from being to becoming, as shown in table 5.2. There are other similar terms for agency, such as self-efficacy and perceived control, but they fail to specify explicitly their links to the relational dynamics. The purpose of this chapter is to show how particular features of self-experience can be linked to the emerging stabilities present in communication processes at different time scales. Authorship can also be experienced in non-verbal frames in parent– infant communication (Fogel, 1993; Stern, 1985). Stern (1985) refers to the sense of self that involves an awareness of agency and narrative emotions as the core self, which begins around the third month of life. As an example, my colleagues and I made weekly videotapes of mother– infant communication for the first year of life as mother and infant played with toy objects. The examples below came from one weekly session of one of the dyads at three months. In this frame, the infant is lying on her back on the floor and her mother is sitting beside her holding a toy rattle. The infant is capable of putting her hand in her mouth yet does not have the ability to reach for objects. In the sequence described below, however, there is a transfer of the object from the mother’s to the infant’s hand. How does this occur? The infant is looking at the rattle with her hand in her mouth, as mother moves the rattle in the infant’s line of vision but out of the infant’s reach. The infant’s hand comes out of her mouth and moves toward the mother. At the same time, the mother begins to move the object toward the infant’s hand. A period of ten seconds ensues in


105

which the mother adjusts the orientation of the rattle while the infant opens her palm and watches. The mother then places the rattle in the infant’s hand. The infant grasps it and then brings the hand, containing the rattle, back to her mouth.

In this frame, the infant is oriented toward the self (hand-in-mouth) and at the same time toward the object held by the mother. The infant’s acquisition of the object, with the mother’s assistance, is then integrated into the same pattern of self-directed activity as the hand with the object is returned to the infant’s mouth. The infant’s sense of authorship in getting the object to her mouth arises as her orientations form into events and events form into a frame during a communication with the mother. It would be difficult to imagine any sense of authorship for the infant outside of a frame or narrative construction. One must have at least a minimal cause–effect sequence for the experience of self as agent. Another instance of the same frame occurs a few moments later, after the infant drops the rattle. The mother moves the rattle in the infant’s line of vision and out of reach as the infant looks at the rattle. This time, however, her hand is not in her mouth but is held out in front of her. The mother continues to move the rattle and the infant looks off to the side and puts her hand in her mouth. The mother calls her name and she turns to look at the mother as her hand comes out of her mouth. At this point, the mother tries to insert the rattle into the infant’s palm but the infant looks away once again. The infant turns to look at the rattle and the mother tries again but the infant closes her palm, displays a distressed expression, and turns to the side, fussing.

By contrasting these two examples, we can see the variability in the emergence of authorship between instances of the frame. In the first example, the emotion – attentive interest in the object – is oriented simultaneously toward self and toward the object. The continuation of that emotion depends, first of all, upon the changing communicative events within the frame, specifically, upon the mother and infant’s coregulated mutual adjustment of their hands. The continuation of the emotion of attentive interest also appears to depend upon the connection of that coregulated communication with the infant’s orientation toward the self. The object transfer event depends, in other words, upon the incorporation of the object into the infant’s self-orientation (moving the hand from the mouth, to get the object, and back to the mouth). Note also that the temporal contour of the emotion during the successful object transfer appears to be a steady focus of attention on the coordination of action. The infant is more likely to have a sense of self as author of the narrative when the communication is coregulated, the orientations become events, and the events address the emotional concerns of the self over the course of the frame (see table 5.2).

106

Alan Fogel

In the second example, the infant does not appear to have a coherent sense of authorship. This lack of a sense of authorship/agency is connected to the distinction, made earlier, about whether orientations are directed or receptive, whether one perceives a sense of control by acting toward the environment or whether the environment is perceived as acting on the individual. When orientations are open, the result is a coregulated balance between directed and receptive stances, a dialogue between self and other. When receptiveness occurs in this context, it can be perceived as a part of the sense of authorship. In the second example, however, the experience of receptiveness is not pleasant for the infant and most likely is accompanied by a loss of a sense of perceived control over the sequence of events. The unsuccessful transfer also has a different temporal contour of emotional experience, beginning with fluctuation rather than steady attention (wavering) and ending with a rapid build-up (crescendo) of distress. In the following example, also from my research, there is a balance of being directed and receptive and a corresponding sense of authorship. Once infants are able to reach and grasp objects (about five months), they are reluctant to release them. It takes another three or four months for infants to learn how to give as well as to receive objects. The following example comes from the same dataset of weekly videotape recordings. It reports on an observation of a different mother–infant dyad when the infant was nine months (see Fogel, 1993, for more details about this example). The example is the first weekly session in which we observed the infant voluntarily release the object into the mother’s hand. Mother and infant are seated across from each other at a child-sized table. The infant has a toy fork in his hand and holds it out toward his mother. As he moves the fork toward his mother, she begins to move her hand toward his hand. She opens her hand in a palm up gesture just underneath the infant’s hand. The infant orients his hand as if to place the fork into the mother’s but does not release the fork. He looks intently at his hand and at the fork and gradually begins to open his hand. The tines of the fork catch on the mother’s open palm as she slowly begins to pull her hand back toward herself, but without closing her palm. At the same time and at the same rate, the infant’s hand continues to open. At some point, the fork falls into the mother’s open hand and she grasps it and brings it up to her chest, smiling. As she does this, the infant watches the movement of the fork away from him while his hand remains extended outward and open. Finally, the infant gazes at his mother’s face and begins to smile.

In this third example, like the first example, mother and infant communicative actions were coregulated into a gently increasing temporal contour. The mother’s movements of her hand toward her body, for


107

example, were coordinated with the infant’s opening of his palm. The infant’s orientation toward concerns of the self was seen over the course of the frame in the intensity of the gaze at his own actions with the object and the care with which he released the object into the mother’s hand. The infant’s emotional experience of joy at the accomplishment of the object transfer, reflecting a sense of authorship within the frame, emerged from this intensity of interest in combination with the coregulated communication with the mother. This third example also shows the first instance of a new developmental achievement: the infant’s voluntary transfer of an object to the mother. In everyday communication, orientations serve the participants by providing an initial stance from which communication becomes possible and a relative openness to emergent and creative processes that allow communication to be spontaneous and to address the concerns of the participating selves. We can see from this example that these same properties of orientation additionally allow for the emergence of a developmental change, the emergence of novel events and frames that have never before appeared in the communication system. In the weeks following the novel object transfer event in the last example, a new frame of give-and-take appeared, allowing the emergence of new orientations that did not previously exist in this dyad and new ways of experiencing the authorship of the self-in-relation (the self who can give and receive with another). These new forms of mutual activity also brought new emotional and self experiences with them. Development is conceptualized as a change in the process by which a system’s constituents change each other to create a newly emergent frame, a change in the process of change. As frames change developmentally, so do the corresponding experiences of self and emotion. Identity It is possible for the individual to develop a sense of oneself over developmental time. Because any particular instance of framing involves an experience of self, and because framing involves the reenactment of similar patterns that have occurred in the past, there is the possibility during frames for the individual to perceive the similarities between the present experience of self and the past experiences of self. Identity is the experience that the self endures through time and that this self through time has preferential orientations that enter into particular forms of creativity, particular propensities for change and stability, and predispositions for particular patterns of framing (Bosma, 1995; Erikson, 1968; Hermans and Kempen, 1993; Lewis, 1995). Previous authors used the term identity to mean the particular

108

Alan Fogel

manifestation of self through time that begins during adolescence and involves a growing commitment to career, gender role, or religion. Without negating that use of the term, one can see more general instances of between-frame comparisons that begin as early as the end of the first year of life. During Piaget’s third substage of sensorimotor development, infants begin to compare and to integrate past and present schemes. During the fourth sensorimotor substage, infants begin to use one frame as a means for another. It is at this age, for example, that infants begin to tease their parents, showing that they can take a past frame and transform it into a means to create something emotionally different in the present (Reddy, 1991). These developments indicate that infants develop a sense of their own agency and an understanding of the causal links between current actions and past actions. This is also the age at which secondary intersubjectivity emerges, the ability to refer to a different frame outside the context of the current frame, as in pointing to an object in order to bring it into the discourse between mother and infant (Trevarthen and Hubley, 1978). According to Stern (1985), this is the period of emergence of the shared self. The coherence of the past and present in relation to the self as a whole, however, does not develop until the age of four years. After this age, children can consistently remember past experiences of themselves and weave them into a narrative account about themselves. ‘‘The child’s representational system can begin to ‘temporalize’ what were previously successive and unrelated states of the self into an organized, coherent autobiographical self-concept’’ (Provinelli and Simon, 1998: 189). This has been called the ‘‘proper self’’ (Provinelli and Simon, 1998), the ‘‘extended self’’ (Neisser, 1991), the ‘‘narrative self’’ (Stern, 1985), and the ‘‘autobiographical self’’ (Nelson, 1992). What is common across all these forms of self-awareness is the sense of an identity between frames. It could be the current frame in relation to the past (e.g. I am the self who typically is shy in social situations). Alternatively, it may be the current frame in relation to cultural frames (e.g. I am the self who is similar to a shy character in my family or community, or in literature, TV, or film). Identity is also about one’s relationship to cultural definitions of identity (I am the self who can more readily remain active in old age because there are more cultural models for successful aging). Identity is experienced with respect to reference emotions involving the relative success of making these comparisons or of meeting the standards implied by the comparisons. These emotions include satisfaction versus dissatisfaction, harmony versus conflict, achievement versus failure, frustration versus elation, and approval versus disapproval. One can accept one’s self as shy in social situations, for example, or feel dissatis-


109

fied with one’s self because of it. One can feel in harmony with cultural standards or in conflict. The self-comparison emotions can be observed in infants beginning at the end of the first year and the comparisons with others or with cultural standards (e.g. gender roles) emerge by the third year, both inside and outside the family (Emde and Oppenheim, 1995; Kitayama et al., 1995; Reddy, 1991). Although the topic of identity as typically construed brings up the issue of a cognitive representation of the self through time, I am suggesting that identity is an emotional experience, not cognition or representation. Following along the theme of this chapter, identity is the emotional perception of the coherence of a relationship over a developmental time scale. Identity, like orientation and authorship, arises in one’s experience of being a participant in a communicative relationship. Ideally, identity is created and recreated dynamically, always partly orientational, a balance between being and becoming. On the other hand, identity can become lost at the pole of becoming or centered on the self in the form of a personality disorder (cf. table 5.2 and Lewis and Ferrari, this volume). On the side of becoming, identity can lose coherence, especially during periods of developmental change (Dunne, 1996). Individuals may experience periods during which they have lost their identity, as during transitions in career, marital status, or gender role (Erikson, 1968). Ricoeur (1996), for example, defines ‘‘narrative identity’’ as the personal stories that endure against the background of life changes, a definition that could apply at any age. Identity emerges in the retelling of something about the self in the past. This is because, as shown earlier, the self is defined in the process of communication. In autobiographical writing, authors do not merely recount the past but are ‘‘deciding what to make of the past narratively at the moment of telling’’ (Bruner, 1990: 122). Identity also arises when one selects existing cultural narratives as frames in which to tell one’s own story. An individual may frame his or her identity in terms of cultural narratives for gender, speaking as females from the perspective of stories of communion or as males from the perspective of stories of independence (Gergen and Gergen, 1995; Miller et al., 1992; Mistry, 1993). The sense of identity can be perceived as an event in real time, as one reflects upon oneself across developmental time. Identity then, as an event, is partially orientational over a developmental time scale, becoming stabilized into an event over a period of years. In these coregulated dynamics, self perspectives become magnified and minified in relation to others’ narratives about the self and the resulting emotions (see Bosma, 1995; Haviland and Kahlbaugh, 1993;

110

Alan Fogel

and Lewis, 1995; for more detailed descriptions of identity/personality development). One can move this discussion to the cultural level. The types of identity narrative in a culture also develop over historical time, presumably by the same types of change dynamics discussed earlier. During the nineteenth century in Western Europe and North America, gender identity was defined with respect to the differences between men and women. In the mid-twentieth century, androgyny became more salient as a theme in gender narratives. In the latter decades of the twentieth century, gender narratives involve the self-construction of one’s own gender identity and sexual orientation (Gergen and Gergen, 1995). One of the features of the end of the twentieth century is rapid cultural change in identity narratives. When cultural change is the backdrop for individual and relationship changes, the indeterminism of selfdevelopment is increased and places increasing demands on individuals to maintain an open orientation to change at all levels in order to allow identity to develop freely (Cole, 1996; Erikson, 1968; Gergen, 1991). Time and the self The experience of self, then, can occur with respect to different time scales: events, frames, and development, as shown in table 5.1. One can expand the ideas in table 5.1 by including longer time scales. The developmental time scale, mentioned above, refers to the changes that occur every few years. The individual can become aware of cumulative changes, leading to increasingly complex senses of self and emotion. The particular form of identity that begins in adolescence seems to require a time scale of fifteen years or more before the individual becomes aware of the frame comparisons related to the self’s role commitments. This may be due in part to the fact that one needs the lengthy perspective on self and also in order to integrate the complex array of identity frames of the culture into a view of the self. According to Erikson’s (1950) theory of psychosocial development, each stage of the life course brings with it a newly emergent sense of self, involving different types of self experiences and different kinds of emotions. Preschool children, for example, take on an identity in the eyes of others, feeling pride and shame (emotions related to their view of how others perceive them) for the first time. As young adults make choices about marriage or career, they acquire sociocultural identities and feel emotions of commitment or alienation. The older adult becomes aware of an identity of a person who dies, of the uniqueness of his or her life course, and experiences emotions such as wisdom and detachment. As people near endings, especially those at the end of life, emotional


111

Table 5.3. Summary of different types of emotion according to type of change process Type of change process

Emotional experience

Orientation to event

Orientational emotions: approach vs. withdrawal, directed vs. receptive, open vs. closed Narrative emotions: love vs. hate, respect vs. disrespect, security vs. insecurity, togetherness vs. loneliness, safety vs. threat, pride vs. shame Reference emotions: harmony vs. conflict, satisfaction vs. dissatisfaction, approval vs. disapproval, achievement vs. failure, elation vs. frustration Change emotions: certainty vs. uncertainty, stability vs. instability, determinism vs. indeterminism, order vs. chaos

Event to frame

Frame to frame comparison

All changes

awareness becomes heightened as well as a corresponding appreciation for one’s unique life course (Carstensen, Isaacowitz, and Charles, 1999). In addition to the time scale of life-course duration, as described in the last paragraph, change itself has an emotional meaning for the self, at all time scales (Fogel et al., 1997). Personal histories are processes of change in time, and the change itself is one of the things immediately experienced. ( James, 1976: 25) A critical dimension in defining and describing emotional experience, therefore, focuses on the concept of changing states. (Stein, Trabasso, and Liwag, 1993: 281)

The emotions of change are certainty versus uncertainty, stability versus instability, determinism versus indeterminism, order versus chaos. Note that these emotions reflect the types of experience along the pole of being and becoming (see table 5.2). Order emerges on the being side while chaos may be felt on the becoming side. These emotions of change, as well as experiences that appear at different time scales of emergence, are summarized in table 5.3. The emotions of change refer to the relative changeability of the other emotions. During transitions in romantic relationships, for example, the relational emotions may fluctuate rapidly, between security and insecurity or between joy and despair. These fluctuations create the change emotions of uncertainty and chaos. It is by these emotions of change during transitions in togetherness or intimacy that one recognizes the importance of the relationship for the self.

112

Alan Fogel

The maintenance of creativity and orientational flexibility in the face of these uncertainties reflects a balance of being and becoming and is thought to facilitate optimal developmental change (Antonofsky and Sagy, 1990; Block and Block, 1980; Csikszentmihalyi, 1990; Csikszentmihalyi and Rathunde, 1998; Isen, 1990; Malatesta and Wilson, 1988). Change, therefore, can be experienced either as highly creative or as distressing and disorganizing, depending upon how individuals are supported in their important relationships while in the process of changing. When most creatively balanced, individuals may seek change, magnifying their emotions of fear or success, uncertainty or certainty, in order to continue acting near the edge of chaos. At other times, people may find change more aversive, keeping their emotions within prescribed boundaries and maintaining the stability of previously ‘‘safe’’ frames for communication and emotion when faced with perturbations that are likely to lead to change (Antonofsky and Sagy, 1990; Bu¨tz, Duran, and Tong, 1995; Fogel and Lyra, 1997; Lewis, 1995; Rogers, 1961; Stern, 1985). Both positive and negative emotions, then, may be important for navigating periods of change because all emotions provide information about the self’s concerns in the relationship. The idea that emotions can be usefully magnified in the service of change runs counter to traditional emotion narratives from Western culture. In these narratives, emotions are irrational impulses, originating from inside the individual, driving us to act in involuntary ways. Traditional Western narratives for mature identity, on the contrary, reflect themes of rationality, self-determination, and responsibility. ‘‘Emotions, with their alleged irrationality, would seem to seriously undermine this cultural ideal of personhood’’ (Fischer and Jansz, 1995: 61). In adopting these traditional narratives, we devalue the opportunities of change, times when events become more orientational and emotions more intense and unpredictable (Bu¨tz et al., 1995). There is a diverse variety of therapeutic methods – ranging from psychotherapy, to body and movement awareness, to meditation and spiritual practices – involving the enhancement of flexibility and creativity in the face of pain or changing, unpredictable circumstances (Antonovsky, 1993; Rogers, 1961; Wilber, 1979). The goal is finding a balance, which in the perspective of this chapter is optimal creativity and optimal stability, a balance between being and becoming, and a heightened awareness of being a coparticipant in a relational process. These approaches can help individuals develop an alternative cultural identity with narratives for connection between persons in caring relationships with others and with the environment and for trusting the inherent value of emotional experience to guide one through the currents of chaos and change.


113

Conclusions In this chapter, I have presented a view of communication, self, and emotion based on the primacy of relationships and their change dynamics. Each time scale allows the individual to perceive different aspects of the self. It is as if these different avenues of self-awareness are the natural consequence of the time it takes for events on different time scales to emerge in relationships and for the psyche to detect them. In addition, single events, authorship events, and identity events can all be perceived in the present and at the same time. Single events can change a frame more readily than they can change a culture or an identity but, theoretically, every single event is in some way part of the events that change at all levels. One of the issues left unexplored in this chapter is a detailed accounting of the coregulation that occurs across these different levels. Also unexplored in this chapter is the relationship of the authorship/ identity system to events that emerge over longer time scales. These include historical, phylogenetic, geological, and cosmological time scales. The way in which events are constituted from orientations depends in part upon the sensorimotor, emotional, and cognitive systems that evolved on this planet. The main narratives of culture depend as much on interpersonal relationships and their changes as on relationships with the earth and its changes. Another unexplored topic is the historical pathways by which orientations and events enter into frames and by which frames enter into identities. Each specific single event or frame in a relationship has a history. In some cases, events and frames form and later disappear, with no perceptible effects on the system. In other cases, particular frames and events become amplified into seeds for developmental and historical change (Fogel and Lyra, 1997; Lyra and Winegar, 1997). These relational–historical dynamics are part of a more complete understanding of communication systems and the self. More can be said about the role of the body in self-awareness. The sense of self and identity emerges from relational dynamics because the body is specialized to create emotional experiences for the temporal contours and relational meanings of events within frames. And reciprocally, the particular features of orientations, events, frames, and cultures owe their identity to the possibilities of emotion as experienced by the human body. Research methods were not discussed in detail. In general, research that follows from the theoretical principles in this chapter must be focused on the process of change over time by which orientations coalesce into events. This requires a sequential analytic approach as

114

Alan Fogel

illustrated in the narratives taken from my research. It is essential to track how a psychological process unfolds over time – at all time scales – in the context of coregulated communication between the person and the environment. It is important to understand what kinds of sequences promote open compared to closed orientations and how developmental creativity can be facilitated in relationships. Writing this chapter was an opportunity to author my own theoretical identity on the problem of how the self is created in the midst of relational changes, to become aware of my position in the scholarly community, and to realize where I am incomplete. I was comforted by the work of many others who have authored narratives of relationship and the indeterministic creativity of emotion and self. The individualistic narrative forms of the scientific culture are changing as they coregulate with narratives of dynamic systems and relationships. In this fluid culture of change, the challenge is to create a relatively stable narrative identity granting autonomy within connection, accepting self-loss against the background of self-emergence, preserving individuality as a part of the human relational ecology. Note 1 This work was funded in part by a grant to the author from the United States National Institute of Mental Health (MH48680 and MH57669). I am grateful to the following individuals for their comments on this chapter: Kari Applegate, Trevor Burnsed, Jacqueline Fogel, J’lene George, Ilse de Koeijer, Tom Malloy, Andrea Pantoja, Cory Secrist, and Dankert Vedeler.

COM M E N T A R Y

The personal experience of coherence Jeroen Jansz

Introduction As soon as we conceptualize ‘‘the self’’ as a dynamic structure that is rooted in communicative relations, we bump into the question of how to account for the continuity and consistency most people experience across time and situations. Fogel addresses this perennial question at the very beginning of his chapter when he asks how people can have a sense of ‘‘stability over time if psychological experience is fundamentally relational and dynamic?’’ (p. 93). His answer focuses on the role of emotions: people perceive consistency in their emotional experiences and this contributes to a sense of stability of the self through time. In this commentary I will first criticize Fogel’s proposal with respect to emotions, and then propose two alternative candidates for sustaining the sense of stability. The first is individual embodiment. Fogel already touches upon the importance of embodiment, but I will attribute a more fundamental role to embodied being than he does. My second candidate for sustaining stability is autobiographical memory. The recollections of people’s personal past are organized in such a way that they generate consistency and continuity in normal individuals.

Emotional experiences are context bound and subject to evaluation In this section I will take issue with Fogel’s argument of the perceived consistency in emotional experiences. According to me, it is difficult to experience consistency in emotions because the emotion process is as much subject to the dynamisms of communicative interaction as selves are. Emotions unfold in a particular communicative context, and are always evaluated with respect to their appropriateness. This generally leads to adjustments in the emotional experience. I will first briefly discuss the inextricable link between emotions and evaluation, and then sketch the standards that are used in evaluation. Emotions are in most cases judged in terms of the social context (or the culture at large), 115

116

Jeroen Jansz

but individuals may also employ a more personal standard across time and situations. The importance of evaluation with respect to emotion stands out at the beginning of the emotion process: an emotion unfolds when an event or a situation is appraised as significant for the (personal) interests of the individual (Frijda, 1986). This kind of evaluation is an intuitive and swift cognitive operation, which proceeds almost automatically. Once the emotion is experienced, it is under constant scrutiny by the person who has the feeling. It generally results in decisions about what to do about the emotion and the event that caused it (Lazarus, 1991b). In the same time span, the appropriateness of the emotion is evaluated. Someone who experiences an emotion will judge whether the emotion conforms with or transgresses the feeling rules in that context (Hochschild, 1983). These rules specify the normative relationship between the situation that evokes the emotion (and in which it is experienced and expressed) and the emotion itself. An emotion is only seen as appropriate if it emerges in the right context, at the right time, and in response to the right antecedent. As a matter of consequence, emotions are strongly regulated by the possibilities and constraints in particular social situations. Other standards of evaluation are employed in addition to the feeling rules. In this volume about identity and emotions the standards that are derived from the individual’s identity are of particular interest. The cultural conception of the person is the source model of self and identity (Harre´, 1983). Western personhood can be characterized briefly by an emphasis on rationality, autonomy, and responsibility (Fischer and Jansz, 1995). These building-blocks of identity contrast sharply with the characteristics that Western culture attributes to emotions. It is widely assumed that emotions are impulsive bodily forces that overwhelm individuals all of a sudden and result in irrational and irresponsible behavior (Fischer and Jansz, 1995; White, 1993). As a consequence, individuals who experience an emotion will be well aware of the potential threat it implies for their status of being. In order to be able to establish the actual damage emotionality has done (or will do) to their respectability, people evaluate the emotion in that particular context. Their evaluation is based on the relevant feeling rules, but it is enriched with evaluative standards that are derived from their actual or aspired identity. In our own research we focused on situations in which individuals judge a particular emotion as being at odds with their identity. Teachers at Dutch high schools were interviewed about their (professional) identity and about the emotions they experienced in class. The professional identity of a teacher requires, among other things, coping with the particularities of pupils’ behavior. What pupils do (or fail to

Commentary on Fogel

117

do) may, however, incite emotions that conflict with the aspired professional identity. This is what happens, for example, when the poor performance of a ‘‘dull’’ child causes irritation in a teacher who is strongly motivated to support disadvantaged pupils. For him, irritation is not the kind of reaction a supportive teacher should have. In our interviews with the teachers we found that this kind of a collision between a particular emotion and the teacher’s identity produces a new feeling, that is, a negative affect of a special kind. We have coined the term emotional dissonance to describe this particular feeling of unease that motivates people to reduce the unpleasant dissonance (Jansz and Timmers, in press). The experiential effects of the conflict between identity and a particular emotion underline the point that both are dynamic processes that are subject to change. On this Fogel and I agree. But our ways part when Fogel says that ‘‘identity is an emotional experience, not cognition or representation . . . identity is the emotional perception of the coherence of a relationship over a developmental timescale’’ (p. 109). I would say that identity is not a subclass of emotion because identity and emotion unfold at different levels: in judging the appropriateness of emotions, identity is used as a standard. In addition, I underline the representational nature of knowledge about oneself. The ‘‘facts and figures’’ about oneself as a person are by all means represented in that part of the cognitive system that we tend to call memory (see below). If we must look for consistency in the sense of self we can therefore better focus on the stability implied in the construction of identity rather than on emotional experiences. Consistency and continuity are, after all, important aspects of the notion of identity (Bosma, 1995). I will now link the sense of stability to our embodied being as singular entities. Embodied being The fact that people experience themselves as (rather) stable and permanent over time has a solid physical basis: their bodies and embodied experiences do not change very quickly over time. It is important to note that the body always functions as a point of location. In Fogel’s words: Since the individual occupies a location in physical and psychological space, the body is perceived as fundamentally in an orientation because all movement is from one location and toward another. (p. 97)

In social interaction the physical location of the body comes to function as a point of reference. A newborn baby is addressed as if he or she is a person long before the child can make any self-reference in its

118

Jeroen Jansz

words and actions. Parents, carers, family-members and others choose the physical location of the child’s body to address the newborn in the second person singular. In due course, all kinds of characteristics, traits, and capacities are attributed to the embodied child. It takes years before children have the capacity to refer to themselves and their bodies in terms of ‘‘I.’’ In this sense the ‘‘you’’ is older than the ‘‘I’’ (Shotter, 1989). The phenomenological continuity of the body that is present from an early age onwards is enriched with the continuity that is taken for granted in social interaction (Harre´, 1998). A particular identity is ascribed to a person in social interaction, and this identity is physically located in the individual’s body. This underlines the point that persons construct their identities in social interaction, and both actors and audiences locate the identity in the physical body. Furthermore, the identity is kept ‘‘in’’ mind once it has been constructed. These personal memories are the subject matter of the next section. Autobiographical consistency In living their lives, people accumulate a huge amount of information in their memory. The capacity of people to recollect their lives is called ‘‘autobiographical memory’’ (Baddeley, 1992). The actual autobiographical memories are of a varied nature. Most are episodic, because they are concerned with events that gained significance in the individual’s past. Other autobiographical memories are semantic. They are concerned with factual knowledge of personal significance. Next to episodes and semantics, people keep other kinds of memories in store like, for example, feelings, thoughts, movements, and, of course, odors and tastes. The present argument about personal stability draws attention to three important aspects of autobiographical memory. First, its narrative nature; second, the role of imagery; and third, the accuracy of autobiographical remembering (cf. Rubin, 1996). The narrative nature of autobiographical memory stands out in many research studies as well as in ordinary experience (Barclay, 1993). As soon as people start recalling their personal past, they generate storylike sequences with a beginning, a middle, and an end. The narrative structure of autobiographical memories brings a sequential order into information that might have been endlessly varied and chaotic at the moment of storage. Schank and Abelson (1995) claim that recall is narrative as a result of the narrative representation of autobiographical memories in the cognitive system. Other researchers have shown that this claim is too radical: personal recollections are also represented in non-narrative ways (Brewer, 1996). This brings us to the second component.

Commentary on Fogel

119

Reports of autobiographical memories frequently involve vivid images of the personal past. Imagery is in particular related to the sense of ‘‘reliving’’ an episode that is often experienced during autobiographical recall (Brewer, 1996). In recall, the imagery becomes part of a narrative structure as soon as people try to communicate the visual representation of the earlier episode. When they tell a story about a past experience they translate the image into a narrative language that will dominate their recollections. The fact that images are traded for narrative accounts is probably due to the fact that people are accustomed to making sense of their experiences by employing stories (Baumeister and Newman, 1994). The third aspect of autobiographical memories is their alleged accuracy. Reports of personal recollections show that people have more confidence in the accuracy of their autobiographical memories than in memories that are not related to the self (Rubin, 1996). The high perceived accuracy of autobiographical memories is generally explained in terms of the active reconstruction that takes place in the memory process. When people recollect their personal past they establish a narrative order in their lives whatever the actual course of their lives has been. Thus, ‘‘whereas our lives may not be coherent, our stories are’’ (Schank and Abelson, 1995: 34). The final result is that autobiographical remembering preserves ‘‘a sense of being a coherent person over time’’ (Barclay, 1996: 99). In conclusion, the argument I submit here holds that coherence in the sense of self is not preserved by consistency in emotional experiences but rather by three other constants in individual experiences. The first one is the immediate and constant experience of being embodied singularly. This is linked to the second constant: the body is the point at which to locate the individual in social reference. This underlines the point that consistency and continuity are to a large extent structured by practical patterns of interaction. The narrative nature of autobiographical recollections is the third, and final, contribution to the coherence people say they experience in their sense of self. It is worthwhile to investigate the links between identity and autobiographical memory in detail. There are strong indications that a focus on memory will help us to gain a more precise understanding of the relation between self, identity, and emotion (Christianson and Safer, 1996).

COM M E N T A R Y

The personal experience of coherence Jeroen Jansz

Introduction As soon as we conceptualize ‘‘the self’’ as a dynamic structure that is rooted in communicative relations, we bump into the question of how to account for the continuity and consistency most people experience across time and situations. Fogel addresses this perennial question at the very beginning of his chapter when he asks how people can have a sense of ‘‘stability over time if psychological experience is fundamentally relational and dynamic?’’ (p. 93). His answer focuses on the role of emotions: people perceive consistency in their emotional experiences and this contributes to a sense of stability of the self through time. In this commentary I will first criticize Fogel’s proposal with respect to emotions, and then propose two alternative candidates for sustaining the sense of stability. The first is individual embodiment. Fogel already touches upon the importance of embodiment, but I will attribute a more fundamental role to embodied being than he does. My second candidate for sustaining stability is autobiographical memory. The recollections of people’s personal past are organized in such a way that they generate consistency and continuity in normal individuals.

Emotional experiences are context bound and subject to evaluation In this section I will take issue with Fogel’s argument of the perceived consistency in emotional experiences. According to me, it is difficult to experience consistency in emotions because the emotion process is as much subject to the dynamisms of communicative interaction as selves are. Emotions unfold in a particular communicative context, and are always evaluated with respect to their appropriateness. This generally leads to adjustments in the emotional experience. I will first briefly discuss the inextricable link between emotions and evaluation, and then sketch the standards that are used in evaluation. Emotions are in most cases judged in terms of the social context (or the culture at large), 115

116

Jeroen Jansz

but individuals may also employ a more personal standard across time and situations. The importance of evaluation with respect to emotion stands out at the beginning of the emotion process: an emotion unfolds when an event or a situation is appraised as significant for the (personal) interests of the individual (Frijda, 1986). This kind of evaluation is an intuitive and swift cognitive operation, which proceeds almost automatically. Once the emotion is experienced, it is under constant scrutiny by the person who has the feeling. It generally results in decisions about what to do about the emotion and the event that caused it (Lazarus, 1991b). In the same time span, the appropriateness of the emotion is evaluated. Someone who experiences an emotion will judge whether the emotion conforms with or transgresses the feeling rules in that context (Hochschild, 1983). These rules specify the normative relationship between the situation that evokes the emotion (and in which it is experienced and expressed) and the emotion itself. An emotion is only seen as appropriate if it emerges in the right context, at the right time, and in response to the right antecedent. As a matter of consequence, emotions are strongly regulated by the possibilities and constraints in particular social situations. Other standards of evaluation are employed in addition to the feeling rules. In this volume about identity and emotions the standards that are derived from the individual’s identity are of particular interest. The cultural conception of the person is the source model of self and identity (Harre´, 1983). Western personhood can be characterized briefly by an emphasis on rationality, autonomy, and responsibility (Fischer and Jansz, 1995). These building-blocks of identity contrast sharply with the characteristics that Western culture attributes to emotions. It is widely assumed that emotions are impulsive bodily forces that overwhelm individuals all of a sudden and result in irrational and irresponsible behavior (Fischer and Jansz, 1995; White, 1993). As a consequence, individuals who experience an emotion will be well aware of the potential threat it implies for their status of being. In order to be able to establish the actual damage emotionality has done (or will do) to their respectability, people evaluate the emotion in that particular context. Their evaluation is based on the relevant feeling rules, but it is enriched with evaluative standards that are derived from their actual or aspired identity. In our own research we focused on situations in which individuals judge a particular emotion as being at odds with their identity. Teachers at Dutch high schools were interviewed about their (professional) identity and about the emotions they experienced in class. The professional identity of a teacher requires, among other things, coping with the particularities of pupils’ behavior. What pupils do (or fail to

Commentary on Fogel

117

do) may, however, incite emotions that conflict with the aspired professional identity. This is what happens, for example, when the poor performance of a ‘‘dull’’ child causes irritation in a teacher who is strongly motivated to support disadvantaged pupils. For him, irritation is not the kind of reaction a supportive teacher should have. In our interviews with the teachers we found that this kind of a collision between a particular emotion and the teacher’s identity produces a new feeling, that is, a negative affect of a special kind. We have coined the term emotional dissonance to describe this particular feeling of unease that motivates people to reduce the unpleasant dissonance (Jansz and Timmers, in press). The experiential effects of the conflict between identity and a particular emotion underline the point that both are dynamic processes that are subject to change. On this Fogel and I agree. But our ways part when Fogel says that ‘‘identity is an emotional experience, not cognition or representation . . . identity is the emotional perception of the coherence of a relationship over a developmental timescale’’ (p. 109). I would say that identity is not a subclass of emotion because identity and emotion unfold at different levels: in judging the appropriateness of emotions, identity is used as a standard. In addition, I underline the representational nature of knowledge about oneself. The ‘‘facts and figures’’ about oneself as a person are by all means represented in that part of the cognitive system that we tend to call memory (see below). If we must look for consistency in the sense of self we can therefore better focus on the stability implied in the construction of identity rather than on emotional experiences. Consistency and continuity are, after all, important aspects of the notion of identity (Bosma, 1995). I will now link the sense of stability to our embodied being as singular entities. Embodied being The fact that people experience themselves as (rather) stable and permanent over time has a solid physical basis: their bodies and embodied experiences do not change very quickly over time. It is important to note that the body always functions as a point of location. In Fogel’s words: Since the individual occupies a location in physical and psychological space, the body is perceived as fundamentally in an orientation because all movement is from one location and toward another. (p. 97)

In social interaction the physical location of the body comes to function as a point of reference. A newborn baby is addressed as if he or she is a person long before the child can make any self-reference in its

118

Jeroen Jansz

words and actions. Parents, carers, family-members and others choose the physical location of the child’s body to address the newborn in the second person singular. In due course, all kinds of characteristics, traits, and capacities are attributed to the embodied child. It takes years before children have the capacity to refer to themselves and their bodies in terms of ‘‘I.’’ In this sense the ‘‘you’’ is older than the ‘‘I’’ (Shotter, 1989). The phenomenological continuity of the body that is present from an early age onwards is enriched with the continuity that is taken for granted in social interaction (Harre´, 1998). A particular identity is ascribed to a person in social interaction, and this identity is physically located in the individual’s body. This underlines the point that persons construct their identities in social interaction, and both actors and audiences locate the identity in the physical body. Furthermore, the identity is kept ‘‘in’’ mind once it has been constructed. These personal memories are the subject matter of the next section. Autobiographical consistency In living their lives, people accumulate a huge amount of information in their memory. The capacity of people to recollect their lives is called ‘‘autobiographical memory’’ (Baddeley, 1992). The actual autobiographical memories are of a varied nature. Most are episodic, because they are concerned with events that gained significance in the individual’s past. Other autobiographical memories are semantic. They are concerned with factual knowledge of personal significance. Next to episodes and semantics, people keep other kinds of memories in store like, for example, feelings, thoughts, movements, and, of course, odors and tastes. The present argument about personal stability draws attention to three important aspects of autobiographical memory. First, its narrative nature; second, the role of imagery; and third, the accuracy of autobiographical remembering (cf. Rubin, 1996). The narrative nature of autobiographical memory stands out in many research studies as well as in ordinary experience (Barclay, 1993). As soon as people start recalling their personal past, they generate storylike sequences with a beginning, a middle, and an end. The narrative structure of autobiographical memories brings a sequential order into information that might have been endlessly varied and chaotic at the moment of storage. Schank and Abelson (1995) claim that recall is narrative as a result of the narrative representation of autobiographical memories in the cognitive system. Other researchers have shown that this claim is too radical: personal recollections are also represented in non-narrative ways (Brewer, 1996). This brings us to the second component.

Commentary on Fogel

119

Reports of autobiographical memories frequently involve vivid images of the personal past. Imagery is in particular related to the sense of ‘‘reliving’’ an episode that is often experienced during autobiographical recall (Brewer, 1996). In recall, the imagery becomes part of a narrative structure as soon as people try to communicate the visual representation of the earlier episode. When they tell a story about a past experience they translate the image into a narrative language that will dominate their recollections. The fact that images are traded for narrative accounts is probably due to the fact that people are accustomed to making sense of their experiences by employing stories (Baumeister and Newman, 1994). The third aspect of autobiographical memories is their alleged accuracy. Reports of personal recollections show that people have more confidence in the accuracy of their autobiographical memories than in memories that are not related to the self (Rubin, 1996). The high perceived accuracy of autobiographical memories is generally explained in terms of the active reconstruction that takes place in the memory process. When people recollect their personal past they establish a narrative order in their lives whatever the actual course of their lives has been. Thus, ‘‘whereas our lives may not be coherent, our stories are’’ (Schank and Abelson, 1995: 34). The final result is that autobiographical remembering preserves ‘‘a sense of being a coherent person over time’’ (Barclay, 1996: 99). In conclusion, the argument I submit here holds that coherence in the sense of self is not preserved by consistency in emotional experiences but rather by three other constants in individual experiences. The first one is the immediate and constant experience of being embodied singularly. This is linked to the second constant: the body is the point at which to locate the individual in social reference. This underlines the point that consistency and continuity are to a large extent structured by practical patterns of interaction. The narrative nature of autobiographical recollections is the third, and final, contribution to the coherence people say they experience in their sense of self. It is worthwhile to investigate the links between identity and autobiographical memory in detail. There are strong indications that a focus on memory will help us to gain a more precise understanding of the relation between self, identity, and emotion (Christianson and Safer, 1996).

C HA P T E R 6

Affective processes in a multivoiced self Hubert J. M. Hermans and Els Hermans-Jansen1

In studying affective processes, one can start from the assumption that during human development, affective responses are increasingly influenced by the relation individuals have with themselves. To illustrate this, let’s look at a crawling infant’s fascination for a bouncing ball. Initially, the ball is not experienced as part of the infant’s self, but the influence of self-reflectivity becomes explicit when the child is approximately two years old. When playing with another child, it may shout: ‘‘Ball mine!’’ The child has developed a special relationship with the ball, considering himself its ‘‘possessor.’’ The ball falls within an invisible boundary between Mine and not-Mine. The child’s definition of himself as the possessor not only indicates that the self is explicitly involved, it also has immediate repercussions for the affective processes both between the child and the ball and between the child and others. Still later, when the child has become a member of a sports club, he may consider himself a ‘‘good soccer player’’ and feel very proud of this. This self-evaluation indicates that the child, being able to adopt the complex language-games of a community, applies to himself a particular standard which is, as an organizing principle, accepted by a group or community. In Mead’s (1934) terms, the child is organizing his self by ‘‘taking the role of the other’’ and his feelings are also organized by this role-taking. As the above examples suggest, the existence of primary affect (e.g. joy in playing with a ball) is influenced by the development of the self (‘‘ball is mine’’), which strongly colors and organizes the child’s, and later the adult’s, secondary affective responses to a great variety of situations. Self and affect: dynamics and complexity Any concept of the self is unthinkable without the distinction between I and Me, originally made by James (1890) and later adopted as one of the classic distinctions in the psychology of the self (Rosenberg, 1979). In 120

Affective processes in a multivoiced self

121

James’ view, the I is the ‘‘self-as-knower,’’ whereas the Me is the ‘‘self-as-known.’’ The I has three features: continuity, distinctness, and volition. The continuity of the self-as-knower manifests itself in a ‘‘sense of sameness’’ through time (James, 1890: 332). A feeling of distinctness indicates that the I exists separately from others. Finally, the sense of personal volition shows the I as an active processor of experience, expressing itself through continuous appropriation and rejection of thoughts. These three features imply the self-reflectivity essential for the self-as-knower (Damon and Hart, 1982). James (1890), identifying the Me as the ‘‘self-as-known’’ or the ‘‘empirical self,’’ observed a gradual transition between ‘‘Me’’ and ‘‘Mine.’’ He expressed this in a frequently cited quotation saying that the empirical self is composed of all that the person can call his or her own, ‘‘not only his body and his psychic powers, but his clothes and his house, his wife and children, his ancestors and friends, his reputation and works, his lands and horses, and yacht and bank-account’’ (p. 291). As this quotation indicates, the self is not simply ‘‘within the skin,’’ but rather extends to the outside world. When someone makes a remark about ‘‘my mother’’ or ‘‘my child,’’ the feelings aroused by this remark are similar to when the remark is directed at ‘‘myself.’’ James (1890) was explicit about the relationship between self and affect: ‘‘The words ‘me’ . . . and ‘self,’ so far as they arouse feelings and connote emotional worth, are objective designations, meaning all the things which have the power to produce in a stream of consciousness excitement of a certain peculiar sort’’ (p. 319, emphases in the original). In summary, any theory of the self in the Jamesian tradition implies two fundamental principles. First, that the self is reflective, assuming an I which becomes aware of itself as a Me. Second, that the self is not a ‘‘thing in itself,’’ not a self-contained entity, but a relationship, often very intense, between a person and those parts of the environment which have personal value and affective significance. A combination of the two principles leads to the proposition of a reflective self which is affectively extended towards the outside world. The concept of a self as a self-reflecting agency, continuously involved in I-world connections, as Boesch (1991) calls it, has profound implications for understanding the relationship between self and affect. A person who reacts fearfully may feel ashamed of his fear, being aware of other people watching him. When someone blushes with shame, that person may be ashamed of his shame so that the initial affective response is intensified. People may even correct or suppress an initial feeling by fighting against it with anger and thus ‘‘give themselves a kick.’’ In other words, as a result of self-reflectivity, an initial feeling may become intensified, combined with, or even suppressed by other

122


feelings. Feelings evoke feelings as a result of one’s own judgment or the supposed judgment of other people. Looking at feelings this way, they become phases or subphases of dynamic processes which take the form of highly complex affective structures. A self-reflecting person not only responds to her own affective impulses, she is also able to remember, foresee, and compare experiences which are seen as similar, different, or contrasting. Humans are, as Becker (1973) has argued, ‘‘time-binding animals,’’ being capable of relating their present to past or future. This active and continuous process of relating deeply influences present experiences. When someone says ‘‘I am already sixty-nine years old,’’ the word ‘‘already’’ may indicate that this person feels his life has run fast and little time is left. Such a statement may imply the ambivalent feeling of satisfaction about things achieved and dissatisfaction about running out of time. This condensing of experiences dispersed across time and space results in affective structures which can only be understood by virtue of the specifically human capacity of transcending the immediate situation. This complex affective structuring was recently revealed by a former Dutch prime minister who was asked how he felt at the end of his career. He answered that he felt ‘‘dishonestly happy’’ and explained that, although he felt happy for the rich life he had lived, he worried at the same time about the ‘‘unequal distribution of luck in the world.’’ In summary, the use of terms like ‘‘already,’’ ‘‘again,’’ ‘‘still,’’ and ‘‘the first time,’’ suggest that great realms of human affect are part of the dynamics of the self. With its reflectivity and I-world connection, the self evokes complexities of similar, different, and contrasting feelings which together form highly dynamic structures. Self-feelings should not simply be understood as anonymous or automatic processes, but, rather, as affective processes in which human agency and intention (James’ ‘‘volition’’) play an organizing role. The multivoicedness of the self Whereas the preceding considerations correspond closely with James’ founding work, we will make two additional theoretical steps in order to explain the multivoicedness of the self. The first step concerns the basic character of time and space in the narrative construction of the mind; the second concerns the metaphorical importance of the notion of voice.


123

The narrative turn in the psychology of the self Social psychologist Sarbin (1986b) made an important theoretical step by translating the I–Me distinction into a narrative framework. He proposed that the pronoun I stands for the author and Me for the actor. The self as author, the I, imaginatively constructs a story in which the Me functions as the protagonist, and other figures that are part of Mine as antagonists (e.g. ‘‘my wife,’’ ‘‘my father,’’ ‘‘my friend,’’ ‘‘my enemy’’) (Verhofstadt-Dene`ve, 1995, 1999). Sarbin (1986b) views the narrative as a way of organizing episodes, actions, and accounts of actions in time and space, as an achievement that brings together fact and fiction in coherent patterns. Our fantasies and daydreams are organized as stories, and our plannings, our rememberings, even our loving and hating, are guided by narrative plots. Bruner’s (1986, 1990) writings on narrative have been particularly influential in drawing on the distinction between the imaginative quality of the narrative mode and the deductive quality of the logicoscientific mode. The narrative mode leads to ‘‘good’’ stories, gripping drama, and believable historical accounts; the logico-scientific mode, also called ‘‘argumentation,’’ refers to the ability to see formal connections between statements as part of a reasoning process. Whereas the narrative mode emphasizes human intention as an organizing principle behind specific events and actions, the logico-scientific mode seeks to transcend the particular by reaching for increasingly higher levels of abstraction. Whereas the logico-scientific mode aims to make abstraction from the specific location of people in action, the narrative mode puts the generalities of the human condition into the particulars of experience, and attempts to locate experience in time and space (Bruner, 1986). The particulars of time and space as central ingredients of storytelling have also been discussed by Jaynes (1976), who also used the distinction between I and Me in describing the self as ‘‘mind-space.’’ The I constructs an analog space and metaphorically observes the Me moving in this space. When we plan to visit someone, we imagine the house we are going to and we see ourselves talking to a friend. In the functioning of our consciousness, the I sees the Me as the main figure in the story of our life. Seated where I am, Jaynes explains, I am writing a book and this fact is imbedded in the story of my life: ‘‘Time being spatialized into a journey of my days and years’’ (p. 63). Like Sarbin (1986b), Jaynes points to the capacity of stories to include not only the course of actions but also the motives behind these actions. Thieves may narratively explain their act as being due to poverty, poets to beauty, scientists to truth. The narrating I observes the Me as an actor

124


located in time and space and relates a story about the movements of the Me, including its motivation, as parts of a narrative construction. Voices in Bakhtin’s polyphonic novel In his book Problems of Dostoyevky’s Poetics (1929/1973), originally published in Russian, Bakhtin develops the thesis that Dostoyevsky – one of the most brilliant innovators in the history of fictional writing – created a new form of artistic thought, the polyphonic novel. The principal feature of the novel is the retreat of the author as the central organizing figure who is ‘‘above’’ his writings and has an overview of all his characters. In the polyphonic novel there is not one single author, but several authors or thinkers (Raskolnikov, Myshkin, Stavogin, Ivan Karamazov, the Grand Inquisitor) each having his own voice and telling his own story. The hero is not simply the object of Dostoyevsky’s finalizing vision, but comes across as the author of his own ideology. The novel is composed of a number of independent and mutually opposing points of view embodied by characters involved in dialogical relationships. Dostoyevsky himself is only one of the many characters. Each character is ‘‘ideologically authoritative and independent,’’ which means that each character is perceived as the author of his own legitimate ideological position instead of being an object of Dostoyevsky’s all-encompassing vision. This ‘‘retreat of the omniscient author,’’ so characteristic of modern novels (Spencer, 1971), implies that the characters are not ‘‘obedient slaves,’’ in the service of Dostoyevsky’s intentions, but are capable of standing beside their creator, disagreeing or even rebelling against him. As in a polyphonic composition, the several voices or instruments have different spatial positions and accompany and oppose each other in a dialogical fashion. Logical versus dialogical relationships In order to understand Dostoyevsky’s polyphonic novel, it is necessary to establish the difference between logical and dialogical relationships. Bakhtin gives the following example (see also Vasil’eva, 1988). Consider two phrases that are completely identical, ‘‘life is good’’ and, again, ‘‘life is good’’. In terms of Aristotelian logic, these two phrases are related in terms of identity; they are, in fact, one and the same statement. From a dialogical point of view, however, they are different because they may be seen as two remarks voiced by two spatially separated people, who in this case entertain a relationship of agreement. The two phrases are identical from a logical point of view, but different as utterances: the first is a statement, the second a confirmation. Similar-


125

ly the phrases ‘‘life is good’’ and ‘‘life is not good’’ can be analyzed. In logical terms, one is a negation of the other. As utterances from two different speakers, however, a dialogical relation of disagreement exists. According to Bakhtin, agreement and disagreement are, like question and answer, dialogical forms. To better understand Bakhtin’s position it should be added that he certainly does not reject the rules of logic: ‘‘Dialogical relationships are totally impossible without logical and concrete semantic relationships, but they are not reducible to them; they have their own specificity’’ (Bakhtin, 1929/1973: 152). The notion of dialogue opens for Bakhtin the possibility of differentiating the inner world of one individual in the form of an interpersonal relationship. By transforming an ‘‘inner’’ thought of a particular character into an utterance, dialogical relations spontaneously occur between this utterance and the utterance of imaginal others. Dostoyevsky’s novel The Double may serve as an example. The second hero (the double) was introduced as a personification of the interior thought of the first hero (Golyadkin). This externalization of a thought in a spatially separated opponent instigates a fully developed dialogue between two relatively independent parties. Such a dialogical narrative not only includes space and time, but temporal relations are even translated into spatial relations. The result is that temporally dispersed events are contracted into spatial oppositions which are simultaneously present. In Bakhtin’s terms: ‘‘This persistent urge to see all things as being coexistent and to perceive and depict all things side by side and simultaneously, as if in space rather than time, leads him [Dostoyevsky] to dramatize in space even the inner contradictions and stages of development of a single person . . .’ (p. 23, emphasis added). Imaginal dialogues in daily lives In her book Invisible guests, Watkins (1986) describes the powerful influence of imaginal dialogues on people’s daily life. We may find ourselves speaking to the photograph of someone we miss, to our cat or dog, or to our reflection in the mirror. Even when we are outwardly silent, we may be talking with our mothers or fathers, opposing our critics, or questioning some personification of our conscience. After a critical discussion of the theories of Piaget, Vygotsky, and Mead, Watkins (1986) concludes that most psychological theories view imaginal phenomena only through the eyes of the ‘‘real.’’ The imagination is often seen as a tricky opponent of the real, as little more than a mimic or a helpmate, always ready to respond to the real. Most theories give a clear ontological priority to actual or ‘‘real’’ others, usually considering imaginal others as derivative from or subordinate to them.

126


Caughey (1984), another student of imaginal phenomena, is also critical of the narrow assumptions that many Western social scientists make about imaginal phenomena. As a cultural anthropologist he performed field work in Micronesia and Pakistan, and compared his observations with the daily life of Americans. He concluded that it is erroneous to think that only people in cultures with thunder gods and fertilizing spirits live in an imaginal world. The daily world of the average American, Caughey observed, is also populated by a throng of beings with whom no physical interaction exists. He identified at least three categories of imaginal figures: (a) media figures with whom the individual engages in imaginal interactions (e.g. pop stars and other celebrities); (b) purely imaginary figures produced in dreams or fantasies; and (c) imaginal replicas of parents, friends, or lovers who are treated as if they are really present. For Caughey, the interactional nature of these relationships was reason to speak of an (imaginal) ‘‘social world’’ rather than a purely ‘‘inner world.’’ Given the personal importance of these imaginal interactions, Caughey was critical of the identification of ‘‘social relationships’’ with only ‘‘actual social relationships.’’ He saw this conception as incomplete because ‘‘such a supposedly objective approach to social organization actually represents an ethnocentric projection of certain narrow assumptions in Western social science’’ (p. 17). Discussing the multivoicedness of the self, it is important to note that the individual is not only able to talk about the variety of imaginal others, but also to talk with them as relatively independent parts of an extended self. Cassirer (1955) described the role of a guardian spirit in mythical conciousness, emphasizing that the spirit might be closely associated with someone, and perhaps even – like thoughts – inhabit his body or govern his behavior. Even in this case, Cassirer continued, the spirit should not be conceived ‘‘as the man’s I, as the ‘subject’ of his inner life, but as something objective, which dwells in man, which is spatially connected with him and hence can also be spatially separated from him’’ (p. 168, quoted by Watkins, 1986: 93). All these observations, from both Western and non-Western cultures, suggest that the imaginal is not simply ‘‘outside the I.’’ Rather, together with a multiplicity of imaginal figures the I constructs a multivoiced world. In other words, imaginal figures are voiced positions in a self-space, and sometimes these positions entertain dialogical relationships. The preceding observations referring to a process of intersubjective exchange with an imaginal figure suggest that the frequently used term ‘‘projection’’ is an unsatisfactory label. It reduces the process to only one voice and, consequently, to only one author. Moreover, it ignores the fact that as well as the person creating an imaginal figure, the


127

imaginal figure is creating the person (see, for example, the many accounts by artists of an inspirational figure). I-positions in a multivoiced self The metaphor of the polyphonic novel expands on the narrative conception of the I as author and the Me as observed actor. Whereas in Sarbin’s (1986b) conception of the self-narrative, a single author (the I) is assumed to tell a story about him- or herself as an actor (the Me), the polyphonic novel as metaphor for the self goes one step further. It permits one individual to live in a multi-authorial world, in a multiplicity of worlds with each world having its own author telling a story relatively independent of the authors of the other worlds. Moreover, the several authors have the possibility to enter into dialogue. Elaborating on the polyphonic metaphor and its implication of spatialized dialogue, Hermans, Kempen, and Van Loon (1992) conceptualized the self in terms of a dynamic multiplicity of relatively autonomous I-positions in an imaginal landscape; the I has the ability to move from one position to the other in accordance with changes in situation and time. The I fluctuates between different and even opposed positions, and has the capacity to imaginatively endow each position with a voice so that dialogical relations between positions can be established. The voices function like characters in a story, involved in a process of question and answer, agreement and disagreement. Each of them has a story to tell about their own experiences from their own standpoint. As different voices, they exchange information about their respective Me’s, resulting in a complex, narratively structured self. In the following section we will translate the preceding considerations into a more specific psychological theory (valuation theory) and concrete procedure (self-confrontation method) allowing us to study the dialogical self in an empirical way. Valuation theory: personal meanings in a multivoiced narrative Valuation theory (Hermans, 1987a, b; Hermans and Hermans-Jansen, 1995) was originally developed to study the ordering and development of experiences into a narrative structure. The underlying view of the person was inspired by philosophical–phenomenological thinking (James, 1890; Merleau-Ponty, 1945) and, later, by Bakhtin’s work. In valuation theory, the self is conceived as an ‘‘organized process of valuation.’’ The process aspect refers to the historical nature of human experience and implies a spatio-temporal orientation: the person lives in the present and is, from a specific point in space and time, oriented

128


towards both past and future. The organizational aspect emphasizes that people not only orient to different aspects of their spatio-temporal situation but also tell different stories about their past, present, and future depending on the actual or imaginal positions from which these stories are told. Through the act of self-reflection and telling, the person is able to bring different experiences and views together into a composite whole. In this composite whole some parts become more influential than others. The central concept of valuation is very open to the everyday world of the individual. It refers to anything people find important when talking about their life-situation. It is any unit of meaning that has a positive (pleasant), negative (unpleasant), or ambivalent (both pleasant and unpleasant) value to them. Valuations can include a broad range of phenomena: a dear memory, a difficult problem, an unreachable ideal, an intriguing dream, and so forth. The person is differentially oriented towards present, past, and future, resulting in different valuations. The different valuations are combined into a valuation system which expresses an individual’s story from the perspective of one position. When someone is invited to tell his or her story from the perspective of different positions (e.g. I as a teacher, I as the person I am in my dreams), they are supposed to be able to formulate different valuation systems. In other words, an individual may tell a different personal history and value this history differently, depending on the position from which it is told. The theory assumes that each valuation has an affective connotation represented by a particular pattern of affect. When we know which types of affect are characteristic of a particular valuation, we know something about the valuation itself, which also implies that the affective meaning of a valuation cannot be separated from it. Moreover, the affective component reveals the organization of a specific valuation system, and, in the case of several valuation systems, the organization of the different systems as a differentiated whole. As part of the affective component of valuation theory, the latent– manifest distinction was introduced. At the manifest level, a large number of personal valuations come and go. This phenomenological richness varies not only across individuals but also within a single individual across time and space. At the latent level, a limited number of basic motives are assumed to be in operation. Active in everyone in all periods of life, these motives are reflected (latently) in the affective component of the valuation system. Study of the affective component can therefore reveal which particular motive is active in a particular valuation or valuation system. The basic motives reflect the person as a motivated storyteller (Hermans and Hermans-Jansen, 1995).


129

Two motives have been elaborated: the striving for self-enhancement, or S motive (i.e. self-maintenance and self-expansion), and the longing for contact and union with the other, or O motive (i.e. the search for alliance with somebody or something else). This distinction was the result of a literature review in which the perspectives of various authors on the basic duality of human experience were considered. For example, agency and communion for Bakan (1966); autonomy (or selfdetermination) and homonomy (or self-surrender) for Angyal (1965); Bindung (solidification) and Lo¨sung (dissolution) for Klages (1948); power and intimacy for McAdams (1985). Note that in valuation theory the basic motives are considered not as causal explanations but rather as basic themes in people’s self-narratives. Moreover, the supposition of two basic motives has no exhaustive pretensions. It is rather a choice on the basis of theoretical parsimony. When a valuation represents a realization of the S motive (e.g. ‘‘I passed a difficult test’’), the person experiences a feeling of strength and pride in connection with the valuation. Similarly, when a valuation represents a realization of the O motive (e.g. ‘‘I enjoy the moments with my friend’’), feelings of intimacy and tenderness are associated with the valuation. In other words, the latent motivational base becomes manifest in the affective pattern associated with a particular valuation. Generalization and idealization Two concepts, generalization and idealization, particularly represent the organizational aspect of valuation theory and play a central role in the methodology presented in the next section. The more a particular valuation generalizes as part of a system, the more it determines the ‘‘general feeling.’’ When people tell how they generally feel, it is highly probable that particular experiences color this general feeling more than others. Not all valuations are equally influential in the system. The more generalizing power a valuation has, the more influential the affective component of this valuation is in determining how someone generally feels during a particular period of life. Similarly, valuations differ in the extent of idealization. This concept refers to the fact that certain valuations better fit how someone would like to feel (the ‘‘ideal feeling’’) than others. Valuations that color the ideal feeling are often different from those that influence the general feeling. When people are faced with personal problems which affect their selves to a significant degree, the ideal feeling typically has an affective modality that is in contrast to the affective modality of the general feeling.

130


Self-confrontation method: organization and reorganization of valuations The self-confrontation method is an idiographic procedure based on valuation theory. It is designed to study relations between valuations and types of affect and the way these variables become organized into one or more systems (Hermans and Hermans-Jansen, 1995). The procedure involves elicitation of a set of valuations and then association of these valuations with a standardized set of affect terms. The result is a matrix in which each cell represents the extent to which a specific affect is characteristic of a specific valuation. As we will see, the self-confrontation method will enable us to study valuational and affective structures belonging to I-positions in detail. The valuations (rows in the matrix) are elicited using a series of open-ended questions. The main questions are intended to bring out some important units of meaning for the past, present, and future (for details see Hermans and Hermans-Jansen, 1995). The questions invite individuals to reflect on their life and to mention those concerns that are most relevant from the perspective of the present situation. Also, subjects are encouraged to phrase the valuations their own way so that the formulations reflect the intended meaning. The typical form of expression is the sentence as the basic unit of text. In the second part of the investigation, a standard list of affect terms (columns in the matrix) is presented. Concentrating on the first valuation, subjects indicate on a 0–5 scale the extent to which they experience each affect in relation to the valuation. The subject, working alone now, rates each valuation with the same list of affect terms. The different valuations can then be compared according to their affective profiles. Once the affective ratings for the different valuations have been obtained, a number of indices that represent the motivational structure of the valuation system are calculated. 1. Index S is the sum of the scores for four affect terms expressing self-enhancement. 2. Index O is the sum of the scores for four affect terms expressing contact and union with the other. 3. Index P is the sum of the scores for four positive affect terms. 4. Index N is the sum of the scores for four negative affect terms. (Note that the scores for each of the four indices S, O, P, and N range from 0 to 20 for each valuation.) In summary, the degree of S and O affect indicates the extent of realization of the basic motives, whereas the degree of positive and negative affect reflects the extent in which these motives meet superable or insuperable obstacles.


131

5. For the extent of generalization (G) of a valuation, the following question is asked at the end of the valuation construction phase: ‘‘How do I generally feel these days?’’ This question does not ask for a specific valuation but is intended to assess the ‘‘general feeling.’’ The person answers directly with the list of affect terms that was used for the characterization of the valuations. The product-moment correlation between the pattern of affect that belongs to a specific valuation and the pattern of affect that belongs to the general feeling reflects the extent of generalization of this valuation. The more positive the correlation, the more this valuation is supposed to generalize. 6. For the assessment of the idealization (I) of a valuation, the question is: ‘‘How would I like to feel?’’ The correlation between the affective profile belonging to a specific valuation and the affective profile belonging to the ideal feeling indicates the ideal quality of this valuation. The strength of the correlation indicates the extent of idealization. The more positive the correlation, the higher the idealization. When a valuation has an affective profile that contrasts with the ideal feeling, this is expressed in a negative (minus) correlation. (G and I represent the last rows in the matrix, see table 6.1). (For reliability and validity data of all the indices presented so far, see Hermans, 1987b.) The results are discussed with the person, usually one week later so that the psychologist can analyze and study the data and the person can digest the many impressions, associations, and thoughts evoked by the self-investigation. The discussion has the quality of an intensive selfreflection and profound dialogue with the interviewer. The person and the interviewer base their discussion on the overall picture provided by the system of elicited valuations. A second (and sometimes a third and a fourth) self-investigation usually follows, typically after some months. In this case, however, the subjects do not start ‘‘from scratch.’’ Instead, the interviewer reads the original questions and, after each question, produces the statements that they provided in the preceding self-investigation. The subjects are instructed to consider, for each statement separately, whether they would still respond the same. Where this is not the case, the interviewer offers various options: an old valuation may be reformulated (modification); replaced (substitution); discarded altogether (elimination); or a new response may be added (supplementation). This way, subjects have considerable freedom and indirectly point to the constant and changing elements in their own valuation systems.

132


Sharon’s I-positions: telling and retelling a self-narrative In one of our projects on the multivoiced self, we considered opposite personality sides as I-positions, approaching them as voices. The central question we posed was: ‘‘Can you consider two sides of your personality, which you see as opposite, with one of the sides more dominant than the other side?’’ One subject may say: ‘‘I have an open side, but also, more hidden and less familiar, a closed side.’’ Others may state different opposites, for example, ‘‘serious versus playful,’’ ‘‘active versus passive,’’ ‘‘hard versus soft,’’ etc. After describing the opposites in their own terms, we invite them to focus on the first, dominant side of their personality and to think and feel about themselves, say, as an open person, about their past, present, and future and tell about important experiences and circumstances from this particular perspective (I-position). This results in a valuation system of this subject as an open person. Next, they are invited to reflect about themselves as a closed person, and tell about their life from this particular perspective. In other words, the person constructs two valuation systems, each with their own content and affective organization, which then can be studied and followed across time. Sharon, a 34-year-old married mother of two children and part-time teacher, contacted a psychotherapist (the second author) in a period of distress. The immediate reason was a nightmare, in which she saw her husband and two children at an unnatural distance, experiencing herself as if she were dead: ‘‘In this dream I thought: ‘perhaps I’m dead now . . . they will manage, it makes no difference to me’; I felt no sorrow, no anger, I felt nothing.’’ She became convinced that this dream symbolized the way she often experienced herself in daily life: being continuously busy, feeling increasingly alienated from her own feelings. Because the dream frightened her, she told her mother, but her mother didn’t respond. Sharon increasingly became aware of her dream as revealing an important but neglected part of her life, both present and past, and therefore she decided to contact a psychotherapist. After some sessions the psychotherapist proposed that she investigate herself using the self-confrontation method. As part of this investigation, Sharon was asked the question about the existence of two opposite personality sides (see above). She phrased her dominant side this way: ‘‘My I which puts my respect outside of myself’’ and explained that she always tried to conform to the high expectations that others, in particular her parents, had of her achievements and visible presentation. Her opposite side, which she considered less dominant, even suppressed, was labeled ‘‘My own world of feelings.’’ This side represented her deeper, affective life from which she felt closed-off, as


133

indicated by her dream. The valuations and affective indices which Sharon formulated from the perspective of her ‘‘outside’’ (Position A) and her ‘‘inside’’ (Position B) are presented in the left part of table 6.1. A direct insight into the difference between Sharon’s two opposite positions (A and B) can be found by inspecting the affective properties of her ideal feeling. Whereas for Position B this is high for both S and O, for Position A it was high in S and low in O, suggesting that from the perspective of this position, she is oriented to an ideal characterized by high self-enhancement and low contact and union. Looking at the specific valuations of Position A, we see that valuation 1, concerning ‘‘being important,’’ is associated with much self-enhancing affect in combination with a high level of positive affect. Also, this valuation correlates highly (.89) with the ideal feeling, suggesting that ‘‘being important’’ has an ideal quality from the perspective of this position. Strikingly, as part of valuation 3, the same theme has a high level of negative affect in combination with low levels of both self-enhancement and contact. Moreover, valuation 3 strongly generalizes within this position (A), which suggests that this valuation is most influential within Position A. Apparently, Sharon finds herself in an internal conflict: what she remembers as positive in the past (valuation 1), is perceived as negative in referring to her future (valuation 3). In other words, from the perspective of Position A, Sharon is oriented to selfenhancement but does not see much prospect for pursuing this ‘‘outside goal’’ in the future (see also the contrast between general and ideal feeling). Whereas Position A is governed by the self-enhancement motive, Position B focuses primarily on contact and union (see the high levels for O affect for valuations 6, 7, 8, and 9). This strong orientation is also indicated by the high level for O affect in the ideal feeling. Despite this strong orientation towards contact and union, the valuations high in O do not generalize (see the low levels of G for valuations 6, 7, 8, and 9). Instead, valuations 4 and 5, referring to contact with her parents, show low levels for both S and O, and strongly generalize within the system of Position B (.93 and .86, respectively). Apparently, Sharon is longing for contact and union, but cannot fulfill her desire because she feels her awareness of her parents’ inability to express their emotions is blocking her. This lack of fulfillment is also represented by the affective contrast between general and ideal feeling. In summary, Sharon’s self-investigation shows that whereas Position A is oriented to self-enhancement, Position B is primarily oriented to contact and union, both orientations being unrealized. The two positions represent disagreeing voices in Sharon’s self and are associated by contrasting affect.

Table 6.1. Sharon’s valuations and corresponding affective indices for positions A, B, and C Position A/Time 1 My I which puts my respect outside of myself (past) 1. My work, my educational training, that is, my visible appearance, that was most important for my parents (present) 2. I’ve a desire for work and for a position in society (future) 3. Self-confidence, which I get from my work, knowledge, and being important, is my outside goal General feeling Ideal feeling

S

O

P

16

1

16

15

1

2

4 12

N

Position C/Time 2 My own world of feelings as basis for a well-functioning outside G

I

7

−.23

.89

9

11

.22

1

1

18

3 2

2 11

S

O

P

N

G

I

My work, that’s what I can 15 do, that’s what I am, and, moreover, I work with a group in which people are very involved with each other

7

13

0

.75

.75

.66

I’ve a desire for work and for a position in society

12

0

11

9

.06

−.05

.92

−.32

I’m satisfied with what I have now; in my private life and in my work, I no longer put myself under pressure to achieve more

18

7

18

1

.70

.78

19 3

— −.38

−.38 —

N

G

I

S

O

P

N

G

I

Position B/Time 1 My own world of feelings S (past) 4. I’ve missed having a mother who had feelings and emotions and showed them (past) 5. My father also didn’t show his emotions; I’ve the suspicion that he has them but can’t cope with them (present) 6. When I talk with my children, when I’m angry, powerless or distressed, I want them to see something of this (present) 7. When I was sick, and Sheila [daughter] was looking for something, I loved her very much and felt sad that I couldn’t care for her (present) 8. After the nightmare I felt again that I loved my children; I also felt in touch with the people around me

P

0

0

0

17

.93

−.94

[valuation eliminated]

2

3

0

13

.86

−.83

[valuation eliminated]

12

20

14

4

−.62

.80

When I talk with my children, when I’m angry, powerless or distressed, I want them to see something of this

9

13

11

6

.38

.50

8

18

13

17

.29

−.16

I’m feeling intensely in touch with my children and people around me

10

14

17

3

.75

.83

9

17

15

13

.12

.10

[valuation combined with valuation 7]

15

14

2

−.80

.84

My feelings and emotions come sooner to the surface; it becomes clear to me more quickly where they come from and to which aspects of my life they are related

13

8

12

4

.73

.65

3 20

4 16

20 0

— −.90

−.90 —

General feeling Ideal feeling

15 18

14 18

15 20

8 1

— .82

.82 —

(future) 9. I wish that my feelings 14 would leave these extremities, that I could become aware of them in time, and get more of a grip on this General feeling Ideal feeling

O

3 17

Note: S = affect referring to self-enhancement; O = affect referring to contact with the other; P = positive affect; N = negative affect; G = generalization; I = idealization

136


After her self-investigation, Sharon continued seeing the therapist on a two-weekly basis. After six months they decided to perform a second self-investigation to examine whether the positions and valuations from Investigation 1 had been changed. When asked to consider the two personality sides as distinguished in Investigation 1, Sharon now proposed to combine Positions A and B into a third Position C, which she described as ‘‘My own world of feelings as basis for a well-functioning outside.’’ She explained that her original Positions A and B, which she had experienced as strongly opposed and inconsistent, had been losing much of their conflicting character so now they could be combined. The valuations belonging to Position C are presented at the right side of table 6.1. As can be seen from table 6.1, Sharon eliminated some of the valuations and modified others, in accordance with the procedure described earlier. Comparing the first self-investigation (Time 1) with the second (Time 2), we see a conspicuous change in the general feeling. Being strongly negative at Time 1, it is positive with high levels for both S and O at Time 2. The valuations with the highest degree of generalization are nos. 1, 3, 7, and 9. Valuation 9 is quite significant from a psychotherapeutic point of view, telling that Sharon has learned to bring her feelings ‘‘sooner to the surface’’ and has developed a capacity to relate her feelings to specific experiences. As a result of her increased psychological skills, Sharon has succeeded in bringing her ‘‘outside’’ and her ‘‘inside,’’ represented by two strongly contrasting Positions A and B at Time 1, together into a more integrative Gestalt, represented by Position C at Time 2. This change is also manifest in valuation 1: at Time 1 this valuation referred to her ‘‘visible appearance,’’ the modified valuation at Time 2 combined an achievement element (‘‘. . . what I can do . . .’’) with a social element (‘‘I work with a group in which people are very involved with each other’’). The overall change is reflected by the high correspondence between the affective modalities of the general and ideal feelings at Time 2. Notably, Sharon had deleted all references to her parents which played such an important role at Time 1 (see valuations 1, 4, and 5). Despite the valid reasons which may exist for concentrating on her positive social relationships, Sharon’s parents are of direct significance to her well-being because they are not simply ‘‘external’’ persons, but part of her self. Sharon recognizes the emotional problems of her parents in herself, as exemplified by her nightmare in which she found herself empty of feelings, a problem which is similar to the problem perceived in her parents. On the assumption that the parents are relevant parts of Sharon’s self, the question that remains is whether Sharon is capable of developing her self-system further without


137

working through her relationship with her parents. On the other hand, it should be emphasized that Sharon has made progress in coping with her feelings (see valuation 9 at Time 2), so that the possibility exists for her to develop further the coherence of her self without directly focusing on the relationship with her parents. From a theoretical point of view, Sharon’s case illustrates some of the ideas and concepts discussed at the beginning of this chapter. She did a self-investigation not from ‘‘one and the same I.’’ She did her selfinvestigation not from one centralized position, as is usual in interviews or test administration. Rather, we approached Sharon as a multivoiced individual, able to reflect on her self from two opposite positions and tell about significant meaning units from the perspective of these positions. She answered the questions in the self-confrontation method twice, coming up with different answers and producing different affective profiles. The two initial positions can be considered as two disagreeing voices or oppositional characters in a multifaceted selfnarrative. To a large extent, Sharon’s emphasis on solidarity and intimacy with significant others (Position B) can be understood as an answer to the threatening shallowness and emptiness of her ‘‘outside’’ (Position A). In the second investigation, she made a significant step in the direction of agreement between the two voices as combined in a third, more integrative system (Position C). As Josselson (1995) would say, from the latter position she tells a self-narrative superseding and encompassing the preceding ones. Different voices, different affect As we have seen in our case study, voices, or, using a more spatial term, I-positions, are not simply ‘‘internal affairs’’ representing an intrapsychic world of thoughts and feelings. As Voloshinov (1929/1986) has argued, a word is a two-sided act. It is not only determined by the speaker but also by the imaginal or real person for whom the word is meant. The word is the product of the reciprocal relationship between speaker and listener, between addresser and addressee. In other words, a word is a bridge. As Sharon’s case study shows, I-positions are embedded in social relationships. Her positions can be understood as localized speakers entertaining active, reciprocal relationships with the social environment and with each other. Not only her first position (her outside), but also her second position (her feeling world rooted in her inside) were embedded in reciprocal relationships with the environment, as the valuations formulated from the perspective of this position show. Her ‘‘own feeling world’’ explicitly referred to her relationship with the

138


people around her. These reciprocal relationships are even at the heart of so-called ‘‘inner feelings’’ and are a challenge to our inside–outside dichotomies. As Sampson (1993) has argued, the boundaries between inside and outside or between intrapersonal and interpersonal are by no means as neat as we had believed: ‘‘What is inside is simultaneously outside, and vice versa’’ (p. 134). People who are ‘‘outside’’ are internally reconstructed as imaginal others and are also present when we define ourselves as being ‘‘alone.’’ In fact, we are never alone, as our minds are continuously populated by the voices of ‘‘external’’ people. As we have seen in this chapter, imaginal others can be considered as examples of I-positions. This concept, however, can be applied in more ways. Personality traits or Kelly’s (1955) personal constructs can also be conceived of as I-positions (Hermans, 1996a, b). In Sharon’s case study, we have examined idiosyncratically phrased personality sides as Ipositions. All these approaches suppose that people can tell a variety of stories about themselves, from the perspective of voiced positions localized in an imaginal space. Given the blurring boundaries between inside and outside, this imaginal space is closely intertwined with people’s actual space. By implication, interpersonal and intrapersonal relationships are always side-by-side and even interwoven. The term I-position can be best understood as a reaction to the traditional Cartesian split between subject and object. As Overton (1998) explains, Descartes’ Cogito ergo sum was a historically defining moment because it set the subject (I) against the (external) object, and, hence, created the overarching dichotomy that was to become known as the Cartesian split. This became the origin of ontological dualism: subject versus object, mind versus body, thinking versus space were each presented as independent, isolated realities. As a consequence of the split, psychologists conceived the I as a soul, supposed to have an existence by itself, unified in itself, and separate from externally located space. As discussed at the beginning of this chapter, James (1890) corrected the traditional notion of the separate soul by conceiving of a dynamic self which was explicitly extended to the environment. His emphasis, however, remained on the I as unified in itself and consistent over time. It should immediately be added that James was well aware of the multiplicity of the self as demonstrated by his discussion of the ‘‘rivalry and conflict of the different selves’’ (p. 309). Despite James’ insight into the multiplicity of the self, Bakhtin’s (1929/1973) dialogical approach can be seen as a decisive step beyond James’ and even Mead’s theorizing, because he accepted the full consequence of the linguistic nature of the human being. The I-position concept can be seen as a correction to the Cartesian


139

split by bringing space back into the self. The same idea can be found in Johnson’s (1985) book The body in the mind. The word ‘‘in’’ reflects a conceptual attempt to transcend the boundaries of the traditional Cartesian self. Similarly, the term ‘‘I-position’’ is devised as a spatial term suggesting that space is in the mind and not only outside. Moreover, a position is always in relation or opposition to something else. A position never exists separately from an opposite position. Just as a particular meaning can only be defined by its opposite, as Rychlak (1988) has so eloquently argued, a position can only be properly understood by taking its counterposition(s) into account. When I am in contact with another person, the other person is an opposite position reconstructed as part of my extended self. Both my own place (here) and the place of the other (there) function as positions in a real or imaginal space. Moreover, as Harre´ and Van Langenhove (1991) have argued, the verbs positioning and repositioning have, in contrast to the more static term role, the theoretical advantage that they are sensitive to the negotiations, contradictions, agreements, and disagreements among and within people which suppose an iterative moving to and fro between divergent or opposite stances in an extended field of positions. The preceding considerations have immediate consequences for the way we conceive of the relationship between self and emotion. There are two ways of considering this relationship. One is to start from a particular conception of emotions and then examine its implication for the domain of the self. Another is to start from a particular conception of the self and explore its implications for the functioning of emotions. The present contribution is based on the second option. At the beginning of this chapter we argued that the capacity of self-reflection, together with the I–world relationship, deeply influences our (secondary) affect. The succeeding discussions of the multivoiced, dialogical self as elaborated in valuation theory again emphasize the implications that a conception of the self has for affective processes. As illustrated by Sharon’s case, different voices carry different affective structures (or profiles). Instead of a single-voiced investigation, Sharon did a multivoiced investigation with the result that not only different and contrasting valuations were found between positions, but also different and contrasting affective profiles. One and the same person can experience opposite feelings depending on the position evoked in a self-investigation. A multivoiced view on the self involves someone who is able to shift to and fro between positions and to experience different affects accordingly. This view implies that there is not one center from which feelings are experienced, but more centers or positions from which feelings are structured and contrasted and from which they receive their specific meaning.

140


As part of his plea for a narrative conception of the mind, Bruner (1986, 1990) holds that the self is ‘‘distributed.’’ Thinking of the self this way means we are forced to dissociate ourselves from a monolytical, centralized, and unified conception of the self, which has for so long dominated the psychology of the self, both before and after James. The multivoiced approach implies that attributing an isolated emotion (e.g. anxiety) as a trait or state to a particular person is inadequate for understanding affective processes. Such attributions should be rephrased as part of the dynamic interplay between and within positions. An emotion or emotional process receives its specific meaning in the context of other emotions within a particular position, and, moreover, in the context of the emotions of other positions (see the dynamic and meaningful contrasts between the feelings within and between Positions A and B for Sharon). Emotions receive their subjective meaning as part of a dynamic process of positioning and repositioning and this process arouses different, contrasting, and contradicting feelings, not only at different moments but even simultaneously. It is our thesis that this requires a complex and dynamic view of the self and, as a consequence, of the process of emotion. Note 1 We thank Sue Houston for her editorial remarks. In this chapter Hubert Hermans as a personality psychologist cooperated with Els Hermans-Jansen as a psychotherapist. The therapeutic assistance to the client of the case study was provided by the second author.

COM M E N T A R Y

Affective processes in a multivoiced self in action Leni Verhofstadt-Dene`ve

Introduction The theoretical framework of Hermans and Hermans-Jansen, including the so-called valuation theory, is grounded in James’ well-known I–Me concept of the self with the I as the reflecting ‘‘self-as-knower’’ and the Me as the reflected ‘‘self-as-known.’’ An important aspect of this view is that the Me is much wider than the mere result of reflection on the self (in its narrow sense); it also contains the whole social and material world which has a certain personal value and affective significance for the person. This view implies an I–world connection with a strong and complex relational and affective component in the self-reflecting person in present, past, and future situations. Hermans and HermansJansen: ‘‘With its reflectivity and I–world connection, the self evokes complexities of similar, different, and contrasting feelings, which together form highly dynamic structures’’ (p. 122). An original turn (inspired by Sarbin) is the interpretation of the self as a sort of narrative psychodramatic system in which the I and the Me function as ‘‘author’’ and ‘‘observed actor’’ respectively. The self as author (the I) constructs an imaginary story in which the Me functions as (1) the protagonist, and (2) other figures that are part of Mine as antagonists (or imaginal others e.g. father, sister, partner). I presume that Hermans and Hermans-Jansen do not restrict these antagonists to significant others but include the whole social and material world to which the I attributes a personal affective value. From this point of view the I constructs a multivoiced world in which the individual is not only able to talk about the variety of imaginal others (antagonists) but also to talk with them as relatively independent parts of an extended self. ‘‘In other words, imaginal figures function as voiced positions in a selfspace, and, at times, these positions entertain dialogical relationships with each other’’ (p. 126). It is interesting to note that this valuation theory functions as a basis 141

142


for the Self-Confrontation Method (SCM), an idiographic procedure which is constructed ‘‘to study the relation between valuations and types of affect and the way in which these variables become organized into one or more systems’’ (p. 130). The procedure involves elicitation of a set of valuations and subsequent association of these valuations with a standardized set of affect items. The valuation theory is highly attractive, as it is based on the organized but complex subjective phenomenological cognitive–affective narratives of the person. In the elicitation of a set of valuations the person has the opportunity to take diverse (also contradictory) I-positions, which over time can develop ‘‘into a more integrative Gestalt.’’ However, one can question whether or not the SCM remains on a merely cognitive, rational self-reflective level. It mainly stresses a reflection on cognitions and emotions from self and others and the dialogues between those constructions. Even when in the SCM the focus is explicitly on affects, what remains is mainly a reflection on affects and not necessarily an intense involvement with those affects. The ‘‘psychodrama’’ in the SCM exclusively works on an internal level with a strong cognitive–rational accentuation. A similar limitation concerns the important aspect of conflict experience. For example: ‘‘Sharon did her self-investigation not from one centralized position . . . Rather we approached Sharon as a multivoiced individual, able to reflect on her self from two opposite positions and tell about significant meaning units from the perspective of these positions’’ (p. 137) Real conflict experience is not at stake here, but the reflection on and dialogue with different I-positions seems again crucial for personality integration. This highly cognitive accent (which no doubt has value of its own) could effectively be enriched with cognitive–affective components by means of the introduction of real action and drama techniques, in which the protagonist not only reflects on, or engages in dialogue with, but also becomes diverse antagonists or imaginal others in a specific time and space. In the following comments I will try to defend this thesis by first explaining the importance of conflict for personality development and secondly by illustrating how, in association with the SCM, different sorts of conflicts could be worked with through the introduction of action and drama techniques. The role of conflict in personality and personality development Although Hermans and Hermans-Jansen do pay attention to the possibility of conflicts between different parts of the person, between differ-

Commentary on Hermans and Hermans-Jansen

143

ent I–Me positions, they mainly treat them from a clinical perspective, with a strong rational–cognitive accent. In this view conflict is not seen as the driving force in psychological development. In developmental theories such as that of Piaget (1949), conflict is seen as an essential characteristic in the developmental process. Conflicts within the ‘‘multivoiced self’’ can drive the individual’s development. Oppositions and conflict, from this perspective, can be considered a necessary (but not sufficient) condition for development. Such an interpretation is underpinned by my own follow-up research, therapeutic practice, and by theoretical analyses on the basis of dialectical developmental psychology (Brown, Werner, and Altman, 1998; Conville, 1998; Verhofstadt-Dene`ve, 1985; Verhofstadt-Dene`ve and Schittekatte, 1996). In this view development is seen as a self-organizing process, consisting in dialectical developmental movements, which are inherent in the qualitative development of each living organism (see also Piaget, 1968; Prigogine and Nicolis, 1977). Or in the words of Piaget, referring to the dialectical course of cognitive development: ‘‘c’est une de´marche ine´vitable de la penseé’’ (p. 104). A comparable view is to be found in the rather rigid epigenetic model of development as described by Erikson (1968): every stage of life is built on a crisis, the content of which is characterized by a positive and a negative pole. There is a strong resemblance here to Kelly’s personality theory and its bipolar constructs as the cornerstones of an active and constantly evolving system of constructs. The experiencing of oppositions causes the entire system of constructs to shift, so that change, activity and personality development can take place (Kelly, 1955). An explicitly dialectical bias is also inherent in Vygotsky’s approach (1979; Zender and Zender, 1974), since he looks upon crises and conflicts as criteria marking off the stages of a dialectical development. It is worth noting that in Vygotsky’s view the crisis (or negative phase) in the child’s psychological development acquires a pronounced positive meaning. The basic concept of contradiction and conflict as stimuli for developmental processes is related to Kohlberg’s (1976) notion of cognitive conflict as the stimulus of moral development. His way of thinking was supported by Lawrence Walker’s (1983) experiments. Walker found that a range of conditions which trigger cognitive conflicts stimulate the transition from one stage to the next in the moral development of children. In my theoretical framework dialectical movements and self-organization determine respectively the process and direction of qualitative developments.

144


Conflicts between what? The idea that conflicts drive development has been further elaborated in my phenomenological–dialectical model of personality (Verhofstadt-Dene`ve, 1988, 1995, 1997, 1999, 2000). In this model the phenomenological aspect refers to the unique subjective content which every human being attributes to him- or herself and to the surrounding world and is also based on the William James I–Me paradigm. The dialectical aspect refers to the underlying process of interaction and conflict between these contents, and their development. Among the multitude of I–Me personality constructions six main domains or images (comparable with fundamental I-positions in Hermans and Hermans-Jansen) are focused on, each corresponding with a specific question: The Self-Image: ‘‘Who am I, with my potential and my shortcomings?’’ The Alter-Image: ‘‘What are the others like?’’ The Meta-Self: ‘‘How is one’s own person viewed by the others?’’ These three basic images are counterbalanced by three ideal-images: The Ideal-Self: ‘‘Who would I like to be or become?’’ The Ideal-Alter: ‘‘What should the others be or become like?’’ The Ideal-Meta-Self: ‘‘How should the others perceive me?’’ For each image a distinction can be made between an external aspect (the way we behave externally) and an internal aspect (what we think and feel at the same time). In this model the person can also be seen as a ‘‘multivoiced self’’ with six main parts which can interact and be in conflict. On the basis of the model, many types of conflict can be distinguished according to the nature of the opposing poles. I shall just give some examples: – There may be inter-dimensional oppositions, for example between the Self-Image and the Meta-Self. – Similarly, there may exist intra-dimensional oppositions, for example between external and internal contents of one dimension. – Oppositions may be experienced between the subjective phenomenological constructions of oneself and of the world, on the one hand, and possible alternative interpretations, on the other. – Vivid oppositions can also be felt between the three time dimensions. Within the scope of these comments I will, in the presentation of action and drama techniques, confine myself to the first and the second kinds


145

of oppositions. For more information I refer to Verhofstadt-Dene`ve (1995, 1997, 1999, 2000). How to stimulate conflicts and development? As Hermans and Hermans-Jansen have shown, the use of the SCM can trigger self-organizing processes in the ‘‘multivoiced self’’ by means of the stimulation of cognitions, emotions, and language. This happens on an internal level, but from my clinical experience and didactic group work with my students I have learned that direct spatial, external action in the context of an intense and supportive group situation can lend vital support to such internal self-organizing processes. To me it seems that the SCM is too exclusively aimed at an internal, mental action level: the ‘‘protagonist’’ has to construct his or her personal life story purely in the imagination. Such internal action can be effectively supported and will receive more impetus by the application of action and dramatechniques, within a supportive audience. Psychodrama can enable the ‘‘protagonist’’ to literally step into his or her own personal universe. Following Bruner’s line of thought on the distinction between the narrative and logico-scientific mode (in Hermans and HermansJansen), one could say that both in the SCM and in psychodrama the emphasis is mainly on the ‘‘narrative mode,’’ although the ‘‘logicoscientific mode’’ is not absent. Still, psychodrama (by literally staging a specific and concrete situation) probably meets Bruner’s requirement of the narrative approach more explicitly: the narrative mode puts the generalities of the human condition into the particulars of experience, and attempts to locate experience in time and space (Hermans and Hermans-Jansen: 123). In order to facilitate internal, dialectical self-organizing processes as well as the individual’s personality development, therefore, a structured application of action and drama techniques is suggested. Action techniques, however, are much more than simple action tricks. They must be grounded in an appropriate theory of personality and personality development. It is for this reason that the phenomenological– dialectical personality model (see above) was developed. Conflicts and development: the application of action and drama techniques There are various theoretical frameworks and methods for activating the I–Me reflection. The SCM on the basis of the Valuation Theory, on the one hand, and action and drama techniques based on the developmental psychotherapeutic view outlined above, on the other, are both valuable methods. The latter, due to its action-orientated character and

146


by its direct affective-situational impact, seems best suited to trigger intensive affective processes in a multivoiced self. A short practical description based on a variant of the psychodrama method as described by Moreno (Moreno, 1946, 1959; Moreno and Moreno, 1969; Verhofstadt-Dene`ve, 1999) and Elefthery (Moreno and Elefthery, 1982) is given below. Case presentation Kevin (K) is a seventeen-year-old boy, the eldest of five children, with poor results at school; he finds it difficult to make friends and is rather hot-tempered. His father is an alcoholic and unemployed; he behaves aggressively towards his children and particularly towards his wife, and there are many conflicts between father and K. Mother cannot cope with her family and relies on K to support her. Specific event: father comes home drunk and wants his wife to give him money. He bullies her and K gives his father a push so that he has a nasty fall. The father turns out to be permanently paralyzed and will never be able to walk again. After this event, K becomes very withdrawn; he leaves home from time to time, plays truant and wanders around aimlessly; he even makes an attempt at suicide. His father is rebellious but has stopped drinking. The mother takes care of her husband to the best of her abilities. The two parents suffer from K’s behavior.

One major task of the director of a psychodrama (Dir.) is to create an atmosphere providing the greatest possible feeling of security and unconditional respect for all group members. The idea is that through an in-depth I–Me reflection participants themselves should (as in the SCM) find solutions to their own problems. This will be illustrated by means of some examples from sessions where K was the protagonist in a group of eight adolescents. The case study shows how, during the psychodrama session, conflicting I–Me constructions appear, and how this conflict results in psychological change. Example 1 (inter-dimensional): Self-Image versus Alter-Image/Meta-Self K shows clear signs of depression. As mentioned above, he has an extremely negative self-evaluation and strong feelings of guilt ( = SelfImage). Before action techniques are applied, the protagonist has to create for himself a concrete familiar situation. Dir.: Kevin, you’re going to see your father now. Where do you want to meet him? What time is it? What are you going to do? K: He’s in the kitchen in his wheelchair, it’s 7 p.m. . . . Dir.: OK, tell us what the kitchen looks like . . . Where’s the door, where’s the worktop? (K describes the kitchen, and a few chairs and a table are brought in. These simple objects considerably enhance Kevin’s affective involvement in this specific situation.)


147

Dir.: I know that it’s difficult for you to meet your father, but here you can safely give it a try, you will not be alone, we will all help you. If you like you can choose somebody from the group to represent your father (‘‘Antagonist’’). K: OK, I take Bob, because he is the tallest of the group . . . (Bob enters the inner group-space). Dir.: OK, Kevin, come here and stand behind Bob. You will now try to become your father and tell who you are, using the I-form ( = roletaking, cf. Mead: ‘‘taking the role of the other’’). Take your time, Kevin . . . Bob, you will listen very carefully so that you know how to represent Kevin’s father. K: I am Kevin’s father. I am thirty-six. I am paralyzed, tired, I’ve lost all my strength and vitality. I watch TV all day long, I feel angry and sad. The only reason I go on living is my youngest son, Jerry, he’s the only one left who loves me . . . ( = a part of Kevin’s Alter-Image: his father and brother).

This Alter-Image is comparable to Hermans and Hermans-Jansen’s I-position in a multivoiced self within an imaginal landscape. We now proceed to Kevin’s Meta-Self (cf. Hermans and Hermans-Jansen: another ‘‘I-position’’): Dir. (addressing Kevin in the role of his father): Well, father of Kevin, what do you think about your eldest son? K (still playing the role of his own father): He is the cause of all this misery . . . he beat me . . . I might have been dead . . . I wish I was . . . As a child he was such a lovely kid . . . But how much he has changed! . . . Yet, I think I still love him . . . ( = Kevin’s Meta-Self). (Kevin is finding it hard to cope and starts crying softly . . . ) Dir.: OK, Kevin, take your time . . . Come here, you can now be yourself again, you are no longer your father . . . Who are you now? K: I’m Kevin, I had a difficult moment just now. I don’t know why, I don’t usually cry, but it came so suddenly! ( = again Kevin’s SelfImage).

The former actions are an example of a dialectical process, namely a double negation movement following a three-stage course. The first stage (or thesis) had Kevin’s Self-Image as its starting point. Through the role taking in the second stage (or anti-thesis) the Self-Image is disregarded and the Alter-Image and the Meta-Self are focused on. This is the movement of the first negation (or the negation of the Self-Image). K becomes his own Alter-Image and Meta-Self, though historically the Self-Image is still present. This vividly incites the cognitive–affective opposition between Self-Image and Alter-Image/Meta-Self. This strong experience of opposition often goes together with a crisis or a catharsis. This was indeed the case when K, in the role of the father, felt his feelings for the father and the pain and the love of the father at the same time. In the third stage (or synthesis), through the second negation (or

148


the negation of the Alter-Image and Meta-Self), the protagonist returns to his own Self-Image. In psychodramatical practice this process is not at all a rigid threestage movement. Mostly there is a repeated toing and froing between the different poles, which can lead to successive partial synthesis stages. For Kevin, after this dialectical movement, his Self-Image, his AlterImage, and his Meta-Self had changed mainly through the experience of conflict in connection with the first negation. In the next session he said to the group members: ‘‘The guilt feelings are lessened and the image of my father is not so negative anymore; I can now think about him quietly without being completely upset.’’ What is important in this process is that the experiencing of a conflict incited by the first negation can stimulate change and integration of the two opposite poles in the synthesis stage, or at least open them up to a more intense I–Me reflection. The above example describes the experience of oppositions between images (or I-positions) of the person. The following example illustrates oppositions within one image. It still concerns Kevin’s relationship with his father. This illustrates how the opposition between internal and external Self-Images can be used. Example 2: External versus internal Self-Image (The following scene is a dialogue in the kitchen.) K: Pa, I’m off for a walk. Bob (as Kevin’s father): All right son, but be sure to be back before midnight. Dir. (to K): Could your father say that? K: No, my father would never say that! Dir.: OK, reverse roles! Kevin, you’ll be your own father now. Bob, you’ll be Kevin, and you repeat the last few words he has just said.

Role reversal is necessary at this stage because we are proceeding on K’s subjective phenomenological constructions. The dialogue therefore must fit perfectly with K’s usual environment, for if it does not we would have a kind of role play with a much weaker affective and cognitive impact. The I constructs a ‘‘multivoiced world’’ but within psychodrama this construction is stimulated by action within a temporal and spatial, affectively involved situation. Bob (as K): Pa, I’m off for a walk. K (as his father): Yes, do, and get out of sight for the rest of the week as well! Dir.: Role reversal once again! K (as himself, addressing his father): Oh shut up! Look at yourself!


149

What did you do with your life . . . Misfit! Dir. (to K): What are you thinking of now? What do you feel? Go one step to the left and try to say what is going on in your mind, your father will not hear you now. (K now has the possibility of expressing his internal part acting from a new I-position, as if another person was speaking from a different space.) K (thinking aloud): I feel bad, guilty . . . Why am I saying all this? I know he feels miserable . . . and strangely enough I feel pity for him . . . Sometimes I think I still love him . . . Is this the little boy in me?

For adolescents (and not only for adolescents) it can be much easier to express harsh and insolent behavior to significant others than affective approach attitudes (of course, the reverse can also be true). By freezing the external action (cf. ‘‘exterior dialogue’’) and stimulating the protagonist to reflect on the simultaneous internal affects and thoughts (cf. ‘‘internal dialogue’’) the protagonist can vividly experience the divergent poles. Comparable with an intra-psychic role reversal he can move from the exterior to the interior following the same three-stage process back and forth. Through this dialectical movement he can discover that both have sense and can enrich each other. K, for example, discovered that his harsh behavior towards his father was grounded in pain, guilt, anxiety, and love. Action and drama sessions are meant to offer an opportunity to express such hidden internal contents and to actively work with them. This way of working is perfectly in tune with Hermans and Hermans-Jansen’s reference to Watkins’ ‘‘Invisible Guests’’ and Bakhtin’s ‘‘interior and exterior’’ dialogues referring to Dostoyevsky’s novel The Double. ‘‘The second hero (the double) was introduced as a personification of the interior thought of the first hero (Golyadkin). This externalization of an interior thought in a spatially separated opponent instigates a fully developed dialogue between two relatively independent parties’’(ibid., p. 125). This was exactly what was put into practice with K. Conclusion The notion that conflicts drive development is inherent in the phenomological–dialectical personality model. Action and drama techniques as applied within this framework are certainly not a panacea. Practice does show, however, that thanks to the combination of speaking, thinking, feeling, and acting, this method offers a very strong stimulus within the complex process of self-actualization and personality development. Both the self-confrontation method and psychodrama techniques

150


can produce a picture of an individual’s personality structure and both can also engender personality development. The accent of the two methods seems different, however. The self-confrontation method probably encourages a more systematic survey of the personality structure at different developmental moments, whereas psychodrama, thanks to its strong and specific affective-relational, emotional, and cognitive involvement, probably has a more direct impact on personality development. Through the construction of a concrete space and time dimension within psychodrama, an extra relational accent can be created. The protagonist has the opportunity to objectify and exteriorize his or her image of the antagonist. This encounter intensifies and surpasses the imaginary self-reflective dimension. Thanks to this intense physical and mental action the protagonist is able to experience the self and significant others as part of a multivoiced self not only mentally but also to meet and be those significant others in vivid cognitive–emotional situations in different times and spaces. Through the process of becoming the other, the protagonist feels the differences and disharmonies intensively but also becomes aware of similarities and harmony between Self and Other. Through action, personal identity is given an explicit relational component. All this means that the internal psychodrama activated in the protagonist by the SCM, with the I as author and the Me as actor, can be strongly affectively supported and intensified if the person as a multivoiced self could effectively move from one I-position to the other, in order to really meet and become the antagonists in a concrete time and space experience. A combination of the Self-Confrontation Method and psychodrama therefore offers considerable promise for the future.

COM M E N T A R Y

Affective processes in a multivoiced self in action Leni Verhofstadt-Dene`ve

Introduction The theoretical framework of Hermans and Hermans-Jansen, including the so-called valuation theory, is grounded in James’ well-known I–Me concept of the self with the I as the reflecting ‘‘self-as-knower’’ and the Me as the reflected ‘‘self-as-known.’’ An important aspect of this view is that the Me is much wider than the mere result of reflection on the self (in its narrow sense); it also contains the whole social and material world which has a certain personal value and affective significance for the person. This view implies an I–world connection with a strong and complex relational and affective component in the self-reflecting person in present, past, and future situations. Hermans and HermansJansen: ‘‘With its reflectivity and I–world connection, the self evokes complexities of similar, different, and contrasting feelings, which together form highly dynamic structures’’ (p. 122). An original turn (inspired by Sarbin) is the interpretation of the self as a sort of narrative psychodramatic system in which the I and the Me function as ‘‘author’’ and ‘‘observed actor’’ respectively. The self as author (the I) constructs an imaginary story in which the Me functions as (1) the protagonist, and (2) other figures that are part of Mine as antagonists (or imaginal others e.g. father, sister, partner). I presume that Hermans and Hermans-Jansen do not restrict these antagonists to significant others but include the whole social and material world to which the I attributes a personal affective value. From this point of view the I constructs a multivoiced world in which the individual is not only able to talk about the variety of imaginal others (antagonists) but also to talk with them as relatively independent parts of an extended self. ‘‘In other words, imaginal figures function as voiced positions in a selfspace, and, at times, these positions entertain dialogical relationships with each other’’ (p. 126). It is interesting to note that this valuation theory functions as a basis 141

142


for the Self-Confrontation Method (SCM), an idiographic procedure which is constructed ‘‘to study the relation between valuations and types of affect and the way in which these variables become organized into one or more systems’’ (p. 130). The procedure involves elicitation of a set of valuations and subsequent association of these valuations with a standardized set of affect items. The valuation theory is highly attractive, as it is based on the organized but complex subjective phenomenological cognitive–affective narratives of the person. In the elicitation of a set of valuations the person has the opportunity to take diverse (also contradictory) I-positions, which over time can develop ‘‘into a more integrative Gestalt.’’ However, one can question whether or not the SCM remains on a merely cognitive, rational self-reflective level. It mainly stresses a reflection on cognitions and emotions from self and others and the dialogues between those constructions. Even when in the SCM the focus is explicitly on affects, what remains is mainly a reflection on affects and not necessarily an intense involvement with those affects. The ‘‘psychodrama’’ in the SCM exclusively works on an internal level with a strong cognitive–rational accentuation. A similar limitation concerns the important aspect of conflict experience. For example: ‘‘Sharon did her self-investigation not from one centralized position . . . Rather we approached Sharon as a multivoiced individual, able to reflect on her self from two opposite positions and tell about significant meaning units from the perspective of these positions’’ (p. 137) Real conflict experience is not at stake here, but the reflection on and dialogue with different I-positions seems again crucial for personality integration. This highly cognitive accent (which no doubt has value of its own) could effectively be enriched with cognitive–affective components by means of the introduction of real action and drama techniques, in which the protagonist not only reflects on, or engages in dialogue with, but also becomes diverse antagonists or imaginal others in a specific time and space. In the following comments I will try to defend this thesis by first explaining the importance of conflict for personality development and secondly by illustrating how, in association with the SCM, different sorts of conflicts could be worked with through the introduction of action and drama techniques. The role of conflict in personality and personality development Although Hermans and Hermans-Jansen do pay attention to the possibility of conflicts between different parts of the person, between differ-


143

ent I–Me positions, they mainly treat them from a clinical perspective, with a strong rational–cognitive accent. In this view conflict is not seen as the driving force in psychological development. In developmental theories such as that of Piaget (1949), conflict is seen as an essential characteristic in the developmental process. Conflicts within the ‘‘multivoiced self’’ can drive the individual’s development. Oppositions and conflict, from this perspective, can be considered a necessary (but not sufficient) condition for development. Such an interpretation is underpinned by my own follow-up research, therapeutic practice, and by theoretical analyses on the basis of dialectical developmental psychology (Brown, Werner, and Altman, 1998; Conville, 1998; Verhofstadt-Dene`ve, 1985; Verhofstadt-Dene`ve and Schittekatte, 1996). In this view development is seen as a self-organizing process, consisting in dialectical developmental movements, which are inherent in the qualitative development of each living organism (see also Piaget, 1968; Prigogine and Nicolis, 1977). Or in the words of Piaget, referring to the dialectical course of cognitive development: ‘‘c’est une de´marche ine´vitable de la penseé’’ (p. 104). A comparable view is to be found in the rather rigid epigenetic model of development as described by Erikson (1968): every stage of life is built on a crisis, the content of which is characterized by a positive and a negative pole. There is a strong resemblance here to Kelly’s personality theory and its bipolar constructs as the cornerstones of an active and constantly evolving system of constructs. The experiencing of oppositions causes the entire system of constructs to shift, so that change, activity and personality development can take place (Kelly, 1955). An explicitly dialectical bias is also inherent in Vygotsky’s approach (1979; Zender and Zender, 1974), since he looks upon crises and conflicts as criteria marking off the stages of a dialectical development. It is worth noting that in Vygotsky’s view the crisis (or negative phase) in the child’s psychological development acquires a pronounced positive meaning. The basic concept of contradiction and conflict as stimuli for developmental processes is related to Kohlberg’s (1976) notion of cognitive conflict as the stimulus of moral development. His way of thinking was supported by Lawrence Walker’s (1983) experiments. Walker found that a range of conditions which trigger cognitive conflicts stimulate the transition from one stage to the next in the moral development of children. In my theoretical framework dialectical movements and self-organization determine respectively the process and direction of qualitative developments.

144


Conflicts between what? The idea that conflicts drive development has been further elaborated in my phenomenological–dialectical model of personality (Verhofstadt-Dene`ve, 1988, 1995, 1997, 1999, 2000). In this model the phenomenological aspect refers to the unique subjective content which every human being attributes to him- or herself and to the surrounding world and is also based on the William James I–Me paradigm. The dialectical aspect refers to the underlying process of interaction and conflict between these contents, and their development. Among the multitude of I–Me personality constructions six main domains or images (comparable with fundamental I-positions in Hermans and Hermans-Jansen) are focused on, each corresponding with a specific question: The Self-Image: ‘‘Who am I, with my potential and my shortcomings?’’ The Alter-Image: ‘‘What are the others like?’’ The Meta-Self: ‘‘How is one’s own person viewed by the others?’’ These three basic images are counterbalanced by three ideal-images: The Ideal-Self: ‘‘Who would I like to be or become?’’ The Ideal-Alter: ‘‘What should the others be or become like?’’ The Ideal-Meta-Self: ‘‘How should the others perceive me?’’ For each image a distinction can be made between an external aspect (the way we behave externally) and an internal aspect (what we think and feel at the same time). In this model the person can also be seen as a ‘‘multivoiced self’’ with six main parts which can interact and be in conflict. On the basis of the model, many types of conflict can be distinguished according to the nature of the opposing poles. I shall just give some examples: – There may be inter-dimensional oppositions, for example between the Self-Image and the Meta-Self. – Similarly, there may exist intra-dimensional oppositions, for example between external and internal contents of one dimension. – Oppositions may be experienced between the subjective phenomenological constructions of oneself and of the world, on the one hand, and possible alternative interpretations, on the other. – Vivid oppositions can also be felt between the three time dimensions. Within the scope of these comments I will, in the presentation of action and drama techniques, confine myself to the first and the second kinds


145

of oppositions. For more information I refer to Verhofstadt-Dene`ve (1995, 1997, 1999, 2000). How to stimulate conflicts and development? As Hermans and Hermans-Jansen have shown, the use of the SCM can trigger self-organizing processes in the ‘‘multivoiced self’’ by means of the stimulation of cognitions, emotions, and language. This happens on an internal level, but from my clinical experience and didactic group work with my students I have learned that direct spatial, external action in the context of an intense and supportive group situation can lend vital support to such internal self-organizing processes. To me it seems that the SCM is too exclusively aimed at an internal, mental action level: the ‘‘protagonist’’ has to construct his or her personal life story purely in the imagination. Such internal action can be effectively supported and will receive more impetus by the application of action and dramatechniques, within a supportive audience. Psychodrama can enable the ‘‘protagonist’’ to literally step into his or her own personal universe. Following Bruner’s line of thought on the distinction between the narrative and logico-scientific mode (in Hermans and HermansJansen), one could say that both in the SCM and in psychodrama the emphasis is mainly on the ‘‘narrative mode,’’ although the ‘‘logicoscientific mode’’ is not absent. Still, psychodrama (by literally staging a specific and concrete situation) probably meets Bruner’s requirement of the narrative approach more explicitly: the narrative mode puts the generalities of the human condition into the particulars of experience, and attempts to locate experience in time and space (Hermans and Hermans-Jansen: 123). In order to facilitate internal, dialectical self-organizing processes as well as the individual’s personality development, therefore, a structured application of action and drama techniques is suggested. Action techniques, however, are much more than simple action tricks. They must be grounded in an appropriate theory of personality and personality development. It is for this reason that the phenomenological– dialectical personality model (see above) was developed. Conflicts and development: the application of action and drama techniques There are various theoretical frameworks and methods for activating the I–Me reflection. The SCM on the basis of the Valuation Theory, on the one hand, and action and drama techniques based on the developmental psychotherapeutic view outlined above, on the other, are both valuable methods. The latter, due to its action-orientated character and

146


by its direct affective-situational impact, seems best suited to trigger intensive affective processes in a multivoiced self. A short practical description based on a variant of the psychodrama method as described by Moreno (Moreno, 1946, 1959; Moreno and Moreno, 1969; Verhofstadt-Dene`ve, 1999) and Elefthery (Moreno and Elefthery, 1982) is given below. Case presentation Kevin (K) is a seventeen-year-old boy, the eldest of five children, with poor results at school; he finds it difficult to make friends and is rather hot-tempered. His father is an alcoholic and unemployed; he behaves aggressively towards his children and particularly towards his wife, and there are many conflicts between father and K. Mother cannot cope with her family and relies on K to support her. Specific event: father comes home drunk and wants his wife to give him money. He bullies her and K gives his father a push so that he has a nasty fall. The father turns out to be permanently paralyzed and will never be able to walk again. After this event, K becomes very withdrawn; he leaves home from time to time, plays truant and wanders around aimlessly; he even makes an attempt at suicide. His father is rebellious but has stopped drinking. The mother takes care of her husband to the best of her abilities. The two parents suffer from K’s behavior.

One major task of the director of a psychodrama (Dir.) is to create an atmosphere providing the greatest possible feeling of security and unconditional respect for all group members. The idea is that through an in-depth I–Me reflection participants themselves should (as in the SCM) find solutions to their own problems. This will be illustrated by means of some examples from sessions where K was the protagonist in a group of eight adolescents. The case study shows how, during the psychodrama session, conflicting I–Me constructions appear, and how this conflict results in psychological change. Example 1 (inter-dimensional): Self-Image versus Alter-Image/Meta-Self K shows clear signs of depression. As mentioned above, he has an extremely negative self-evaluation and strong feelings of guilt ( = SelfImage). Before action techniques are applied, the protagonist has to create for himself a concrete familiar situation. Dir.: Kevin, you’re going to see your father now. Where do you want to meet him? What time is it? What are you going to do? K: He’s in the kitchen in his wheelchair, it’s 7 p.m. . . . Dir.: OK, tell us what the kitchen looks like . . . Where’s the door, where’s the worktop? (K describes the kitchen, and a few chairs and a table are brought in. These simple objects considerably enhance Kevin’s affective involvement in this specific situation.)


147

Dir.: I know that it’s difficult for you to meet your father, but here you can safely give it a try, you will not be alone, we will all help you. If you like you can choose somebody from the group to represent your father (‘‘Antagonist’’). K: OK, I take Bob, because he is the tallest of the group . . . (Bob enters the inner group-space). Dir.: OK, Kevin, come here and stand behind Bob. You will now try to become your father and tell who you are, using the I-form ( = roletaking, cf. Mead: ‘‘taking the role of the other’’). Take your time, Kevin . . . Bob, you will listen very carefully so that you know how to represent Kevin’s father. K: I am Kevin’s father. I am thirty-six. I am paralyzed, tired, I’ve lost all my strength and vitality. I watch TV all day long, I feel angry and sad. The only reason I go on living is my youngest son, Jerry, he’s the only one left who loves me . . . ( = a part of Kevin’s Alter-Image: his father and brother).

This Alter-Image is comparable to Hermans and Hermans-Jansen’s I-position in a multivoiced self within an imaginal landscape. We now proceed to Kevin’s Meta-Self (cf. Hermans and Hermans-Jansen: another ‘‘I-position’’): Dir. (addressing Kevin in the role of his father): Well, father of Kevin, what do you think about your eldest son? K (still playing the role of his own father): He is the cause of all this misery . . . he beat me . . . I might have been dead . . . I wish I was . . . As a child he was such a lovely kid . . . But how much he has changed! . . . Yet, I think I still love him . . . ( = Kevin’s Meta-Self). (Kevin is finding it hard to cope and starts crying softly . . . ) Dir.: OK, Kevin, take your time . . . Come here, you can now be yourself again, you are no longer your father . . . Who are you now? K: I’m Kevin, I had a difficult moment just now. I don’t know why, I don’t usually cry, but it came so suddenly! ( = again Kevin’s SelfImage).

The former actions are an example of a dialectical process, namely a double negation movement following a three-stage course. The first stage (or thesis) had Kevin’s Self-Image as its starting point. Through the role taking in the second stage (or anti-thesis) the Self-Image is disregarded and the Alter-Image and the Meta-Self are focused on. This is the movement of the first negation (or the negation of the Self-Image). K becomes his own Alter-Image and Meta-Self, though historically the Self-Image is still present. This vividly incites the cognitive–affective opposition between Self-Image and Alter-Image/Meta-Self. This strong experience of opposition often goes together with a crisis or a catharsis. This was indeed the case when K, in the role of the father, felt his feelings for the father and the pain and the love of the father at the same time. In the third stage (or synthesis), through the second negation (or

148


the negation of the Alter-Image and Meta-Self), the protagonist returns to his own Self-Image. In psychodramatical practice this process is not at all a rigid threestage movement. Mostly there is a repeated toing and froing between the different poles, which can lead to successive partial synthesis stages. For Kevin, after this dialectical movement, his Self-Image, his AlterImage, and his Meta-Self had changed mainly through the experience of conflict in connection with the first negation. In the next session he said to the group members: ‘‘The guilt feelings are lessened and the image of my father is not so negative anymore; I can now think about him quietly without being completely upset.’’ What is important in this process is that the experiencing of a conflict incited by the first negation can stimulate change and integration of the two opposite poles in the synthesis stage, or at least open them up to a more intense I–Me reflection. The above example describes the experience of oppositions between images (or I-positions) of the person. The following example illustrates oppositions within one image. It still concerns Kevin’s relationship with his father. This illustrates how the opposition between internal and external Self-Images can be used. Example 2: External versus internal Self-Image (The following scene is a dialogue in the kitchen.) K: Pa, I’m off for a walk. Bob (as Kevin’s father): All right son, but be sure to be back before midnight. Dir. (to K): Could your father say that? K: No, my father would never say that! Dir.: OK, reverse roles! Kevin, you’ll be your own father now. Bob, you’ll be Kevin, and you repeat the last few words he has just said.

Role reversal is necessary at this stage because we are proceeding on K’s subjective phenomenological constructions. The dialogue therefore must fit perfectly with K’s usual environment, for if it does not we would have a kind of role play with a much weaker affective and cognitive impact. The I constructs a ‘‘multivoiced world’’ but within psychodrama this construction is stimulated by action within a temporal and spatial, affectively involved situation. Bob (as K): Pa, I’m off for a walk. K (as his father): Yes, do, and get out of sight for the rest of the week as well! Dir.: Role reversal once again! K (as himself, addressing his father): Oh shut up! Look at yourself!


149

What did you do with your life . . . Misfit! Dir. (to K): What are you thinking of now? What do you feel? Go one step to the left and try to say what is going on in your mind, your father will not hear you now. (K now has the possibility of expressing his internal part acting from a new I-position, as if another person was speaking from a different space.) K (thinking aloud): I feel bad, guilty . . . Why am I saying all this? I know he feels miserable . . . and strangely enough I feel pity for him . . . Sometimes I think I still love him . . . Is this the little boy in me?

For adolescents (and not only for adolescents) it can be much easier to express harsh and insolent behavior to significant others than affective approach attitudes (of course, the reverse can also be true). By freezing the external action (cf. ‘‘exterior dialogue’’) and stimulating the protagonist to reflect on the simultaneous internal affects and thoughts (cf. ‘‘internal dialogue’’) the protagonist can vividly experience the divergent poles. Comparable with an intra-psychic role reversal he can move from the exterior to the interior following the same three-stage process back and forth. Through this dialectical movement he can discover that both have sense and can enrich each other. K, for example, discovered that his harsh behavior towards his father was grounded in pain, guilt, anxiety, and love. Action and drama sessions are meant to offer an opportunity to express such hidden internal contents and to actively work with them. This way of working is perfectly in tune with Hermans and Hermans-Jansen’s reference to Watkins’ ‘‘Invisible Guests’’ and Bakhtin’s ‘‘interior and exterior’’ dialogues referring to Dostoyevsky’s novel The Double. ‘‘The second hero (the double) was introduced as a personification of the interior thought of the first hero (Golyadkin). This externalization of an interior thought in a spatially separated opponent instigates a fully developed dialogue between two relatively independent parties’’(ibid., p. 125). This was exactly what was put into practice with K. Conclusion The notion that conflicts drive development is inherent in the phenomological–dialectical personality model. Action and drama techniques as applied within this framework are certainly not a panacea. Practice does show, however, that thanks to the combination of speaking, thinking, feeling, and acting, this method offers a very strong stimulus within the complex process of self-actualization and personality development. Both the self-confrontation method and psychodrama techniques

150


can produce a picture of an individual’s personality structure and both can also engender personality development. The accent of the two methods seems different, however. The self-confrontation method probably encourages a more systematic survey of the personality structure at different developmental moments, whereas psychodrama, thanks to its strong and specific affective-relational, emotional, and cognitive involvement, probably has a more direct impact on personality development. Through the construction of a concrete space and time dimension within psychodrama, an extra relational accent can be created. The protagonist has the opportunity to objectify and exteriorize his or her image of the antagonist. This encounter intensifies and surpasses the imaginary self-reflective dimension. Thanks to this intense physical and mental action the protagonist is able to experience the self and significant others as part of a multivoiced self not only mentally but also to meet and be those significant others in vivid cognitive–emotional situations in different times and spaces. Through the process of becoming the other, the protagonist feels the differences and disharmonies intensively but also becomes aware of similarities and harmony between Self and Other. Through action, personal identity is given an explicit relational component. All this means that the internal psychodrama activated in the protagonist by the SCM, with the I as author and the Me as actor, can be strongly affectively supported and intensified if the person as a multivoiced self could effectively move from one I-position to the other, in order to really meet and become the antagonists in a concrete time and space experience. A combination of the Self-Confrontation Method and psychodrama therefore offers considerable promise for the future.

CH A P T E R 7

Old–new answers and new–old questions for personality and emotion: a matter of complexity Jeannette Haviland-Jones, David Boulifard, and Carol Magai1 We are very fortunate to see the beginning of an exciting time in psychology, an era in which we may bring together complex intuitive ideas with appropriate technical tools for expanding them. This may be particularly interesting in the area of emotion and personality development. The present authors have described changes in theories of identity and emotion from dichotomous sets to systems of sets across the twentieth century (Haviland and Kahlbaugh, 1993) and have also examined the impact of emotion on personality development (Magai and McFadden, 1995). Here we intend to take a further step in this direction with some analyses of the development of identity, or selfconcept, in two individuals as these are interwoven with emotion. With a dynamic mathematical model of emotion we will also show how emotional processes may be variable without the necessity for variable cognitive content in the analysis of the stimulation that led to an emotional change. We will use our examples to show how different contexts at the origin of a fearful stimulus might lead either to fear or to anger without any necessity for awareness by the individual. In this chapter we will first briefly review a few instances of our past research in which we employed some of the older tools of categorization and linear analysis. We will use these examples to demonstrate the questions that arose that we could not answer with these tools. For illustrative purposes we will then present two cases using Anne Frank’s diary and Albert Ellis’ autobiography and we will focus on the impact of one emotion on self-concept or identity development, that of fear and its neighbors such as anxiety and terror. These cases will offer clues about the functions of emotion in identity and lead us to consider a simple dynamic model for emotion, again focusing on fear, to bring us to new questions.

151

152


Old answers At the very beginning of our research, there was some debate about whether emotions could be studied at all as various writers referred to emotions as epiphenomena. To refute these points, we and other researchers began with a very concrete system that involved classifying facial movements into categories according to the systems of Ekman (1972), Izard (1971), and our own early coding system (Haviland, 1976). Using these categories of movement we could track the development of emotion expressions in the young infant (Haviland and Lelwica, 1987; Malatesta, Culver, Tesman, and Shepard, 1989) and into old age (Malatesta-Magai, Jonas, Shepard, and Culver, 1992). The categorical and linear approaches provided interesting answers to some old questions. The categories provided us with insight into mother–child interaction and the development of individual and familial emotional expression. For example, we found that infants match their mothers using different muscles in a familial way. If the mother’s face was primarily active in the brow region, her baby was likely to develop that pattern. We also found that very young infants respond with different facial expressions to their mothers’ repeated expression of a particular emotion such as anger or sadness or happiness but do not necessarily match their mothers. For example, many babies ‘‘froze’’ when their mothers acted angrily. So we learned that emotional patterns and developing responses were sometimes responsive to simple associative connections and sometimes not. How this might be related to later appearing identity was quite obscure and why they sometimes led to one response and sometimes to another was also unclear. Partly because we did not have particularly good tools for following continuity in development, we shifted to an older age group, working more directly on emotion and identity in early adolescence. We knew by the time we designed this project that adolescents’ self-perception of their emotional lives changed developmentally (Stapley and Haviland, 1989). Younger children are oriented to how concrete events influence their feelings – toys breaking, presents arriving, sports games won, and so forth. In early adolescence there is a dramatic change. Older children orient toward their relationships with peers to explain their most salient emotional experiences. Later in adolescence, young people orient toward their personal identity issues as instigators of salient emotion. Memorable emotional experiences may come from self-discovery and self-exploration, self-expectation and selfmonitoring. But the diversity of emotional experiences among individuals begins to emerge here as well. To what this diversity might be attributed was still not clear.

Old–new answers and new–old questions

153

These changes in self-report again suggest that there are qualitatively different meanings to emotional experience for a variety of reasons and that acquiring a particular set of meanings for emotional experience might not be a simple linear process. The investigator cannot always count on a dark and stormy night to inspire fear or a broken doll to inspire anger or a family birthday party to inspire happiness. This is not to claim that there are no fairly universal events, only that they are likely to be limited to extremes – death, mutilation, new life, and so forth. We still do not know quite how to approach this, so the data that we have tell a story that we know is too simple. Consider that it is not only the self-perception of emotion that changes during this time, but also the prominence of the type of emotion that is expressed (Kahlbaugh and Haviland, 1994). Genuine signals of emotional distancing emerge. The youngest adolescents hide their faces with their hair or hands or turn away or examine their fingers during conversation. Under the very same circumstances, older adolescents add more confronting expressions of distance such as contempt – eye rolling, blocking with arms and legs, and even snorting and head tossing. These changes appear to parallel in some way the changes in the adolescents’ sense of identity but they are not conscious, controlled changes. This research alerts us to levels of subconscious change. It also alerts us to the possible importance of repetitive, brief expressions in establishing the styles or tone of personality and of dialogue, even when people are unaware of the changes at the level that we are noting them. Perhaps small iterations of emotion expression across changing contexts lead to large changes in the organization of identity and interpersonal interaction. Perhaps such iterations even lead to categorical shift rather than mere enlargement of the same category. This suggests that it is not necessarily a gross and acknowledged change in the adolescent’s known social status in society, but could easily be an accumulation of small, subconscious changes in the adolescents’ emotions or motivations that is responsible for major changes in identity as well as social status. In our next studies we tried to get a picture of the changes in relations among self-defined roles and emotional traits. We asked children and young adolescents to tell us about their personal and social roles and their personality traits, particularly their emotional ‘‘traits.’’ To get a sense of how they related these aspects of their self-concept to each other we asked them to consider themselves in each role and to tell us how likely each trait was to emerge in that role. For example, only shyness might emerge as the emotion for a new role such as ‘‘entering high school’’ whereas many emotions might emerge in a familiar role such as ‘‘daughter’’ – anything from anger to shyness to happiness and

154


sorrow. Did adolescents just assimilate new roles to old emotional structures – adding the high school role to the daughter role – or is there a different sort of shift? The way that one analyzes such networks of data is to run a hierarchical clustering program (hiclas; see De Boeck and Rosenberg, 1988) that orders the roles so that roles with many defining traits are considered inclusive of most of the self’s possibilities and thus ‘‘higher’’ in the self-concept hierarchy. Roles with very few traits are subsumed by the roles that are associated with more traits. It is also possible to have sets of roles that do not fit together but form separate compartments or even to have roles that do not fit into the hierarchy but become residuals. In any case, the hierarchical clustering technique presents one with a picture of the relations that different roles have to each other in terms of traits. We found that ten- and eleven-year-old children had elaborate hierarchies of roles and traits that were quite predictable. The family roles tended to be the most inclusive, with school and friend roles secondary and very few unique roles such as ‘‘Auntie’s favorite.’’ In adolescence, however, the familiar roles such as daughter actually lost many of their positive emotions and traits while retaining their more negative ones. Because they lost traits, such ‘‘familiar’’ roles were unlikely to remain the most salient or at the top of the hierarchy, as they had been in childhood. New roles such as ‘‘apprentice beautician’’ emerged that did not fit well in the hierarchy and that did not seem as complex as the old ones in terms of their inclusiveness of possible traits. Roles dropped out and were not replaced by complex ones or inclusive ones. This may be a vortex during which unpredictable changes occur and the place of emotional traits seems to play a pivotal role. Clearly we could not use a simple linear additive model. At best, a rise and fall model might work, but what emerged after the fall was not the same as what existed beforehand. Our previous research seemed to tell us that we would not be able to detect how emotional processes entered into identity when taking account of data collapsed across many people. Trying to sum across varied time lines for emotions, roles, interactions, and changing selfperceptions for many people would obscure the developmental processes. It began to seem that we need to take one case (the x case) and then the next case (x + 1) (see Haviland and Kramer, 1991) and so forth, looking for the rules of process, not summation. We were being led to dynamic processes and network or matrix analyses by the lack of linearity in the data themselves.


155

Questions about identity and emotion: two special cases Anne Frank At the beginning of our analyses of Anne Frank’s diary we predicted that we would see how the emotion of fear became a major attractor for events and thoughts, a significant organizer of her self-concept or identity. We assumed that the war circumstances of her life terrified her often and perhaps continuously. We predicted that a search for the words that indicated fear would show that they became more frequent and more related to diverse corners of Anne’s life. Eventually fear would become an attractor in her self-concept. Using a simple model for the development of an anxious clinical syndrome we predicted that when fearful things happen, the person is sensitized to watch for fearful events, the slightly fearful aspects of unrelated events are exaggerated, the person develops an interpretation of life that revolves entirely around fear and anxiety, and so forth. Does Anne Frank’s diary show evidence of such an associative process? Is this a reasonable approach to conceptualizing fear – to think of it as a stable, global category, a singular thing? We know quite a bit about changes in Anne Frank’s emotional expression and also about her styles of cognitively solving everyday problems from an earlier analysis (Haviland and Kramer, 1991). By tracing change across time we had already discovered that cognitive development was preceded by changes in emotional expression. However, across the diary as a whole, different types of emotional expression tended to be more or less related to different cognitive forms, suggesting two processes. On the one hand, the rise in emotionality might predict a kind of disorganization or loosening of previous connections, just as we had seen in the cross-sectional studies. On the other hand, singular emotions, experienced repetitively, seemed to have the potential to become organizers that could affect mental processes. For example, when Anne was expressing her fears she had a tendency to use relativistic logic or thinking. We learned, contrary to our prediction, that the number of references to fear did not rise and fear did not become more variously associated with diverse events. Did it not change its place at all? What is faulty about our predictions? We divided Anne’s diary into four sections so we could examine the self-concept structure at each time and see whether fear was a significant organizer of the self-concept structure at any point. In each section we coded traits, especially emotional traits, and people or events that were described by or associated with the traits. We created data sets to which we could apply the hierarchical clustering techniques as we had

156


done in the previous cross-sectional studies on identity and the place of emotion. In the first section, before the Franks go into hiding, we found a matrix that resembled those collected from other children. Anne’s roles were daughter and student, sister and friend. The events were family- and school-related with just a subordinate pocket for war worries. Most of her roles were related to both negative and positive emotional traits and the most frequent traits were more positive than negative. Anne was a fairly typical child in a happy family at the beginning of the diary. Her thoughts about bombs, the dark of night, being Jewish, death, and war were highly negative and fearful; however, these thoughts made a separate small cluster or section in her self-concept, a cluster that was not integrated with her day-to-day life, nor an integral part of her self-concept. In the next section of the diary in which Anne describes the beginning of their life in hiding, her childhood roles are emotionally disconnected. Her daughter role is largely negative with an emphasis on anger; her school and friend roles only exist as memories or as she learns or fantasizes what the grim fate of her school friends has been. These roles emphasize depressive emotions. Happy moments form no repeating cluster. There is no longer any well-elaborated, complex set of roles for Anne that attracts both positive and negative emotions. She has moved into a period of segmented or disintegrated role structure, aware of being pulled by different sides of her emotional self in different scenes. The cognitive shift into a more relativistic style in which there was no certainty and no possibility of establishing certainty along with resolution of that relativism into absolute statements became common. The cluster of fearful war thoughts also is missing. The elements of it are separate but visible in the mix of fearfulness, anger, and sadness – the beginning of a complex depression syndrome – as if the fear had recruited other emotions, or perhaps they recruited fear. The war is not entirely outside in the night to be afraid of and separate from herself, but has begun to be a small part of her memories of better times. It is not an attractor for anything, however. In the third slice that we took from Anne Frank’s diary there is the beginning of a new organization of self around her relationship with her friend in the annex, Peter, and her sister, Margot. Both of these relationships have happy episodes and contents mixed with anger in Margot’s case and with some fearfulness as regards Peter and her new sexual feelings. A large, diverse cluster now surrounds sadness (not fear) – dead friends, memories of flirting with boys, outsiders, justice, her mother’s feelings of sorrow. Her self-concept is still segmented – one fairly broad and diverse cluster surrounds sadness with lower levels of other negative emotions and smaller clusters are related to


157

different issues. Fear is once again associated directly with bombings, terror of discovery. It was still not a major organizer. In the last section of Anne Frank’s diary a more interrelated selfconcept is developing, one that looks in its organization more like the earliest structure. But this new self-concept is far removed from the happy child we knew at the beginning of the diary. Anne’s roles with Peter and her annex family are complex and high in the hierarchy of clusters. What is surprising is that the most positive parts of herself are now intertwined with the war cluster, to hope for its ending, and beliefs about the ideal life that will follow the end of the war. The least positive or most dreaded aspects of Anne’s self-concept are in another cluster that contains her memories of her old self and her old life, her judgments of her own mood regulation and her personal criticism of her two-sided identity. The beginning self-concept in the diary was set around external objects and people and events in Anne’s young life. The ending selfconcept was set around internal and idealized distant objects. There were intervening glimpses of separated clusters of feelings and roles. None of this supported a linear process of change. None of this supported our original prediction about fear, either. When we come to the end of the diary, Anne’s most positive feelings and thoughts are centered in a wished-for and idealized future that is intertwined with her feared and mourned war experiences. Her most elaborate dreaded self is not associated with actual frightening war experiences, but with her own ability or lack of abilities to control her moods and express her true self. Is fear an organizer of either Anne’s idealized or self-evaluative self-concept clusters? Or is fear a process that is open to changing into other emotional states, highly variable, non-linear in its process? Fear is certainly contained in the idealized cluster and its absence may help define the second, but one cannot argue easily that fears are simply multiplying in Anne’s identity – in spite of her terror through many days and nights. There is an accumulation of depressing events, and an accumulation of Anne’s attempts to counter and control her depressive reactions. Perhaps because she cannot escape to a bigger world, being frustrated by the restraints of living in the annex, she turns to an inward examination of controlling depression and anxiety. Perhaps the constraint, independent of her thoughts about it, changes her long-term feelings. Albert Ellis In the case of Albert Ellis we have another opportunity to ask how fearful events might emerge to command a significant place in identity. This work on a particular emotion and its place in identity comes from a

158


larger study by Magai and Haviland-Jones (in press). We have extensively analyzed the ideology, attachment patterns, and emotions of Albert Ellis over several decades of his life. Ellis is exemplary as a person with a very stable adult self-concept that is associated with fear and the management of fear. Over many years, Ellis assimilates new information to a successful identity structure. He does not go through detectable episodes of disorganization and reorganization in his adult years. Fear (fright, terror, horror, panic, anxiety – the whole panoply of words that Ellis repeats over decades of writing) is a particularly compelling and toxic emotion for him. It is usually an emergency emotion that, when chronically activated, is physiologically enervating and can be ultimately deadly (Selye, 1956). It may have played an important role even in Ellis’ childhood illnesses and his chronic adult somatic problems, but it obviously did not prevent him from living a long and productive life, as he is still working and now is in his ninth decade. This gives rise to the possibility that fearfulness has many faces or functions. Ellis’ (1972) autobiography reveals that he believes he developed coping strategies when very young that were the forerunners of his theory of Rational Emotive Therapy (RET). Around the age of four, he began drilling into his head rules such as: ‘‘Life is full of hassles you can’t control or eliminate’’ (i.e. Things are terribly, frighteningly, uncontrollable) or ‘‘Wait before you panic’’ (i.e. Panic is imminent). Ellis’ childhood rules are obviously tools for managing forms of fearfulness. They are remarkable because there is no expectation that the sources of fright might be located and eliminated; in fact, the first rule is that they cannot be eliminated. The problem is therefore identified as the emotional feeling, as part of oneself, not as the external stimulus for the emotion. This is the sort of identity–emotion interaction that we saw beginning in Anne Frank. For Anne, the terror and sadness of the war became issues of personal mood control as she began to judge herself based on her emotional control. Even more clearly for Ellis, fears and fearlessness are personal traits and need to be controlled internally but this seems to carry even more weight than we had anticipated. In his autobiography, Ellis describes many roles and describes these roles with many traits, including many emotional traits. From this we extracted a self-concept model, as before. We took some liberties to reduce the number of traits by coding together ones that seemed to be used to indicate the same thing. For example, enjoyment and happy were grouped together; stupid, idiotic, and so forth were grouped together. Nevertheless, we ended up with more roles and traits than we had for Anne, probably indicative of a long and rich life. Again, using


159

hierarchical clustering methods, we developed a picture of how roles and traits were associated. For example, Ellis’ clients and himself-in-acrisis are clustered together because they share traits of anger (including hostility) and psychoticism (including emotionally disturbed). There are no roles that are superordinate to this cluster, meaning that Ellis’ descriptions of clients are not a piece of his description of some other role. Fear is the trait that is superordinate to this cluster, as predicted. As it turns out, fear is equally well associated with all the defined roles except for himself-as-a-therapist and himself-as-a-child. These last two roles are distinguished by their association with fearlessness, confrontation, logic, shyness, and happiness, and are not related to fearfulness. In other words, all of Ellis’ roles in the hierarchy are characterized either by fear or by the management of fear. The possible exceptions are some relationship roles that do not fit in any cluster in the hierarchy. This could indicate that the matrix is not as good a fit as we would like – perhaps that we need more data, or possibly that some more complex aspects of Ellis’ identity do not fit this type of matrix. Or it could indicate that many of Ellis’ roles are not systematically related to other roles. In any case, the earlier prediction that fearful experiences could organize self-concept does appear to be upheld in this case leading us to wonder why it is so different. Fear for Ellis, because it is a superordinate emotional trait, does not differentiate roles well; it is omnipresent in the identity construction. Ordinarily we think of fear as restricted to certain types of situations, such as fear of strangers, of illness, of heights, and so forth, as it originally was stimulated by the threat of bombs in Anne Frank’s account. But if fear can associate with any situation, there might be more variation in adult development than we previously suspected (on emotional biases in personality, see Magai and McFadden, 1995). Ellis’ autobiography is a history of stoicism and cleverness triumphing over what anyone else might construe as wretched early circumstances. He almost single-handedly battled forces of neglect, crossed dangerous streets, went to school at an extraordinarily young age, superintended his two younger siblings, and later in young adulthood cured himself of painful shyness, social inhibition, and sexual ineptitude. In addition to managing this collection of hardships, Ellis had major physical trauma. When he was five he was hospitalized for tonsillitis, developed a severe streptococcal infection that required emergency surgery to save his life and culminated in nephritis. He had recurrences of the nephritis, developed pneumonia, and was hospitalized repeatedly. He remembers being in the hospital for almost an entire year during which he was left largely alone for days at a time. Rather than recounting the negative feelings he might have had about

160


this time, he related how he took a scientific interest in the proceedings of one operation, which allowed him to watch pus coming out of his abdomen. From the account of Ellis’ early years one might predict depression and anxiety but Ellis emphasizes his successful management of these feelings. The organization of roles and emotional traits with fear at its center tells us that in order to enable this management, Ellis became vigilant for fearful events, finding evidence of manageable fear in all aspects of his life and continuing to find fear in new roles that were added as he reached maturity. These instances of fearfulness are separated from each other in his self-concept, however. As a homely metaphor one can imagine reciting all the meals one could have with cheese as an ingredient, but one would not imagine mixing them all together to have the ultimate in cheese dishes. Cheese is in every dish, but it has a different array of tastes and textures in each one. Similarly, Ellis has kept his experiences of fear, ubiquitous though they appear to be, separate and manageable. One of the striking things about Ellis’ autobiography is clear evidence of disconnection as an organization strategy, operating at many levels. In fact, he opens the autobiography with an assertion of disconnection: ‘‘I do not believe that the events of my early childhood . . . oriented me to becoming the kind of individual and the type of therapist that I now am’’ (p. 103). And yet elsewhere he ascribes the origins of RET to childhood ideas. For another example, there are a number of inconsistences in the narrative that leave a disconnected picture of his family. Ellis’ engagement and disengagement in love affairs in his early adulthood also mirror the segmentation or separation he values. Even in his cognitive strategies, as analyzed from his theoretical contributions, there is a strong preference for absolute, separable analyses rather than integrated systems of analysis. Segmentation and separation are a pattern across the various modes of looking at Ellis and fear appears to be the key. Fear is not just a simple reaction, here and gone, nor is it simply additive. Something is happening to the experience of fear that is changing its nature and its associations, so it is the backbone of a particular style of information processing. Ellis claimed that his manageable experience of fear happened at the level of gross intention – that he decided at the age of four to think of fear in this manageable way. Should we be persuaded by this? He is not likely to be aware of the continuing ubiquitous, if transformed, function of fear in his self-concept, nor is he likely to be aware of its association with social-cognitive strategies of separation. The belief that one has (or does not have) an emotional trait without necessarily knowing what process might have led one to that conclusion, or how its very existence


161

is controlling one’s thoughts about it, is not restricted to our two cases. As Ohman (1993) pointed out, the source of fearful reactions may be perceived only subconsciously. When the person reacts fearfully, but has no conscious grasp of the connections, he believes that the source is himself and further believes that he is irrational – that there is no real source. The paradox is that such a person may in fact become an unusually fearful person in the sense that he becomes hypersensitive to partial cues of potentially fearful things. Thus far, in our use of old questions and modes of analysis we have stayed at a gross level of categorization in our analyses. We have considered emotions, traits, thoughts, intentions, goals, beliefs, and so on as things that we could identify, separate, and hold constant. There has been some usefulness to this level of analysis but there are puzzling gaps in our predictions at this level. What could we learn about emotion and identity that would take us beyond such often faulty global predictions? To focus on only one problem, why might an Albert Ellis head towards fearfulness and an Anne Frank head towards depression? If we were to narrow our focus and expand our view of one aspect of this process, if we could use a microprocess to analyze just ‘‘fear’’ and its possible transformations, what might change in our understanding and could we then reintegrate with what we find at the grosser level? Microprocesses of fear: one special case When we think of emotions in dynamic systems terms, we necessarily begin to move from gross categories of emotions, thoughts, beliefs, and so on, to microprocesses. What kinds of process would we like to posit for fear? Fear most often seems to be the result of exposure to information that is increasingly intense and which overloads the perceptual and cognitive system. As Tomkins (1962) might have explained it, it is a global response to increasingly rapid and intense change from the person’s point of view, change that cannot be assimilated. ‘‘Our theory posits three discrete classes of activators of affect, each of which further amplifies the sources which activate them. These are stimulation increase [startle, fear, interest], stimulation level [sadness, anger], and stimulation decrease [joy] . . . such a set of mechanisms guarantees sensitivity to whatever is new, to whatever continues for any extended period of time and to whatever is ceasing to happen, in that order’’(Tomkins, 1962: 252). Tomkins suggested that all emotions, not just fear, were activated by changing patterns of neural firing, by which he meant changing receptivity, reactivity, and behavioral activity. Later on we will refer to emotions that involve stimulation increase or decrease as

162


‘‘change-based’’ emotions and those that involve stimulation level as ‘‘level-based’’ emotions. While most theories of emotion have either categories or dimensionality, but not both, they usually do not allow continuous transition from one emotion to another. Tomkins suggested that one emotional state could dissolve or morph into another. His proposal is analogous to color perception. Continuous change in the wavelength of light moves in and out of categories of color because receptors have special sensitivity curves. The stimulation is continuous, but the categories occur because of receptor qualities. Our emotion model could operate similarly with either receptor specificity or biochemical rate of change specificity and not violate modern theories of emotion. Fear occurs because the match in stimulation (itself not necessarily a categorical event) matches the receptors for fearfulness. Neural processes are linked to emotion (see Feyereisen, 1989; LeDoux, 1989; Panksepp, 2000a) and the ‘‘amplification’’ of sensory information results in behavioral changes which in turn produce more information based on the latest summary of stimulation (including the internal process as a source of stimulation) so that emotion is a continuous processing variable as well as a potential state. The ‘‘color’’ model of emotion does move us a bit away from naive psychology’s intuitive understanding of emotions as separate categories of information, but we can come back to the categories just as we can come back to colors. In this model, qualities of the information such as its intensity and rate of intrusion on the processor are part of the message, potentially leading to particular processing routes. It would not just be the content of what is assimilated, but qualities of the assimilation process that influence emotional responses. In our case, too much information intruding too quickly would be a source of anxiety or fear no matter what the content of the information might be. Panksepp (2000a) also proposes that there are different neurological systems or circuits for different emotions such as the ‘‘Rage,’’ ‘‘Fear,’’ and ‘‘Lust’’ circuits; interestingly, he notes that the effects of stimulation and activation of one circuit can spread to another. In the physical sciences we have learned that to study physical process one looks at rates of change and what or how different ‘‘forces’’ induce different rates. For example, to understand the acceleration of bodies in classical physics, time derivatives are used to model how forces are related to the motion of bodies. Influences (for us, this might be sensory input) on rates of change (e.g. patterns of sensory processing) show up ultimately in the state of systems, and the states then may be described in categories (e.g. discrete emotions). It is always possible that human systems do not work in ways analogous to such physical systems, but testing the models will reveal this in the long run.


163

An advantage of a micromodel for continuous emotional processes is that it allows us to demonstrate complex interactions that the computer can follow farther with iterations than we can predict analytically with grosser categories. Perhaps there are multiple outcomes to situations that begin with the same conditions, many more than we can anticipate. Partly, this may occur because random effects are magnified with many repetitions, as they tend to be in dynamic systems, especially more chaotic systems. (While our present model does not take all these possibilities into account, it lends itself to including them.) The systems we have been using historically to study emotions are very simple, categorical, and linear, as we have shown. This probably severely limits our understanding of individual differences. We have attempted to model emotion processes based on the above framework and assumptions. The basic idea in our model is that each emotion is activated by a particular pattern of stimulation and that such patterns feed back upon themselves, producing dynamic systems. Either the level of information or stimulation from low to high or the rate of change in the amount or density of information defines the emotion pattern. To make a demonstration model of this we chose a time series, the familiar sine wave, because it goes through the full range of theoretical possibilities for the model – a range of accelerating rates, decelerating rates, and some relatively stable levels. We then defined boundaries for each emotion along the time series, based largely on our interpretation of Tomkins’ original formulations, as they have been refined by our subsequent years of research. For example, fear was defined as occurring along a band of rapidly accelerating stimulation. Anger was defined such that it would occur when the wave stabilizes at a certain intensity and happiness was defined as being stimulated by a band of deceleration. In this model, any change in sensation occurring internally or externally could change the emotion, which raises the question of time boundaries in a general sense. Frijda and colleagues (e.g. Frijda, Mesquita, Sonnemans, and Van Goozen, 1991) propose that people are aware of moods, if not emotions, as lasting days and weeks. On the other hand, discrete emotions theorists argue that ‘‘real’’emotions last at most a few minutes and that most expressions can be measured in seconds or microseconds (e.g. Ekman, 1972). We suggest two answers to the problem of duration that can occur in our model. First, we suggest that durations are related to the type of emotion (MalatestaMagai and Culver, 1991). Emotions of level – that is, emotions that tell us that nothing is changing (e.g. depression) may be of any duration, but emotions of change, such as surprise, would necessarily be of

164


shorter duration. So part of the model would have to take the type of emotion into account. On the other hand, people’s ways of reporting emotions also must be taken into account. A frequently occurring emotion may be perceived as a constant emotion. Different people may differ in their awareness and in the sophistication of their reporting of different categories of emotion, as Ellis is highly sensitive to cues of anxiety. Some of the differences may be largely a matter of experience, culture, and training. It might be noted that awareness and ability to recognize emotions is another amplifier of information and hence could be construed as an additional stimulus for emotion. Next, we intend to give the reader an example so that some benefits of modeling emotions as microprocessors might emerge and so that we can reflect on the questions that remain in our cases of Anne and Albert. In our example, we suggest that a storm (the stimulus), with lightning followed later by a rumble of thunder, is in the vicinity of our subject, called Anne–Albert. We will show how our model predicts the emotional experience. This is a Gedanken experiment – neither Anne nor Albert has actually reported this experience. We somewhat arbitrarily modeled the stimulus change of lightning and thunder events by tracing the general shape of the lightning flash and that of the ensuing thunder clap with a normal probability curve and a chi square probability density curve. Superimposed on that curve are deviations obtained from random draws from a normal distribution. These manipulations make the stimulus irregular and convey some of the chaotic nature of real-world stimuli rather than working with the smoothing over of information into large categories that we usually prefer. Figure 7.1 shows the patterns of rate change and level in the amount of information in the storm at the top. Then it shows the effect of that stimulus on the processing of the storm as change-based emotions and as level-based emotions. In particular, when the stimulus intensity rises (e.g. lightning flashes) the level of neural response (density of neural firing) will rise. The rate of change will be directly proportional to the difference between the stimulus intensity and current level of neural response, but inversely proportional to the difference between the maximum possible level and the current level. We adopted this functional form because we thought the response to sensory inputs should reflect not only the magnitude of these inputs but also the finite resources of the response system. (For a fuller description: HavilandJones, Magai, and Boulifard, unpublished.) The level of processed information always lags slightly behind the sensory input to produce fear or surprise and the iterative model requires that prior state influences the processing of the present stimulus, so the curves are not exactly the same.


165

100

STIMULUS

80 60 40 20 0 0

1

2

3

4

5

6

7

0

1

2

3

4

5

6

7

8

100

DNF LEVEL

80 60 40 20

LEVEL-BASED EMOTION

0 8

150

100

DISTRESS ANGER PAIN

50

0

CHANGE-BASED EMOTION

0

1

2

3

4

5

6

7

8

150

JOY

100

INTEREST FEAR 50

STARTLE

0 0

1

2

3

4

5

6

7

8

TIME, secs

Figure 7.1. Emotion model time series for storm stimulus

The power of this simple little model lies in the interesting things it tells us about the complex stimulus and the emotions evoked. For example, we can see in the graphs that the lightning evokes roughly equal amounts of startle, fear, and interest but relatively little distress

166


and essentially no anger or pain. There is even some momentary joy as the lightning quickly dissipates. On the other hand, the thunder evokes mostly distress and anger. Why does the model work as it does? What you see in the lightning event is a rapid onset and release of the sensory input. It occurs too quickly for the level of information to rise to the required thresholds for relative stability in level-based negative emotions. What you see in the thunder event is a greater duration of high sensory input levels which allows the information level to reach these thresholds for distress and then anger and to sustain those values. To generalize, the model suggests that brief but dramatic changes result primarily in change-based emotions. On the other hand, slower and more pervasive stimuli result in level-based emotions, if the persistent level of stimulation is high enough. This model provokes many interesting speculations. Obviously, if we change the parameters of the model even a little (as one might when computing individual differences) there could be shifts in the categories of emotions produced. A change in parameter that affected the rate of processing could move fear towards more interest, for example, just as a change in threshold for constant information could change one person’s frequent anger to another person’s more frequent distress. This model also demonstrates that information processing can be related to emotional categories without any need for prior categorical learning or self-evaluation. One does not need to have been thrashed during a thunder storm in order to be angry. It also shows how very small differences in original state or in the stimulus itself could make large categorical differences that would have implications both for immediate behavior and long-term identity scripts. If the model we propose is even close to veridical it would help to make sense of one of the central problems in emotional processing – namely that similar stimuli evoke different emotions in different people. These different responses have generally been attributed to cognitive processes, as Ellis attributes his management of fear to his powerful decision-making abilities. While this is one source of variation in the process, in our model there could be many others. These thoughts led us to a simple variation in the emotion example. Not only may two people respond differently to the same stimulus, but even the same person will not be likely to respond the same way twice to repeated exposure. Our model handles this problem by making the previous state influence the processing of new information. Suppose, for example, young Anne–Albert was already angry when the next thunder and lightning storm unexpectedly hit. Instead of having the storm sequence impinge upon a tabula rasa as we had proposed in


167

100

STIMULUS

80 60 40 20 0 0

1

2

3

4

5

6

7

8

100

DNF LEVEL

80 60 40 20 0

LEVEL-BASED EMOTION

0

1

2

3

4

5

6

7

8

150

100

DISTRESS ANGER PAIN

50

0

CHANGE-BASED EMOTION

0

1

2

3

4

5

6

7

8

150

JOY

100

INTEREST FEAR 50

STARTLE

0 0

1

2

3

4

5

6

7

8

TIME, secs

Figure 7.2. Emotion model time series for storm while upset

the first instance, we now have the subject previously engaged in a personal internal ‘‘storm’’ at the point of origin for our Gedanken experiment. When we look at the figure 7.2 we see that this makes interesting differences. This time there is very little change-based emotion because

168


there is less difference between the new and background stimulus. Fear and surprise, as well as joy after the storm, are reduced because of the change in the original state of the system. On the other hand, distress and anger are pervasive and pain is increased during the lightning as well as the thunder. This little dynamic system for describing fear shows that when the state of origin is rather angry and the external stimulus is fearful, the resultant most frequent response can be distress. Anne, with her anger at having to adapt to the restrictions of the annex, may have been more in this situation than Albert who was free to roam and experiment, with the exception of his hospital stays. When the original conditions for processing are different, it is possible to show that fearful stimuli exacerbate distress, not fear and not necessarily anger, though both fear and anger are possible with a slight change in conditions. This very simple dynamic model of emotional process leads to testable hypotheses that do not emerge obviously from simple analytical approaches of gross categories. We could play with other parameters in this simple dynamic model to reflect individual differences, ad infinitum. We could change the level at which particular emotions shift categories. We could change the range over which each emotion is defined. We could get more complicated and use different parameters for each category of emotion. It is also easy to conceive of different slopes for transitions from one emotion to another. One of the remarkable things about the model is how shifting these parameters – manipulating small, continuous variables – changes the categories of emotions, often in ways that we had not predicted. Already, shifting our conceptualization of emotion to a dynamic system and focusing on a microlevel process reveals whole new types of questions that we were unlikely to ask about developmental emotional process just using grosser categorical approaches. These new questions either bypass or precede the old ones because they do not require that processes include awareness and synthesis of contextual information. Concluding thoughts In our earlier work we asked the old questions and tried to understand identity processes and their relation to emotion by using categorical methods of analysis and linear assumptions of additivity. In the present work, we abandoned these assumptions and approached the issue from a dynamic systems framework and now we find new questions. We modeled an emotion process mathematically with a microprocessor. We found that the distinction between classes of emotion and dimensions of emotion is perhaps an artificial one. We saw that information parameters having mainly to do with rates of change and levels of state


169

are, in themselves, capable of eliciting changing emotion categories. We saw that conscious or goal-oriented knowledge was not even fundamental to the shift. This may tell us something about why our semantic/ categorical sorting approaches would be bound to miss certain occurrences and reasons for surface appearance changes. There is often acknowledgment that emotions, considered as large summaries of contextual interaction, influence a wide variety of developmental processes that are collapsed under the term ‘‘identity.’’ Coming from one end of the spectrum and embedding this idea in their own dynamic model, we have Kunnen and Bosma (2000) putting ‘‘conflict’’ in their dynamic model of identity as a meaning-making developmental phenomenon for identity. ‘‘Conflict’’ is one way of conceptualizing where emotional process enters into the identity system, but it is as yet highly non-specific and does not include our concept of emotion as an ongoing process. Could positive emotional processes also have developmental potential for change, as Magai and McFadden (1995) have suggested? Perhaps they enter the Kunnen and Bosma model as the ‘‘support’’ functions. Such positive processes could be very significant for Anne Frank and Albert Ellis, for both of whom there are many happy moments and events, as well as clear conflicts. But this model does not explain why Anne with many ‘‘conflictful’’ events becomes somewhat depressed, whereras Albert with many ‘‘conflictful’’ events becomes sensitized to fear. This difference in emotional bias has a large effect on identity. Coming from a very different level of analysis, we see the work of researchers like Marc Lewis (1995) also suggesting that emotions aid in the association of concepts, causing them to form semantically meaningful systems leading to self-concept. Lewis argues that some aspect of emotional experience enters into a larger, self-organized form not unlike the self-concept structures we presented for Anne Frank and Albert Ellis. Once again, at even this more limited or perhaps more processoriented level where we consider the interaction of two proposed independent systems – the emotional and the cognitive – we are still missing the possibility that different emotions might enter the processes in different ways to produce not only different outcomes but also different associative processes. It is possible that the emotional and cognitive systems are not independent. Such a possibility is closer to the conclusions from the model of emotional experience we have proposed in which brief emotional changes, occurring beyond a conscious level of awareness, are the dynamic and continuous part of more general informational processing activities. The difference between a human thinker and a machine thinker (at this point in time) would be that the machine thinker would always use the same ways of combining bits and solving

170


problems, whereas the human processor would shift the mode of combining and solving problems because it is dependent upon the rate and intensity of the information. This shift would be called the emotional phase of the process. The concept of ‘‘phase shift,’’ useful in dynamic systems approaches, suggests that an existing system or state can shift in the sense that events or concepts that were stable and related to each other in particular patterns in one phase now change. When Anne Frank’s self-concept space shows evidence of change it seems that it could have been due to repeated emotional experiences and that this would lead to a realization of changing opportunities and threats. Those realizations would then lead to further change, but are not required to initiate change. It was just this sort of concept that we tried to access in our microprocessor model of emotions, admittedly at a very low level. These phase shifts offer enormous flexibility if the phase shifting lends itself to long-term adaptation under diverse circumstances, even though it might also lead to shorter periods of disorganization in self-concept. In our little model, when we showed emotional responses to the storm moving from one emotion to another – in brief and not necessarily conscious processes – we meant to illustrate that emotional process at this level instantiates phase shifts in attending to the stimulus. These very brief emotional phases are pieces of a response pattern in which different processes might be signaled and brought into play in Anne’s or Albert’s reaction to, and possibly even understanding of, stormy events. Beliefs and knowledge about emotions can operate to change emotion as well, in that such beliefs are themselves a part of the information, not just a way of managing emotion. Telling oneself not to jump at the crash of thunder because it is not dangerous is probably minimally effective in preventing fear. If one wanted to diminish fear and knew about the process, then singing loudly before the thunder is probably effective, but then one might respond with increased distress, as our model showed. Having a meta-affective knowledge of oneself and of emotional process in general may enable one to change many situations, thereby changing the emotional process but not always in the way that one predicts. The rise in distress is not intuitively obvious, for example. When we look back on earlier analyses of Anne Frank and Albert Ellis we might have been persuaded that we could understand the processes by asking about them at the semantic level. For example, Ellis has an explanation for developing his particular self-concept and for placing emotions within a very controlled cognitive system. Anne Frank also has explanations for the two sides to her self-concept – an


171

inner one with true feelings and an outer one with managed feelings. As reasonable as these explanations seem to them, our microprocess example encourages us to be more cautious in interpreting emotionality as a global construct or as one that is understood by the person herself or himself. Emotional processes may be about as accessible to consciousness as blood pressure fluctuation, even though the consequences of emotional or of blood pressure change may be perceived and explanations for such change concocted. Our secondary analysis of how we come to be as we are may make a good linear story, but it is not necessarily true especially at the level of understanding emotional process and how it is placed in the identity process of change and stability. As far as we know, we are the first researchers to propose that emotions might be modeled as continuous processes in dynamic systems. An examination of the many possible brief reactive as well as active emotional processes reveals much about emotional process that would not be predictable from more categorical or linear theories. Dynamic systems have the potential to trace enormous quantities of complex data and are made barely possible by the latest technology. This chapter opens up possibilities for us all in how we approach our questions about self-concept and emotion. Shall we ask ourselves or the subject what we believe or shall we study microscopic changes? Obviously it makes a difference. The difference has led us to propose that different forms of emotion are responses to different phases of information process both internally and externally, and that this constant phase shifting occurs on a microlevel as well as on the semantic meaningmaking level. The iteration of these processes leads to the changing formations of self-concept. Note 1 This work was partly supported by a grant to Carol Magai from the Minority Biomedical Research Support Program and the National Institute of Aging (1 SO6 GM54650-01).

COMMENTARY

Emotions as sources of information about the self Peter G. Heymans

Haviland, Boulifard, and Magai propose interesting points of view on the study of the self and emotions, and illustrate their positions with admirable research. I especially admire their efforts to present parts of the life trajectories of real people and illuminate their plights by analyzing underlying processes. The main plea, as I see it, is an appeal to leave the traditional methods of analysis in the field and switch to non-linear model building accompanied by computer/spreadsheetbased simulations of the model. Such a model-based simulation of the processing of fearful stimuli has been presented by Haviland et al. to demonstrate the power of non-linear models over classical approaches. The Haviland et al. model predicts the course of the intensities of seven emotions over a time period of eight seconds in the life of a single (hypothetical) individual. Below I will try to mitigate the enthusiasm in the Haviland et al.-plea by pointing out four warnings, extending beyond the Haviland et al. model. They concern: (1) anchoring the model-time unit in real developmental time; (2) the Haviland et al. (and my own) plea that the analysis of individual developmental trajectories should be accompanied by a compatible type of (statistical) treatment of the simulated trajectories; (3) the turn toward non-conscious processing of selfrelevant information at microlevel, which is implied in many nonlinear dynamical models, should not divert attention from solving unanswered questions about the cultural construction of the self and its development; (4) classical methods of analysis have not (yet) been exhausted in advancing our understanding of the development of the self, they were just not applied to the right type of data. Time in the modeling and simulation of developmental processes To construct a model one needs to combine in a suitable way assumptions about the processes involved. In order to compare predictions from the model with observed data parameters have to be estimated quantitatively. When the derivation of predictions from the model is 172

Commentary on Haviland-Jones, Boulifard, and Magai

173

mathematically cumbersome, simulations of the model-reality bring relief. One of the more important decisions for the model builder is the choice of a timescale in which model-reality unfolds. Is time measured in (milli-)seconds, hours, months or years? Of course we know that a day has 24 × 60 × 60 seconds. But the interesting ripples in simulated developmental curves which are compared to parts of an observed developmental curve might occur ten thousand seconds after starting the process, while in real life a similar ripple might be observable after ten thousand days. Moreover, at which point in the lifecourse does the simulated process start? Anchoring the time-unit and the starting point more firmly in real-life phenomena would greatly increase the value of (non-linear) model simulations in developmental psychology. In the Haviland et al. simulations presented time runs in seconds, starting from a well-defined event (lightning followed by thunder). But in general, we do not know (yet) which life events can serve as triggers for developmental processes. The estimation of the simulated developmental trajectories Usually several (ten-)thousands of process-unfoldings are run. The results of all these runs are averaged to obtain an estimate of the simulated developmental curve(s). These multiple runs can show quite variable results, due to the random components which are assumed to be part of the model’s assumptions. In fact the unexpectedness of process progression, given certain combinations of parameter values, is what seems to attract many researchers. The outcome of any single simulation-run is to be considered as one possible realization of a developmental trajectory. As such, a single run is comparable to a (part of a) life trajectory of a single individual. The simulated developmental curves presented in many textbook applications are thus averaged over many individual fates. (Note that this averaging is not the case in the Haviland et al. example.) Indices for variation around these mean values should reflect (a) inherent variability in the phenomenon under study, and/or (b) systematic inter-individual differences. Usually only the mean simulated values are given, variances not. In the case of simulating habitual ways of responding of single individuals, variances around the mean developmental curve can only refer to inherent unpredictablity of the process. Haviland et al. seem to downplay this unpredictability when they are showing how the previous emotional state influences an individual’s emotional reactions (comparison of figures 7.1 and 7.2). Systematic inter-individual differences in emotional responding between a hypothetical Anne and Albert are supposed to be explainable by assuming

174

Peter G. Heymans

that (1) both individuals react emotionally according to the Haviland et al. model, and (2) both individuals differ in the habitual emotional state, which serves as partial input to the Haviland et al. model, and (3) this habitual emotional state is the resultant of previous life history. It remains however to be explained how this habitual emotional state has been built up out of the thousands of residues of emotional encounters with life events. There is certainly a possibility for extending the presented Haviland et al. model for present-day concurrent emotions with a mechanism for cumulating residues of emotional encounters into a habitual mood-state. In this last part I see definitely a place for nonlinear modeling, as these models allow for catastrophe-like bifurcations in the course of the unfolding of a developmental process. Indeed, is it not the singular, unique encounters with events that individuals later on report as having been so influential in their lives? The cultural construction of the self and its development ‘‘Shifting our conceptualization of emotion to a dynamic system and focusing on a microlevel process reveals whole new types of questions that we were unlikely to ask about developmental emotional processes just using grosser categorical approaches’’ (Haviland et al., p. 167). I tend to agree, but warn – in my self-chosen role of skeptic – also of the possibility that results obtained at that microlevel might be difficult to relate to the grosser level of observable behavior. Ties between levels of analysis have to be demonstrated as well. Haviland et al. proceed by stating that ‘‘These new questions . . . bypass . . . the old ones because they do not require that processes include awareness and synthesis of contextual information’’ (emphasis added). Indeed, much of the success of psychology consists in having shown that the causal sequence of events leading up to a human perception, judgment, decision or act is not what it appears to be to the individual concerned, but follows a sequence which is (partly) hidden from awareness. However, it is also an undeniable fact that people are aware (erroneously or not) of some connections predisposing to behavior, and that people tend to act on the basis of that awareness, especially in social interaction. An individual’s identity may serve as an example. Let us focus upon identities connected to having succesfully completed so-called developmental tasks, e.g. as described by Erikson’s lifespan theory which can be found in every introductory textbook in (developmental) psychology. Not many people are aware that Erikson’s theory embodies a combination of two powerful cultural narratives. The first is the narrative of man on his earthly way to redemption and heaven being distracted by Cardinal Sins (with an origin in fourth-century contemplation experiences of monks in the


175

Syrian deserts). The second narrative depicts life as a journey in search of the realization of an ideal in which the hero is hindered by obstacles put by representatives of Evil; by fighting and overcoming the obstacles the hero receives honors from the audience (i.e. the narrative of Romance, Murray, 1985; Heymans, 1992). Many people map their life experiences on such a (combined) narrative, or are being mapped when a psychologist analyses their life from an Eriksonian perspective; the benefit of such a mapping is order in one’s set of experiences. Personhood and meaning spring from this order-generating mapping (Heymans, 1994). Let us have a closer look at the developmental task of adolescence: overcoming the ‘‘identity-crisis,’’ made famous by Erikson and his followers. Erikson assigned the Cardinal Sins to a specific phase in life, one for each phase, and used an 8-Sin instead of the standard 7-Sin Doctrine (Capps, 1989). Cardinal Sins were thought by medieval man to be conditions which hindered spiritual progress; these sins could be avoided or overcome by acquiring sin-specific virtues. Erikson postulated that the Cardinal Sin typical of adolescence was Pride, sometimes thought of by Church Fathers to be the basis of all other Cardinal Sins. The virtue to be acquired as antidote against Pride (a tendency to require continuous adulation from others) was Fidelitas (fidelity) conceived as personal predictability over situations. This fidelitas comes quite close to the demonstration of personality traits, which psychologists conceive as trans-situational behavioral consistencies. For Erikson, these situations are mainly diverse social roles which the adolescent is offered for enactment. Solution of the identity-crisis is nothing else than demonstrating relative cross-situational stability of behaviors. When the adult world arranges for a fifteen- to seventeen-year-old to enter into a variety of new situations (roles), none of the persons involved is aware that they are setting up a test of fidelitas for the adolescent. The cultural model structuring the interactions and the judgments is beyond awareness when adults and the adolescent are ‘‘conspiring’’ to make part of the Cardinal Sin model come true again. This example points – I hope – to the need to invest also in searching for mesolevel processes occurring at the nexus of the individual and his social context. Microlevel modeling should not distract researchers from what happens at the interface of culture and the developing person. Incantation-theory was proposed to clarify such processes (Heymans, 1994, 2000). Classical methods of analysis have not been exhausted The emphasis on non-linear dynamical models in developmental psychology has directed attention to the need for (multiple) timeseries data from single individuals (or other units of analysis, e.g. child–mother

176

Peter G. Heymans

dyad). There is no substitute for such data as a basis for statements about intra-individual, patterned changes. The near reflex-like application of psychometric methods of analysis (developed to serve a clear description of inter-individual differences) has led to a depreciation of individual case data, which until now were not frequently collected. Classical methods can help the developmental psychologist to uncover aspects of the developing self, given that multiple timeseries from a single individual are available. Space restrictions permit only some indications of what is possible. Emotions indeed are a powerful source of information about the self. Therefore, part of my favorite research design is to collect timeseries of the intensities of multiple emotions (between 40 and 140) from single individuals, over long stretches of time (e.g. between 60 and 260 consecutive days). This is the same type of data that can serve in tests of non-linear dynamical models. However, I use classical methods, all within the SPSS/Timeseries Statistical Software package. Calculation of (partial) autocorrelations within an emotion timeseries can tell about periodicity in this emotion, just as does spectral analysis. Cross-lagged correlations between timeseries for different emotions tell whether one emotion leads to another emotion, and after how many days. Most importantly, by clustering the patterns of activation-over-time of multiple emotions from a single individual the resultant emotion clusters can be interpreted as ‘‘concerns’’ in the sense of Frijda’s (1986) emotion theory, and the activation of these concerns over time can be described. Internal dynamics of these concern-activations over time can be described by ARIMA-models (see the SPSS/Timeseries package). When information about life events is also available it is possible to determine their effects on concern-activations, e.g. by adding to the relevant ARIMA-model a regression model representing this effect. Cross-lagged correlation analysis of the ARIMA-whitened concernactivation scores will give information about the intra-individual internal dynamical relations between concern-activations. For an example of these analyses see Heymans (2000). In conclusion: the best short-term consequence of the increased attention to non-linear dynamic models in developmental psychology is the need to revalue and collect (multiple) timeseries data from single cases. In my opinion, these cases should be individuals-in-their-socioculturalcontext, and the data should comprise also indicators for processes supposed to occur at the nexus between developmental context and individual.

COMMENTARY

Emotions as sources of information about the self Peter G. Heymans

Haviland, Boulifard, and Magai propose interesting points of view on the study of the self and emotions, and illustrate their positions with admirable research. I especially admire their efforts to present parts of the life trajectories of real people and illuminate their plights by analyzing underlying processes. The main plea, as I see it, is an appeal to leave the traditional methods of analysis in the field and switch to non-linear model building accompanied by computer/spreadsheetbased simulations of the model. Such a model-based simulation of the processing of fearful stimuli has been presented by Haviland et al. to demonstrate the power of non-linear models over classical approaches. The Haviland et al. model predicts the course of the intensities of seven emotions over a time period of eight seconds in the life of a single (hypothetical) individual. Below I will try to mitigate the enthusiasm in the Haviland et al.-plea by pointing out four warnings, extending beyond the Haviland et al. model. They concern: (1) anchoring the model-time unit in real developmental time; (2) the Haviland et al. (and my own) plea that the analysis of individual developmental trajectories should be accompanied by a compatible type of (statistical) treatment of the simulated trajectories; (3) the turn toward non-conscious processing of selfrelevant information at microlevel, which is implied in many nonlinear dynamical models, should not divert attention from solving unanswered questions about the cultural construction of the self and its development; (4) classical methods of analysis have not (yet) been exhausted in advancing our understanding of the development of the self, they were just not applied to the right type of data. Time in the modeling and simulation of developmental processes To construct a model one needs to combine in a suitable way assumptions about the processes involved. In order to compare predictions from the model with observed data parameters have to be estimated quantitatively. When the derivation of predictions from the model is 172


173

mathematically cumbersome, simulations of the model-reality bring relief. One of the more important decisions for the model builder is the choice of a timescale in which model-reality unfolds. Is time measured in (milli-)seconds, hours, months or years? Of course we know that a day has 24 × 60 × 60 seconds. But the interesting ripples in simulated developmental curves which are compared to parts of an observed developmental curve might occur ten thousand seconds after starting the process, while in real life a similar ripple might be observable after ten thousand days. Moreover, at which point in the lifecourse does the simulated process start? Anchoring the time-unit and the starting point more firmly in real-life phenomena would greatly increase the value of (non-linear) model simulations in developmental psychology. In the Haviland et al. simulations presented time runs in seconds, starting from a well-defined event (lightning followed by thunder). But in general, we do not know (yet) which life events can serve as triggers for developmental processes. The estimation of the simulated developmental trajectories Usually several (ten-)thousands of process-unfoldings are run. The results of all these runs are averaged to obtain an estimate of the simulated developmental curve(s). These multiple runs can show quite variable results, due to the random components which are assumed to be part of the model’s assumptions. In fact the unexpectedness of process progression, given certain combinations of parameter values, is what seems to attract many researchers. The outcome of any single simulation-run is to be considered as one possible realization of a developmental trajectory. As such, a single run is comparable to a (part of a) life trajectory of a single individual. The simulated developmental curves presented in many textbook applications are thus averaged over many individual fates. (Note that this averaging is not the case in the Haviland et al. example.) Indices for variation around these mean values should reflect (a) inherent variability in the phenomenon under study, and/or (b) systematic inter-individual differences. Usually only the mean simulated values are given, variances not. In the case of simulating habitual ways of responding of single individuals, variances around the mean developmental curve can only refer to inherent unpredictablity of the process. Haviland et al. seem to downplay this unpredictability when they are showing how the previous emotional state influences an individual’s emotional reactions (comparison of figures 7.1 and 7.2). Systematic inter-individual differences in emotional responding between a hypothetical Anne and Albert are supposed to be explainable by assuming

174

Peter G. Heymans

that (1) both individuals react emotionally according to the Haviland et al. model, and (2) both individuals differ in the habitual emotional state, which serves as partial input to the Haviland et al. model, and (3) this habitual emotional state is the resultant of previous life history. It remains however to be explained how this habitual emotional state has been built up out of the thousands of residues of emotional encounters with life events. There is certainly a possibility for extending the presented Haviland et al. model for present-day concurrent emotions with a mechanism for cumulating residues of emotional encounters into a habitual mood-state. In this last part I see definitely a place for nonlinear modeling, as these models allow for catastrophe-like bifurcations in the course of the unfolding of a developmental process. Indeed, is it not the singular, unique encounters with events that individuals later on report as having been so influential in their lives? The cultural construction of the self and its development ‘‘Shifting our conceptualization of emotion to a dynamic system and focusing on a microlevel process reveals whole new types of questions that we were unlikely to ask about developmental emotional processes just using grosser categorical approaches’’ (Haviland et al., p. 167). I tend to agree, but warn – in my self-chosen role of skeptic – also of the possibility that results obtained at that microlevel might be difficult to relate to the grosser level of observable behavior. Ties between levels of analysis have to be demonstrated as well. Haviland et al. proceed by stating that ‘‘These new questions . . . bypass . . . the old ones because they do not require that processes include awareness and synthesis of contextual information’’ (emphasis added). Indeed, much of the success of psychology consists in having shown that the causal sequence of events leading up to a human perception, judgment, decision or act is not what it appears to be to the individual concerned, but follows a sequence which is (partly) hidden from awareness. However, it is also an undeniable fact that people are aware (erroneously or not) of some connections predisposing to behavior, and that people tend to act on the basis of that awareness, especially in social interaction. An individual’s identity may serve as an example. Let us focus upon identities connected to having succesfully completed so-called developmental tasks, e.g. as described by Erikson’s lifespan theory which can be found in every introductory textbook in (developmental) psychology. Not many people are aware that Erikson’s theory embodies a combination of two powerful cultural narratives. The first is the narrative of man on his earthly way to redemption and heaven being distracted by Cardinal Sins (with an origin in fourth-century contemplation experiences of monks in the


175

Syrian deserts). The second narrative depicts life as a journey in search of the realization of an ideal in which the hero is hindered by obstacles put by representatives of Evil; by fighting and overcoming the obstacles the hero receives honors from the audience (i.e. the narrative of Romance, Murray, 1985; Heymans, 1992). Many people map their life experiences on such a (combined) narrative, or are being mapped when a psychologist analyses their life from an Eriksonian perspective; the benefit of such a mapping is order in one’s set of experiences. Personhood and meaning spring from this order-generating mapping (Heymans, 1994). Let us have a closer look at the developmental task of adolescence: overcoming the ‘‘identity-crisis,’’ made famous by Erikson and his followers. Erikson assigned the Cardinal Sins to a specific phase in life, one for each phase, and used an 8-Sin instead of the standard 7-Sin Doctrine (Capps, 1989). Cardinal Sins were thought by medieval man to be conditions which hindered spiritual progress; these sins could be avoided or overcome by acquiring sin-specific virtues. Erikson postulated that the Cardinal Sin typical of adolescence was Pride, sometimes thought of by Church Fathers to be the basis of all other Cardinal Sins. The virtue to be acquired as antidote against Pride (a tendency to require continuous adulation from others) was Fidelitas (fidelity) conceived as personal predictability over situations. This fidelitas comes quite close to the demonstration of personality traits, which psychologists conceive as trans-situational behavioral consistencies. For Erikson, these situations are mainly diverse social roles which the adolescent is offered for enactment. Solution of the identity-crisis is nothing else than demonstrating relative cross-situational stability of behaviors. When the adult world arranges for a fifteen- to seventeen-year-old to enter into a variety of new situations (roles), none of the persons involved is aware that they are setting up a test of fidelitas for the adolescent. The cultural model structuring the interactions and the judgments is beyond awareness when adults and the adolescent are ‘‘conspiring’’ to make part of the Cardinal Sin model come true again. This example points – I hope – to the need to invest also in searching for mesolevel processes occurring at the nexus of the individual and his social context. Microlevel modeling should not distract researchers from what happens at the interface of culture and the developing person. Incantation-theory was proposed to clarify such processes (Heymans, 1994, 2000). Classical methods of analysis have not been exhausted The emphasis on non-linear dynamical models in developmental psychology has directed attention to the need for (multiple) timeseries data from single individuals (or other units of analysis, e.g. child–mother

176

Peter G. Heymans

dyad). There is no substitute for such data as a basis for statements about intra-individual, patterned changes. The near reflex-like application of psychometric methods of analysis (developed to serve a clear description of inter-individual differences) has led to a depreciation of individual case data, which until now were not frequently collected. Classical methods can help the developmental psychologist to uncover aspects of the developing self, given that multiple timeseries from a single individual are available. Space restrictions permit only some indications of what is possible. Emotions indeed are a powerful source of information about the self. Therefore, part of my favorite research design is to collect timeseries of the intensities of multiple emotions (between 40 and 140) from single individuals, over long stretches of time (e.g. between 60 and 260 consecutive days). This is the same type of data that can serve in tests of non-linear dynamical models. However, I use classical methods, all within the SPSS/Timeseries Statistical Software package. Calculation of (partial) autocorrelations within an emotion timeseries can tell about periodicity in this emotion, just as does spectral analysis. Cross-lagged correlations between timeseries for different emotions tell whether one emotion leads to another emotion, and after how many days. Most importantly, by clustering the patterns of activation-over-time of multiple emotions from a single individual the resultant emotion clusters can be interpreted as ‘‘concerns’’ in the sense of Frijda’s (1986) emotion theory, and the activation of these concerns over time can be described. Internal dynamics of these concern-activations over time can be described by ARIMA-models (see the SPSS/Timeseries package). When information about life events is also available it is possible to determine their effects on concern-activations, e.g. by adding to the relevant ARIMA-model a regression model representing this effect. Cross-lagged correlation analysis of the ARIMA-whitened concernactivation scores will give information about the intra-individual internal dynamical relations between concern-activations. For an example of these analyses see Heymans (2000). In conclusion: the best short-term consequence of the increased attention to non-linear dynamic models in developmental psychology is the need to revalue and collect (multiple) timeseries data from single cases. In my opinion, these cases should be individuals-in-their-socioculturalcontext, and the data should comprise also indicators for processes supposed to occur at the nexus between developmental context and individual.

CH A P T E R 8

Cognitive–emotional self-organization in personality development and personal identity Marc D. Lewis and Michel Ferrari The continuity of identity, despite ongoing change in the person and the world, has challenged thinkers since ancient times. Identity has its roots in the Latin word for same. According to the Oxford English Dictionary (on-line), identity involves ‘‘the sameness of a person or thing at all times or in all circumstances’’ (OED 2.a) and, more specifically, personal identity involves ‘‘the condition or fact of remaining the same person throughout the various phases of existence.’’ Locke (1690), perhaps the first to propose a modern sense of personal identity, wrote that ‘‘The Identity of the same Man consists . . . in nothing but a participation of the same continued Life, by constantly fleeting Particles of Matter, in succession vitally united to the same organized Body’’ (ii. xxvii. Sect. 6). Yet the roots of Locke’s statement go at least as far back as Plato’s symposium (Plato, ~ 390, 207.d). Psychologists in our era have attempted to solve the riddle of identity by proposing the construction of a self structure (Marcia, 1980), concept, or theory (Schlenker and Weigold, 1989), built out of cognitive and social constituents. Whether viewed as a schema of the self, a theory, a set of traits or dispositions, or a hierarchy of defenses and goals, identity is understood by conventional theories as a stable structure built up over development. For such an identity, continuity over time is not difficult to explain. Building-block structures maintain their sameness by virtue of an invariant relation among their parts and an invariant set of functions or transactions with the world. The present volume advances the premise that traditional models of identity, with their constructivist flavor, do not offer adequate explanations. Rather, as Bosma (1995) and others have suggested, the interplay of continuity and change is central to identity, just as it is to personality more generally (e.g. Block and Robins, 1993; Mischel and Shoda, 1995). For example, conventional models have a difficult time explaining identity change, because their explanation of continuity relies on principles of organization that are foreign to natural systems. Change and 177

178


stability are antithetical in these models, but they are different facets of the same phenomena in nature. Indeed, as proposed by Locke over 300 years ago, personal identity can only be understood in relation to the changing, or ‘‘constantly fleeting,’’ substrate out of which it emerges. In keeping with this insight, principles of self-organizing dynamic systems have been proposed as the basis for a new generation of models of identity and identity development (Bosma, 1995; Haviland and Kahlbaugh, 1993; Kunnen and Bosma, 1994; Lightfoot, 1997). The present chapter is one attempt to further this agenda. Our chapter addresses the development and continuity of personality and identity. We begin with principles of self-organization and self-perpetuation in natural systems. We then go on to discuss personality self-organization, modeled as a developmental consolidation of cognition–emotion interactions, and describe its relations with realtime appraisals and mood states. Finally, we move into the realm of reflective, autobiographical, and dialogical activities through which identity continuously emerges as a specialized product of the personality system. The relation between personality and what we usually refer to as ‘‘personal identity’’ can be schematized as implicit identity versus explicit identity. Personality involves interpretations, emotions, goals, and intentions that persist over time (McAdams et al., 1997; Mischel and Shoda, 1995). This sort of identity (i.e. sameness) remains largely implicit, unknown, and unconscious, and it describes young infants and non-human animals. Perhaps more complex and certainly more specifically human is personal identity. This is experienced by individuals as statements, stances, policies, or stories about who they are and how they live their lives. This sort of identity is at least partly explicit and conscious, it reflects the sociocultural matrix in which individuals are raised, and it involves unifying narratives that make sense of – or explicate – disparate actions in one’s past. In order to arrive at a model of explicit identity, we first attempt to establish an understanding of personality, the implicit identity that serves as a foundation for psychological sameness over the lifespan. Continuity despite change in self-organizing systems Continuity of form in the natural world is evident everywhere, despite the massive fluctuations which surround and penetrate it. It is present in the demarcations of species which can last for tens of millions of years despite changes in weather, ecology, and even the genome itself. It is present in climatic conditions that persist for eons, despite extreme seasonal changes and gradual upheavals in the topography of land and

Cognitive–emotional self-organization

179

sea. It is present in ecosystems, with their uncanny capacity to maintain or renew a balance among predators, prey, waste products, and vegetation. But the constancy of these forms is not that of a predetermined structure or rule system. For example, the genome of each species is not a program that dictates adult morphology. Nor is it a product of some external source by way of instruction or copying. The ecosystem in a lake may depart radically from that of the river flowing into it or the lake previous to it in a chain of lakes. Natural forms are not wholes built piecemeal out of parts, and there is nobody monitoring their constancy, to repair it when it fails. Principles of prespecification, programming, instruction, or construction, so central to psychological explanations, have no place in the natural world and in the continuity of its forms. Rather, principles of self-organization and self-perpetuation are seen as the basis for the orderliness of nature (Kauffman, 1993). Models of self-organization focus on the emergence of order from disorder, and in particular the emergence of complex, higher-order forms from simpler, lower-order components. Examples are the emergence of living microbes through the interaction of complex molecules, the emergence of herding behavior out of the interactions of individual animals, and even the emergence of embryonic structure through the interaction of proliferating cells (e.g. Goodwin, 1987). These processes exemplify the convergence or crystallization of global organization from recurring interactions among simpler components, providing structure and pattern in all of nature (Prigogine and Stengers, 1984). There are at least three ways in which principles of self-organization explain growth and change in natural systems. First, and most important, is the emergence of novel, higher-order forms from recursive interactions among lower-order elements. When systems are unstable, these interactions give rise to positive feedback loops that amplify novel coordinations into macroscopic patterns which replace the previous organizational regime (Prigogine and Stengers, 1984). These patterns are actually arrangements of coupled elements that persist because they move energy through the system efficiently. Second, recurring patterns of coupling or coordination change the elements and the connections that give rise to them. These changes facilitate similar couplings on subsequent occasions, so that active habits grow in strength and replace competing organizations (Thelen and Smith, 1994). Third, self-organizing systems shift between alternative patterns of coupling in abrupt changes called phase transitions (Kelso, 1984). Often, small changes in a particular environmental parameter (e.g. heat, concentration) trigger transitions between stable states. But they do not determine them, as causation resides within the system itself.

180


While principles of self-organization are usually applied to understanding growth and change, they must also account for stability, continuity, and the maintenance of order. Particularly in living systems, it is obvious that the maintenance of organization is necessary, not only for survival, but for systems to grow and evolve (Maturana and Varela, 1987). Our discussion of personality and identity will be grounded in developmental considerations, but its greatest challenge is to model continuity. Personal identity exemplifies continuity of form as much as anything in the natural world. We therefore begin by asking how continuity is accomplished in nature. Biological models of self-maintenance are complex and difficult, but a number of general principles can be extracted. Self-continuity results from recurrent interactions – change rather than stasis (e.g. Maturana and Varela, 1987). This is as true of a peaceful nap after a large meal as the ordering of birds in flight or the steadfast continuity of a runner or mountain-climber. Continuity emerges out of enduring or recurring patterns of coupling among constituents that are in flux. This is because self-organized systems are constantly taking in and dissipating energy and information through their transactions with the environment, and the coupling of their constituents funnels this energy along lines of least resistance. Higher-order continuity through lower-order coupling can be found at many hierarchical levels in complex living forms – from cells to cell assemblies, to modular structures such as organs or leaves, to systems of such structures, and finally to the organism as a whole. Thus, the continuity of (self-organized) biological systems resides in an ongoing, macroscopic ordering of microscopic interactions across several scales. How does this ordering come about? The microscopic interactions at any level of a living hierarchy fall into preferred or familiar patterns of coupling, giving rise to an emergent, higher-order form. These patterns of coupling guide the flow of energy through the system (or subsystem) despite decay, perturbation, and even structural change to the system itself. Meanwhile, the higher-order forms that emerge in this process can be described as attractors on the state space of the system, demonstrating its tendency to occupy only a few states despite the multitude of potential states allowed by the physical properties of its constituents. These preferred configurations are constrained by the nature of the elements, their modes of interaction with each other (e.g. damping, activating, competing, catalyzing), and the pattern of connections between them. But also, most importantly, they are constrained by their functional relationship to the whole system – the levels of the hierarchy above and below – and the environment beyond. Such interacting constraints can be thought of as a matrix of complementarities that


181

supports or favors particular patterns and permits spontaneous change from one preferred pattern to another. Complementarities favor the emergence and continuity of orderliness, but they do not act like a set of rules dictating a particular outcome. For example, the population level of a species of birds stabilizes through complementarities across mating patterns, nesting supplies, food supplies, and predation, coupled with overall reproductive success. But there is no set of rules that stipulates what that population level should be. Thus, continuity arises from interactions guided by complementarities among constituents and functions, and these reflect the essential character of the system through its patterned behavior in real time. Concepts such as ‘‘decentered structure’’ (Goodwin, 1987), ‘‘relative stability’’ (Landauer, 1987), and ‘‘dynamic stability’’ (Thelen and Ulrich, 1991) capture continuity of this sort. Self-perpetuation in biological systems is of course much more complex and difficult than implied by this brief discussion. For one thing, mechanisms of self-regulation emerge in self-organizing systems of sufficient complexity and protect them from deviation and change. For another, self-replication is a particular type of continuity, adopted by all biological systems, through which identity is transmitted over time and space. We will not deal specifically with these ideas in this chapter, but they will be important considerations for further theorizing in this area. For now, we can begin to model the emergence of continuity in personality and identity armed with the basic principles of recursion, coupling, and complementarities. A dynamic systems perspective on personality development Before going on to model identity from a self-organizational perspective, it is necessary to establish a more general picture of personality development and emerging personality consistency. Our approach follows the general strategy of dynamic systems (DS) approaches in developmental psychology. These approaches assume that developing systems consolidate through recurrent interactions among psychological, social, or perception-action components, passing through phases of reorganization at transition points (Fogel, 1993; Lewis, 2000b; Thelen and Ulrich, 1991; Van Geert, 1991). Developing systems are thus selforganizing systems, accruing order through increasingly complex coordinations. This orderliness can be represented by attractors on a state space (Lewis, Lamey, and Douglas, 1999; Thelen and Smith, 1994) or by the stabilization of a growth profile or time series (Van Geert, 1994). It is important to note that the term self-organization can describe emergent processes at many time scales. It can refer to the movement of behavior

182


to attractors in real time (e.g. over seconds) and to the emergence and consolidation of attractors in developmental time (e.g. over the course of months or years) (Thelen and Ulrich, 1991). Thus, DS approaches are capable of tracking convergence, reorganization, and crystallization at several scales at once, and the interaction of time scales has been of particular interest to psychologists with a dynamic systems view (e.g. Van Gelder and Port, 1995; Thelen and Smith, 1994). To date, motor development (Thelen and Ulrich, 1991), cognitive development (Smith, 1995; Van der Maas and Molenaar, 1992; Van Geert, 1991), and communicative development (Fogel, 1990, 1993) have been most thoroughly studied using DS methods. However, DS models of personality development are now beginning to appear (Derryberry and Rothbart, 1997; Lewis, 1995, 1997; Magai and Nusbaum, 1996; Schore, 1997, 2000; Cloninger, Svrakic, and Svrakic, 1997). Most of these approaches see personality as being knit together from psychological, psychosocial, and/or biological constituents over developmental time, with early patterns setting the course for the subsequent channeling of learning and experience. Many of these models also describe relations among emotion, cognition, and behavior as central to the development of individual styles. The promise of these approaches is to explain – not just describe – personality as an emerging, consolidating, and continuous organization (Lewis and Granic, 1999a). From this perspective, personality is not specified by genes or environment or constructed out of parts; it is a self-organizing system that converges to its own unique form and perpetuates that form as its implicit identity. Cognition–emotion interactions and personality self-organization According to the present approach, personality development can be described as change, stabilization, and refinement of recurrent patterns of cognition–emotion interactions, represented by attractors on the state space of the psychological system (Lewis, 1995, 1997). Like other theoretical models of personality development, this account has little in common with trait or factor-analytic approaches to adult personality, the latter being unconcerned with the fundamental processes of personality formation and change. More specifically, the idea that personality arises from cognition–emotion interactions derives from theories of emotion and emotional development, though it also resonates with some personality theories (e.g. Mischel and Shoda, 1995). Tomkins (1978) and Izard (1984) viewed personality as the construction of affective-cognitive or ideoaffective structures, arising through the child’s most frequent or important experiences. These structures are defined as linkages between frequently felt emotions and accompanying cognitive


183

interpretations. Affective-cognitive structures are assumed to generate characteristic behaviors that indicate trait-like qualities in the eyes of others (Malatesta and Wilson, 1988). As to the mechanisms underlying these structures, Tomkins and Izard emphasize the cueing of cognitive processes by emotions, whereas cognitive-developmental approaches highlight the opposite causal direction – the elicitation of emotions by cognitive appraisals (e.g. Sroufe, 1979). These two causal directions are brought together by Malatesta and Wilson (1988), who propose a ‘‘recursive and sustaining relationship’’ (p. 103) between cognitive interpretations and emotions underpinning personality development. The premise of cognition–emotion recursion has been crucial for modeling emotional and personality self-organization in real time and development (Lewis, 1995, 1996, 1997). Such modeling efforts have much in common with dynamic systems approaches of Camras (1992), Fogel (1993), Magai and Nusbaum (1996), and recently Izard, Ackerman, Schoff, and Fine (2000) and Mascolo, Harkins, and Harakal (2000). A comprehensive theory of cognition–emotion interactions is currently under construction (Lewis, 2000a). This account specifies three time scales of emotional self-organization: appraisal–emotion episodes in real time, moods that last from minutes to days, and personality patterns that persist over a lifetime. At each of these scales, cognition–emotion configurations are proposed to converge from initial indeterminacy and variability to coherent and lasting states. Thus, appraisals, moods, and personality each develop, giving rise to nested scales of emotional self-organization in microdevelopment, mesodevelopment, and macrodevelopment. According to this account, a cognitive appraisal or interpretation converges in real time from the coordination of (lower-order) elements such as concepts, associations, and perceptions, but it does so only in interaction with a consolidating emotional state. The convergence of an appraisal augments and constrains emotional activation while emotional activation simultaneously augments and constrains attentional processes at work in appraisal (Lewis, 1995). In contrast to conventional appraisal theory, appraisals are not presumed to be independent of emotion, nor to precede emotion in time. Rather, as argued by Frijda (1993b, in press), it is not until the end of an appraisal process that cognitive interpretations are comprehensive and complete. Thus, rather than view cognitive appraisals as eliciting emotions, or emotions as cueing cognitive activities, appraisal and emotion are proposed to arise in tandem (cf. Buck, 1985) and to stabilize through ongoing feedback. This is consistent with the notions of reciprocal causation mentioned by previous theorists (e.g. Malatesta and Wilson, 1988). The synchronization of appraisal and emotion can be described as a

184


macroscopic coupling between the cognitive and emotional systems, giving rise to a coherent emotional interpretation (EI) in real time. Emotional interpretations are similar to Izard’s affective–cognitive structures, but they are construed as temporary organizations rather than lasting structures. In an EI of blame, for example, anger rapidly couples with the sense that someone is at fault. This coupling subsumes more intricate couplings among microscopic cognitive constituents. For example, the sense that someone is at fault results from couplings among an image of a person, an inference of harm, and the attribution of power or intention. In a more complex example, the cognitive couplings that contribute to an EI of inadequacy might include links betwen concepts of ‘‘small,’’ ‘‘helpless,’’ and ‘‘baby,’’ expectancies of humiliation, associations between adulthood and power, and images of parental disdain. However, these cognitive couplings would not cohere unless nested in a global coupling with emotion, such as anger in the first example and shame in the second. In a recent look at emotional self-organization, Izard et al. (2000) suggests that coupling among basic emotion systems creates comprehensive emotion patterns, just as cognitive couplings underpin comprehensive interpretations in the present account. EIs may thus involve the coupling of constituents in both systems. (At a neural level, this is inevitable, but psychological terminology does not make clear distinctions between networks and single entities.) The idea that coherent psychological states arise from coupling within and between cognitive and emotional systems rests partly on neurobiological evidence (Damasio, 1994; Freeman, 1995; Schore, 1997; Harkness and Tucker, 2000). According to this research, cognition– emotion interactions are centered in the relation between the prefrontal cortex (e.g. ventromedial prefrontal or orbitofrontal cortex), which acts as a convergence zone for all cortical regions, and the emotional circuitry of the limbic system, which includes various connected nuclei. Appraisals in the prefrontal convergence zone arise from the entrainment of many other cortical regions, including all perceptual and motor regions, and they mediate perceptual expectancies or preafference, motor rehearsal, and the articulation of specific behaviors (e.g. Freeman, 1995; Schore, 1997). These appraisals also resonate with the limbic system and brain stem, fashioning a macroscopic feedback loop between emotional and ideational activity (e.g. Tucker, 1992). Because EIs are higher-order forms, arising from preferred patterns of coordination among lower-order cognitive and emotional constituents, they can be represented as attractors on a state space of ‘‘possible’’ psychological states. Perceptual categories (Thelen and Smith, 1994), linguistic categories (Smith, 1995), motor coordinations (Hopkins and


185

Butterworth, 1997), cognitive skills (Van Geert, 1994), memories (Abraham, 1995), belief systems (Goertzel, 1995), and communicative frames (Fogel, 1993) have all been proposed as attractors on a psychological or biological state space. More recently, modal emotions (Scherer, 2000) have been added to the list. We suggest that recurrent emotional interpretations (EIs) are psychological attractors whose unique configuration represents an individual personality. The presence of several attractors, representing several EIs, indicates a range of states (cf. Schore, 1994) to which interpretations can converge for the same individual. Thus, personality is depicted as a cluster of probable cognitive–emotional states, indeterminate and variable in real time but stable or continuous over development (Lewis, 1997). Such a depiction is consistent with Mischel and Shoda’s (1995) definition of personality as an individual cognitive-affective system that produces predictable behaviors in each of several classes of situations. Self-organizing EIs are not merely psychological events; they have specific implications for action. The function of emotion in (human and non-human) animals is to induce, urge, and constrain actions that are designed to alter the physical world – and very often the social world – in ways that are conducive to the achievement of goals (Oatley and Johnson-Laird, 1987; Stein and Trabasso, 1992). Thus, blame urges and supports some form of attack in order to get rid of an obstacle, and the shamed individual is propelled to hide some part of him- or herself in order to avoid rejection. According to the present theory, it is the type of relation between emotion and action that gives rise to three distinct scales of emotional self-organization. At the first scale, microdevelopment, the organization that builds over seconds in an emergent EI rapidly dissipates when goals are achieved through actions. We notice a part of a headline in the newspaper held by the passenger beside us on the subway. We feel interest and excitement, peek over her shoulder, notice that the headline reports nothing out of the ordinary, and go back to our daydreaming. However, when actions cannot achieve goals, then the goal state can persevere, and the organization of interpretation and emotion may persist for minutes, hours, days, or weeks (cf. Scheff, 1987; Teasdale and Barnard, 1993). The development of such prolonged states of cognitiveaffective patterning constitutes the second scale of emotional self-organization – what we have termed the mesodevelopment of moods. We are loosely defining moods as lasting states of cognitive– emotional organization. That moods are characterized by reduced cognitive variance is supported by copious data on mood-congruent perception, learning, and interpretation. Moods also imply a limited range of long-lasting emotions or affects, though not necessarily a

186


single, persistent emotion (Frijda, 1993a). For example, in an angry mood, resentment is recurrent and enduring whereas interest, sadness, and happiness are less probable and shorter lasting. Thus, mood can be depicted as a temporary constriction of the state space, representing a cognitive–emotional bias: attractors are strengthened for some EIs and weakened or absent for others, and trajectories between states are more limited (happy mood may be an exception; see Lewis, 2000a). In moods, not only does action (anticipated or actual) fall short of achieving goals, but action tendencies continually arise as long as an emotional constellation remains in place. Thus, action is both urged and obstructed in moods. This hypothesis of a prolonged state of psychological organization, resulting from backed-up action, is congenial with theorists going back at least to Freud and continues to be verified empirically (Horowitz, 1998; Polivy, 1998). It is also consistent with recent neurobiological accounts. According to Freeman (1995, 2000) the corticolimbic feedback or resonance described earlier gives rise to a global intentional state, beginning in the hippocampus and rapidly self-organizing across the entire brain. Action is initiated by this intentionality, fueled by emotion in the limbic system, and articulated and elaborated in the frontal and prefrontal cortex (Freeman, 1995; Tucker, 1992). Thus, intentions keep the brain organized in the pursuit of goals, and emotions serve as their handmaidens, supporting a perpetual readiness for intended actions. This picture is consistent with Frijda’s (1986) emphasis on states of action readiness as key attributes of emotions. If this picture of brain functioning is accurate, then emotional interpretations (EIs) represent global patterns of corticolimbic coherence manifesting recurrent intentional states. Yet these patterns should be shortlived when EIs give rise to successful actions. Freeman’s research into the perception–action cycle demonstrates electroencephalogram (EEG) coherence during the perception phase, dissipating when action is initiated. This dissipation of coherence leaves the brain in a chaotic ‘‘background’’ state from which new coherences can emerge in the next cycle (Freeman and Baird, 1987; Skarda and Freeman, 1987). But what happens if action is ineffective or impossible? If the goal cannot be abandoned (as is typical of humans, with their advanced symbolic and memory capacities), then the orderliness of intention and emotion would persist. In this state, goal-related behaviors might be rehearsed, expected responses monitored, and contingencies checked and rechecked. Rumination is characterized by these mental activities, and research has demonstrated a reliable correlation between the tendency to ruminate and the endurance of depressive mood states (NolenHoeksema and Morrow, 1991; Teasdale and Barnard, 1993). In neural


187

terms, a lasting state of goal-specific corticolimbic organization would resist switching to novel pursuits (Schore, 1997). Moreover, enduring affective states cause the release of neurohormones whose function it is to commit the cortex to mammalian goals such as aggression, mating, caring, and so forth (Panksepp, 1998). Thus, mood states may harness brain activity to an agenda that cannot be readily satisfied. Instead of peeking at somebody’s newspaper, one may feel outraged by a rude passenger or sexually attracted to another. The intentions that correspond with these feelings – aggression and courting – cannot readily be put into action on the subway. They therefore have the potential to resonate for long periods, often without one’s awareness of their origins or their object. The third and longest scale of emotional self-organization is personality development. In previous work, personality development has been described as the evolution and refinement of a cognitive– emotional state space, whereon attractors for EIs emerge, crystallize, or vanish over the course of months and years (e.g. Lewis and Douglas, 1998). As mentioned earlier, EIs are the real-time manifestation of this developmental process, and the means by which personality is displayed by states of shyness, grumpiness, friendliness, and so forth. Moods constitute temporary modifications to the personality state space, strengthening some attractors and weakening others. But how does the state space consolidate in the first place? In other words, how does personality develop? As mentioned at the beginning of this chapter, a universal consequence of recursive self-organizing processes is their tendency to leave some trace behind them. Wherever water pools in the garden during a rain storm it is more likely to pool in the future. Connection strengths in a neural network are adjusted according to patterns of activation, enhancing the probability that those same patterns will recur. In the brain, Hebbian learning describes the increased likelihood of coactivation of neurons that have been coactivated on a previous occasion. And everyday experience tells us that interpreting a novel event in a certain way increases the chance of interpreting similar events in a similar way. These examples demonstrate a principle of change by which the coupling of elements in a self-organizing system alters the elements, their connections with each other, and their relations with other levels of organization – in short, their complementarities – such that similar patterns of coupling are facilitated in the future. This general principle of accrual and crystallization has several implications for personality development. First, if interpretive organization is always embedded within cognitive–affective patterns, then cognition-emotion couplings are the source of personality structure

188


(Izard, 1984; Malatesta and Wilson, 1988). Second, any converging EI will increase the likelihood of its own recurrence, contributing to personality development through the repetitive incidence of similar interpretations and the ongoing accretion of emotional patterning. Third, EIs that recur frequently in mood states will have a particularly strong impact on personality development. The longer cognitive–emotional couplings resonate and the more often they recur, the more deeply they will become entrenched in underlying complementarities. Moreover, neurohormonal activation in moods extends intentional states, maintains cortical configurations for longer periods, and promotes ‘‘longlasting changes in the strengths and durations of synaptic actions’’ (Freeman, 2000: 226). Moods thus have the greatest potential to guide synaptic learning as well as emotional conditioning over long periods, fashioning permanent modifications to global attractors for development (Panksepp, 2000b). It is often assumed that personality starts to crystallize in early childhood. How might this early consolidation be influenced by recurrent mood states? Young children’s moods have not been systematically studied, but no parent would doubt their ubiquity and power. Schore’s work is particularly relevant here. According to Schore (1997, 2000), neuromodulator action during prolonged affective states maintains enduring cortical configurations. These configurations fashion entrenched habits that gradually supersede the brain’s early plasticity. Moreover, it is the affective context of caring that constrains cortical patterning. Habitual appraisals emerge during caring interactions with a characteristic affective tone, consolidate in the first two years, and assume a hierachical dominance over subsequent cortical organizations (Schore, 1994, 2000). According to Harkness and Tucker (2000), more intense emotions and moods, reexperienced over many occasions, have the strongest effects, as when the recurrence of abuse and neglect early in life ‘‘kindle’’ depression or anxiety with increasing predictability over development. Thus, the synaptic architecture of the brain is literally sculpted during states of prolonged affect in infancy and early childhood. Most important from the present perspective, prolonged affect in mood states is very often negative – a consequence of blocked action and unresolved intentions. These states reflect tensions, losses, and frustrations arising from recurrent patterns of familial interaction but they resonate long after interpersonal exchanges are over. Examples include anxiety about parental rejection, jealousy of siblings, shame and anger following punishment, guilt about harm done to others, and sadness resulting from parental distance or withdrawal. These states


189

reflect goals and needs that are difficult to satisfy (yet difficult to relinquish) given the real contingencies of the interpersonal world. The continuities they hold in mind and brain thus tap continuities in the world. That they also fashion continuities at a longer time scale – that of personality development – expresses the psychoanalytic axiom that personality derives from unresolved wishes, but does so in a scientifically plausible manner. Before going on to a discussion of identity, let us summarize how this theoretical model explains personal continuity in psychological functioning. Specifically, how do processes at longer time scales constrain those at shorter ones? We have already defined moods as modifications or refinements to a previously established (personality) state space. What this means is that the connections laid down in personality (macro)development constrain the possibilities for moods (in mesodevelopment). Hostile people tend to have angry moods and shy people tend to have anxious-inhibited moods. Personality also constrains feeling and thinking in microdevelopment, as indicated by the movement of emotional interpretations (EIs) to personality-specific attractors in real time. However, this effect is necessarily mediated by moods (broadly defined) which constrain microdevelopment more immediately. Thus, it is the nesting of mood in personality that constrains EIs in real time, and their joint effects curtail the variance available for making sense of and feeling about the world. One could say that real-time EIs grow out of system orderliness maintained in moods by an inability to achieve goals in the present, and moods grow out of the orderliness maintained in personality by an inability to achieve goals in the past. Thus, continuity is to be found in real-time interpretations that echo the incomplete resolution of longstanding needs and wishes. This account of continuity does not rely on static structures, rules, or programming. As with all natural systems, it relies on hierarchically nested self-organizing processes, by which the emergent products at each level assemble themselves according to complementarities among their elements and constraints from above and below. A psychological state in real time is always emergent, a product of synaptic activation and corticolimbic resonance. However, the pattern that arises is one of only a few that can arise given complementarities that have evolved over longer time scales. The more constraining the complementarities, the more frequent the recurrence of the pattern, as epitomized by psychopathological states such as paranoia and depression. Yet these complementarities do not comprise a structure or rule: they comprise a contoured state space, a matrix of tendencies. According to the present

190


theory, these tendencies express unfinished business, by which the individual remains preoccupied with an idealized version of an imperfect world. The conclusion is that continuity builds on itself wherever goals are blocked, tension is unresolved, and conflict interrupts the smooth flow of action from intention. This may seem like a new idea, but it mirrors the classic psychoanalytic insight that personality is a byproduct of the stance we take toward a world that habitually obstructs our goals. Personal continuity, viewed in this way, is a compensation for the loss of continuity between intentions and actions, and it expresses the uniquely human capacity to hold onto wishes rather than realities. We are hypothesizing that this principle of continuity underlies the crystallization of personality and its manifestation in moods, emotions, and appraisals, creating an implicit identity or sameness over time. The question that remains is how explicit identity can be illuminated by the same principle. Explicit identity: recurrent appraisals of an idealized self Explicit identity, or personal identity as we define it, relies on the continuity of personality but goes beyond it. Personal identity could not be fashioned and maintained without the interpretive and behavioral continuity resulting from a stable personality system. Yet personality stability is not sufficient to explain identity. Animals and infants have stable personalities but do not demonstrate identities as we understand them. We suggest that personal identity is a specialized product of personality that relies on semantic, reflective, and often conscious activities. We also suggest that this identity is not a static cognitive structure, but rather a lineage of recurring self-appraisals intrinsically linked with emotion (Haviland and Kahlbaugh, 1993). Like other appraisals, they embody cognition–emotion resonances embedded in intentional states. Like other appraisals, they should recur more often in the presence of unsatisfied goals. Unlike other appraisals, they are necessarily about the self and are both autobiographical and dialogical in nature. Thus, we construe identity appraisals as the recounting or enacting of a story of oneself to a real or imagined listener. This story is an explicit account of who one is – an explication of self. Yet, epitomizing appraisals that persist and recur, it suggests a wish or a need to be someone that cannot quite be accomplished, rather than a readout of who one really is. Its stability is situated at a juncture between who one is and who one would like to be, and this tension, this non-resolution of personal goals, is the grounds for its remarkable persistence.


191

Identity as autobiography A number of contemporary models of identity stipulate its autobiographical nature, reflecting a growing emphasis on narrative in developmental, social, and personality psychology (Bruner, 1990; Freeman and Brockmeier, in press; Labouvie-Vief, Orwell, and Manion, 1995; McAdams, 1993). Narratives are key to what most people consider their innermost selves and are integral to how they construe their participation in culture (Bruner, 1996; Lucariello, 1995; McAdams, 1993). Narratives are also considered instrumental in orienting individuals’ interpretations of life events, including choices they make as they strive to construct their own personal identity or to assume the identities that culture constructs for them (Chandler, Lalonde, and Sokol, in press; Ferrari and Mahalingam, 1998; Labouvie-Vief et al., 1995). McAdams et al. (1997) suggest that ‘‘contemporary American adults make sense of their sometimes scattered lives by fashioning and internalizing stories that integrate their reconstructed past, perceived present, and anticipated future’’ (p. 678). From adolescence on, people construct such narratives in order to give their lives a sense of unity and purpose. ‘‘Identity, therefore, may be viewed as an internalized and evolving life story, a way of telling the self to the self and others, through a story or set of stories complete with settings, scenes, characters, plots, and themes’’ (McAdams et al., 1997: 678). Good narratives are about action, and identity narratives concern the actions that are most integral to who we think we are (or wish we were). Connecting these actions into a coherent account establishes identity as an autobiography. Narratives are also powerful goads to action – one’s own action and that of others. Indeed, McAdams et al. (1997) suggest that narratives are highly effective in promoting and supporting generative action. Most people actually develop their stories in the service of action, and not at arm’s length from lived events (Bourdieu, 1997). The future is lived as the ‘‘anticipated present’’ (Bourdieu, 1997; Cole, 1996) as one acts to make one’s narratives actual. These actions, in turn, become the object of further narratives in a recursive cycle of action and interpretation. The ability to anticipate events stems from personal narratives but is also acquired tacitly and historically. It thus constitutes what Bourdieu (1997) refers to as socially inherited cultural capital. This capital is manifested in an enactive narrative (an ongoing interpretation of events) that is experienced as the power and possibility to act in a particular sociocultural setting. For example, narratives associated with gender, ethnic identity, or family serve to frame and orient action.

192


Action in particular cultural contexts, in turn, transforms narratives by enriching and validating them, or by posing problems that must be overcome in order to sustain them. Such encultured action often tailors cognitive development to the narratives in which one participates (Ferrari, 1998; Ferrari and Mahalingam, 1998). Thus, identity can be viewed in terms of autobiographical narratives which concern actions in the past, express cultural ideals, and help to guide and organize actions in the present and future. To rephrase this idea in the language of appraisals and emotions, the actions and events that come to mind in self-appraisals are named, referred to, and woven into autobiographical stories in subsequent appraisals in an ongoing lineage. These autobiographical appraisals, like all appraisals, are coupled with emotions in recurring emotional interpretations (EIs). For example, excitement or hope couple with interpretations of accomplishments, and jealousy or guilt couple with interpretations of failed relationships. Emotions make identity narratives compelling and absorbing and are central to their creation (Haviland and Kahlbaugh, 1993). Moreover, these emotions are instrumental in generating and framing further actions, extending our autobiographies into belief systems and intellectual achievements (Magai and Hunziker, 1993). But this is only half the story. Autobiographical accounts, or recurrent appraisals of oneself and one’s actions, change and grow in detail and nuance but remain highly consistent in theme. This consistency conveys the essence of one’s identity. Moreover, personal consistency is demanded by all cultures and it is a psychological necessity for developing individuals (Chandler et al., in press). How do we explain this consistency in explicit identity? Our account of implicit identity stipulates that only unsuccessful or partially successful actions produce lasting states of psychological organization and only unresolved wishes and goals create lineages of recurrent appraisals. In order to explain the consistency of explicit identity, we should therefore examine not only the actions recalled and promoted by autobiographical accounts, but the very immediate actions of telling these accounts, and we should ask whether these actions successfully resolve the goals which initiated them. To do this, we turn now to the dialogical nature of identity appraisals. The act of telling (or enacting) one’s story is inherently dialogical because it presupposes a listener who will be interested and responsive. Thus, the telling of one’s story, whether to real or imagined listeners, must be evaluated in terms of the interpersonal goals inherent in dialogue.


193

The dialogical nature of identity The idea that personal identity is essentially dialogical is a natural extension of the notion of a narrative self, and it follows from the proposition that self-narratives are collaborative efforts involving one or more listeners (Sarbin, 1993). The dialogues in which self-narratives are embedded may take place with anyone, but they often end up taking place with imagined listeners who have a particular voice and perspective (Bakhtin, 1929/1973). Hermans (1996a) argues that the self takes up various positions in such internal dialogues. Voiced positions, each functioning as a separate person, engage in disagreements, conflicts, and confrontations. Thus, each position is a center of initiative or intention, laying the groundwork for the other voices’ reactions (Hermans, 1996a; Hermans and Kempen, 1993). Not surprisingly, the idea of self-dialogue has a long history. It goes back at least as far as Plato’s claim that thinking involves, ‘‘a discourse that the mind carries on with itself about any subject it is considering’’ (Plato, ~ 390). In terms of personal identity, C. S. Lewis (1967) eloquently stated that ‘‘A person cannot help thinking of himself as, and even feeling himself to be (for certain purposes), two people, one of whom can act upon and observe the other. Thus he pities, loves, admires, hates, despises, rebukes, comforts, examines, masters or is mastered by, ‘himself’’’ (p. 187). To return to the language of appraisal, emotion, and action, all emotional interpretations (EIs) include action readiness, or an emergent plan for action, as well as anticipated feedback from the world – an expectation of what will happen when one actually does take action. Identity appraisals include an emergent plan for telling others about oneself and an expectation of how they will respond. When we actually talk about ourselves to others, these anticipations resolve into real dialogues. But most of the time, the telling remains at the level of rehearsal and imagination, and much of what is ‘‘heard’’ internally is the expected response of the other. As we groom ourselves in the mirror or prepare to teach a class, the following voices might be heard: you look terrific today . . . thanks, I’m actually in great shape for someone my age; you look tired . . . well, I’ve been working too hard lately, not surprisingly considering the demands placed on me . . . he carries so much responsibility, we couldn’t do without him; that was a great point you made in class . . . yes, I’m actually an excellent teacher . . . you’re/he’s much brighter than people realize. These internal dialogues permit a constant editing and elaborating of one’s autobiography. They also fall naturally into the intersection set of one’s own and others’ expectations (Bakhtin, 1929/1973).

194


They are quasi-realistic exchanges contextualized in the interpersonal world (even though we would rarely speak them aloud). For Fogel (1993), such internal dialogues are examples of consensual frames, and they express the coregulation of two (or more) participants who adjust to each other in expectable and familiar ways. Consensual frames self-organize in real time, maintaining their configuration within occasions, but, according to Fogel, they also recur across occasions, establishing a developmental pattern or lineage. This is certainly the case with personal identity. In our view, each voiced position in an internal dialogue or frame corresponds with a global intention that entrains cognition–emotion. This means that others’ voices are indeed one’s own, at least for the moment it takes to formulate them. But why are these voices, and the intentions they serve, so consistent across occasions? We can begin to answer this question by examining the position of the ‘‘I’’ and the ‘‘you.’’ The ‘‘I’’ consistently intends to put forth a reasonable story of someone who will be accepted, esteemed, or liked. This is no easy job, because the ‘‘you’’ is often critical rather than sympathetic, or simply non-committal. Thus, wishes to be accepted, admired, and liked, to be close, to avoid rejection, and even to hide unattractive parts of the self cannot be fully or permanently realized, because other people in the world and in imagination are not always accommodating. Yet these wishes are what give rise to internal (and external) dialogues repeatedly, because they propagate the goal of ‘‘explicating’’ oneself in the first place. It may be the very difficulty in achieving these intentions that entrenches them and extends their orderliness over time. Thus, the goal of narrative in dialogue is never quite fulfilled, either internally or externally, because the wish for closeness, acceptance, and admiration is never completely satisfied. There may be another crucial obstruction to the dialogical goals for self-narrative: it is nearly impossible to fashion an account of the self that is coherent and unified given the disparity of positions and voices (Bruner and Kalmar, 1998; McAdams et al., 1997). As a result, the intention to unify what is so obviously disparate cannot be fully realized, and the justification of discrepancies among our various selves is elaborated and extended over many recurring efforts. Narratives are communicative acts with a particular listener in mind. The same story requires major adaptation to appeal to different audiences, who value different things and judge one’s stories (including one’s explanations, justifications, and plans) by different criteria (Ferrari and Mahalingham, 1998). It is commonplace to tell conflicting stories in different contexts (Cooper et al., 1998; Hermans, 1996a), and we do not always voice a particular narrative out loud (Bruner and Kalmar, 1998). Thus,


195

the demand for narrative consistency, highlighted by McAdams et al. (1997), clashes with the need to please different audiences with different accounts of the same actions. The different voices in Hermans’ (1996a) dialogical melee cannot all be satisfied with the same narrative, yet the actions being explained cannot be completely rewritten. Aggressive actions and intentions concerning one’s younger brother must be hidden, or at least disguised, when recounting them to one’s mother, but amplified when recounting them to the other boys in the playground. We suggest that the need to reconcile these accounts constitutes an unrealizable agenda: one that cannot be satisfied through action and which therefore maintains the orderliness of cognition–emotion coupling and behavioral rehearsal. This claim is not far removed from the psychoanalytic idea that tension between wishes and justifications is the crucible of psychological life. The backing up of orderliness resulting from this impasse would then give rise to a richly articulated, historical, and continuous account of an idealized, justified self. However, the explication of self can succeed, as when we admit to the contradictions in our stories and the listener accepts the many facets of who we are. If our model is correct, then the achievement of this dialogical goal ought to dispel explicit identity – at least for a while. As evidence for this prediction, consider the experience of identity when falling in love. It is no accident that the feeling of being completely accepted by another person corresponds with the sense that the self is no longer finite, unitary, describable, or even known. A similar experience may derive from psychotherapy. These ‘‘self-actualizing’’ experiences free us from carrying around a story of ourselves in mind and arguing its validity to a real or imagined respondent. However, the loss of identity may come about much more drastically when meaningful dialogue with real or imagined listeners is truncated by developmental or cultural upheaval. When the dialogue of self-definition comes to a halt, loss of identity can lead to suicide, presumably because the future has no relevance and the present is too painful (Chandler and Ball, 1989). Identity change Given our explanation for continuity of personal identity, how does this identity change? Early in this chapter we reviewed three kinds of change in self-organizing systems: (1) the emergence of novel, higherorder forms that replace their precursors, (2) the entrenchment of particular habits to the exclusion of others, and (3) the abrupt shifting between stable patterns denoted by phase transitions. Identity change

196


in development can be modeled in all three ways, but first we briefly consider identity change in real time. Identity change in real time consists of switching from one story to another, or shifting dialogical frames (Fogel and Lyra, 1997), as implied by the previous discussion. We will not discuss this type of change in detail, but Fogel and his colleagues have thoroughly researched the qualitative distinctions between frames and the discontinuity seen in behavior when dyads switch from one to another (Fogel and Lyra, 1997). The DS notion of phase transitions seems to capture the abrupt and global nature of such change, but research on this topic is still pending (Fogel, 1993: 116). For the present, we note that each of these stories or frames would anchor identity only if it had a historical lineage. Thus, each identity story (as a robust and recurrent EI) would require a cognitive–emotional complementarity that had arisen over many occasions, expressing concerns and wishes intrinsic to one’s personality. This is one obvious way in which real-time identity appraisals makes use of longstanding personality patterns. Change in identity across developmental time would involve change to one or more of these narrative lineages or, importantly, to the intentions with which one recounts and integrates them in dialogue. The three kinds of change each suggests different directions for modeling. First, new identity constellations (either stories or meta-stories) can be viewed as novel, higher-order assemblies that cohere from emotiondriven coordinations among the cognitive constituents of emotional interpretations (EIs) in general and previous stories and dialogues in particular. This general mechanism of self-organization would be responsible for the consolidation of identity in the first place, probably in the second to fourth year of life. It may also be a good way to conceptualize subsequent identity change, especially when new identity patterns start off inchoate and unstable and consolidate over many emotionally significant occasions. For example, in adolescent transitions, needs for autonomy and intimacy can be brought to bear on one’s autobiographical stories, forcing individual narratives to change in order to maintain authenticity (Harter, 1998; Mascolo and Fischer, 1998). Yet, it takes many repeated dialogical appraisals – justifying actions and mending discrepancies – before those narratives become coherent and consistent. Marriage also brings about identity change, sometimes driven by the intention to have children and to place them before one’s own interests. Again, narratives of oneself as a parent take time to self-organize, and they do so by assembling emotionally meaningful constellations of narrative constituents already present in one’s personality or one’s culture.


197

In the second kind of systemic change, recursion strengthens new habits such that they supersede competing tendencies. The retelling of newly fashioned autobiographical accounts to real and imagined listeners is therefore a likely mechanism for identity change. These accounts may initally be imported from peers or the media but then retold repeatedly in one’s imaginary dialogues. Elkind (1974) has emphasized the impact of the ‘‘imaginary audience’’ on the development of adolescent identity. Moreover, actions would seem to be particularly powerful instruments of identity change. We have already suggested that identity narratives both make sense of actions and promote new actions in an ongoing cycle of recursive interpretations. Such recursion would rapidly strengthen complementarities that support one’s favoured story-lines. Finally, new narratives are likely to be problematic, emotional, and unfinished because of their inconsistency with previous accounts. They are thus well suited to capture psychological organization for hours and days in mesodevelopment and to modify personality itself over macrodevelopment. In this way, identity change may provide a leading edge to personality development while at the same time being shaped and constrained by underlying personality complementarities. Finally, the idea of phase transitions, now at the scale of development, may be useful for modeling identity change. Life stories are constantly under revision (Freeman and Brockmeier, in press), but some actions or events act as pivot points in people’s life stories. Dramatic life events can serve as junctures in one’s autobiography, specifically when one story is abandoned and another is begun (Bruner and Kalmar, 1998). These pivotal changes can be modeled as phase transitions, and the events that elicit them as triggers. As discussed earlier, the nature of such changes is determined by the attributes of the system, not the attributes of the triggering events, a point also emphasized by Magai and Nusbaum (1996). Changes in personal identity may also be revealed by changed emotions (Haviland and Kalbaugh, 1993) and emotionally powerful life events have been shown to precipitate identity change (Magai and Nusbaum, 1996). A temporary period of fluctuation at a point of rapid developmental change is the necessary criterion for phase transitions (Van Geert, 1994). Indeed, Magai and Nusbaum (1996) integrate a number of lines of evidence indicating that personal identity and self-concept change abruptly, even explosively, while underlying personality dispositions endure; but personality and identity are not clearly distinguished in their account. It is possible that identity change is more precipitous than personality change, but further research is needed to decide this issue.

198


Conclusion Early in this chapter we reviewed a newly formulated theory of emotional self-organization at three time scales. According to this theory, real-time interpretations, mood-like states, and crystallizing personality dispositions were linked through principles of psychological entrenchment related to the non-resolution of longstanding goals and wishes. Personal identity was then modeled as a specialized product of personality that relies on reflective and often conscious activities that are inherently dialogical and autobiographical. We described identity as a lineage of real-time self-appraisals that generally retains continuity yet is subject to periodic transitions. These appraisals revise and extend narrative accounts of who we are, or who we would like to be, in interpersonal, dialogical contexts. As a final word on the relation between personality and identity, we suggest that identity not only depends on personality as grist for its mill; it also guides the direction of personality change and consolidation, at least in the years following infancy. It thus exists in a mutually causal relation to both personality continuity and personality change. These ideas are consistent with a new family of models that apply principles of self-organizing dynamic systems to psychological processes in general and socioemotional processes in particular. However, much of the present framework is still untested and may need substantial revision before it is authenticated. We present it here, speculatively, in order to demonstrate how self-organization can explain not only developmental change, but also continuity or sameness in cognitive– emotional processes. Some theorists suggest that all dynamical systems have an implicit memory – that is, an accretion of identity – resulting from subtle changes to their elements and their connections when engaged in self-organizing processes (Schwartz and Russek, 1997). This provocative hypothesis will take years to explore fully. However, if true, it would suggest that both implicit and explicit identity express tendencies that extend well beyond human psychology to all natural phenomena.

COM M E N T A R Y

Two faces of identity Carol Magai

Marc Lewis, who was one of the first to apply dynamic systems constructs to an explication of personality development, and who did so in a very exciting way, provides us in the present chapter with more stimulating ideas and further food for thought. Here Lewis and Ferrari distinguish between implicit identity, which refers to personality characteristics, and explicit identity, which is a ‘‘specialized product’’ of the personality system and is self-reflective and dialogic. Both of these ‘‘identities’’ are subject to change as modeled in a dynamic systems framework. This is an interesting and potentially useful distinction; I would like to explore a comparison between these two aspects a bit further. It seems to me that implicit and explicit identity may be differentially stable and differentially permeable to change. An examination of the literatures on ‘‘personality’’ and ‘‘self systems’’ suggests that implicit identity or personality may be more stable than explicit identity or self identity. According to a number of accounts, personality is in large measure grounded in emotional dispositions that have roots in temperament (McCrae and Costa, 1996; Goldsmith, 1994), though regularities and repetitive emotional experiences triggered by environmental conditions also shape dispositions (Malatesta, 1990). In any event, whether or not emotional dispositions of particular individuals accrue largely from temperamental substrates or from repetitive emotional events during early development, it is clear that mood states, which are products of personality, are fairly stable by the time individuals reach adulthood. The literature indicates that stability coefficients for positive and negative affect are quite high. For example, Deiner and Larsen (1994) report that the correlation between positive affect scores over a one-year period was .74; similarly, Ormel and Schaufeli (1991) found the correlation for negative affect over seven years was .57. It seems more likely that explicit identity, given its dialogic nature, might be more susceptible to change, though one suspects that the core emotional dispositions would retain their overriding influence. A number of individuals have noted that even people who appear to undergo a striking change in personality retain important components 199

200

Carol Magai

of their personality. For example, David McClelland (1984) recounted a chance encounter with Richard Alpert, some twenty-five years after their short acquaintance when Alpert was a graduate student studying psychology under him. McClelland describes what at first blush appeared to be a major personality transformation – one that transformed a young, highly intelligent, and charasmatic Harvard assistant professor into Ram Das, a servant of the Lord, under the influence of the Indian guru, Maharaji-ji. Alpert’s whole appearance had changed, as had his attitudes, career, and purpose in life; his intellectual and spiritual inspiration shifted from neo-Freudianism to the eastern psychologies of Buddhism and Hinduism. One rarely sees such striking change, and yet, as McClelland pointed out, fundamental aspects of Alpert’s personality could still be perceived. He still had the same expressive movements, he still had a strong interest in internal psychic states, he was still charming, he was still very much involved in power games, and still felt guilty about being interested in power. This suggests that although Alpert’s explicit identity had changed substantially, there had been far less change in implicit identity. It is these kinds of observations as well as a growing research literature on self-concept that makes us question how the two ‘‘identities’’ relate to one another. As the editors of this volume have noted in their introductory chapter, the construct of self-concept (Lewis and Ferrari’s explicit identity) has undergone significant change within the social and personality literature in recent times. Indeed, within this literature there has been a movement away from viewing the self-concept as a unified, stable, integrated entity, to one that involves seeing the self as a representational system that is large, complex, dynamic, and subject to change. The self is now seen as a cognitively complex entity constituted of self-schemas, which are active, functional structures shaping a person’s emotional and behavioral responses to events (Aron, Aron, Tudor, and Nelson, 1991; Markus and Kitayama, 1991). In a recently completed study (Magai, 1999), we attempted to examine the stability of these two aspects of personality in the same people over an eight-year period. In effect, we were studying both implicit and explicit identity, though we referred to the two aspects of personality as dispositional tendencies (emotion traits) and characteristic adaptations (views of self). Stability coefficients for the emotion traits of anxiety, depression, interest, anger, anger-in, anger-out, and aggression ranged from .47 to .75, with an average of .62; only anger-out showed significant change over the eight years. These findings are in accord with other research, noted earlier, on the robust nature of individual differences in positive and negative affect. On the other hand, our respondents reported moderate changes in perspectives, goals, personality,

Commentary on Lewis and Ferrari

201

feelings, and ways of relating (an average score of 2.9 on a scale of 1 = not at all, to 5 = extremely). Moreover, ratings by family and friends of the informants were significantly correlated with self-reports of change for all variables except goals. Another interesting finding was that this kind of personality change was associated with positive and negative interpersonal life events of an intimate nature, such as marriage, divorce, and death of loved ones that took place during the eight years; moreover, it was not associated with other high and low points in peoples’ lives involving careers, changes in residence, and more distant social relationships. This suggests that interactions with others have a substantial impact on aspects of the self that relate to self-concept and motives, which would seem to support Lewis and Ferrari’s idea that explicit identity is dialogic in nature, and perhaps may even represent an interiorization of dialogues with others. In fact, elsewhere we (Magai and McFadden, 1995; Magai and Nusbaum, 1996) have argued that although intense emotion of a positive or negative nature is a necessary condition of personality change, change is unlikely to be sustained in the absence of interpersonal support and mental elaboration. Lewis and Ferrari suggest that the same three dynamic system principles may govern change in explicit and implicit identity. Given their apparent differential permeability to change it is possible that the underlying dynamic principles may be less similar than would at first appear to be the case. It might help to adopt an ideographic approach to the issue, say contrasting cases that exemplify dramatic, structural change involving change in emotion traits, as in the case of posttraumatic stress disorder, and cases of explicit identity change (as in the case of Richard Alpert), to begin to model these different kinds of identity change (Magai and Haviland-Jones, in press).

COM M E N T A R Y

Two faces of identity Carol Magai

Marc Lewis, who was one of the first to apply dynamic systems constructs to an explication of personality development, and who did so in a very exciting way, provides us in the present chapter with more stimulating ideas and further food for thought. Here Lewis and Ferrari distinguish between implicit identity, which refers to personality characteristics, and explicit identity, which is a ‘‘specialized product’’ of the personality system and is self-reflective and dialogic. Both of these ‘‘identities’’ are subject to change as modeled in a dynamic systems framework. This is an interesting and potentially useful distinction; I would like to explore a comparison between these two aspects a bit further. It seems to me that implicit and explicit identity may be differentially stable and differentially permeable to change. An examination of the literatures on ‘‘personality’’ and ‘‘self systems’’ suggests that implicit identity or personality may be more stable than explicit identity or self identity. According to a number of accounts, personality is in large measure grounded in emotional dispositions that have roots in temperament (McCrae and Costa, 1996; Goldsmith, 1994), though regularities and repetitive emotional experiences triggered by environmental conditions also shape dispositions (Malatesta, 1990). In any event, whether or not emotional dispositions of particular individuals accrue largely from temperamental substrates or from repetitive emotional events during early development, it is clear that mood states, which are products of personality, are fairly stable by the time individuals reach adulthood. The literature indicates that stability coefficients for positive and negative affect are quite high. For example, Deiner and Larsen (1994) report that the correlation between positive affect scores over a one-year period was .74; similarly, Ormel and Schaufeli (1991) found the correlation for negative affect over seven years was .57. It seems more likely that explicit identity, given its dialogic nature, might be more susceptible to change, though one suspects that the core emotional dispositions would retain their overriding influence. A number of individuals have noted that even people who appear to undergo a striking change in personality retain important components 199

200

Carol Magai

of their personality. For example, David McClelland (1984) recounted a chance encounter with Richard Alpert, some twenty-five years after their short acquaintance when Alpert was a graduate student studying psychology under him. McClelland describes what at first blush appeared to be a major personality transformation – one that transformed a young, highly intelligent, and charasmatic Harvard assistant professor into Ram Das, a servant of the Lord, under the influence of the Indian guru, Maharaji-ji. Alpert’s whole appearance had changed, as had his attitudes, career, and purpose in life; his intellectual and spiritual inspiration shifted from neo-Freudianism to the eastern psychologies of Buddhism and Hinduism. One rarely sees such striking change, and yet, as McClelland pointed out, fundamental aspects of Alpert’s personality could still be perceived. He still had the same expressive movements, he still had a strong interest in internal psychic states, he was still charming, he was still very much involved in power games, and still felt guilty about being interested in power. This suggests that although Alpert’s explicit identity had changed substantially, there had been far less change in implicit identity. It is these kinds of observations as well as a growing research literature on self-concept that makes us question how the two ‘‘identities’’ relate to one another. As the editors of this volume have noted in their introductory chapter, the construct of self-concept (Lewis and Ferrari’s explicit identity) has undergone significant change within the social and personality literature in recent times. Indeed, within this literature there has been a movement away from viewing the self-concept as a unified, stable, integrated entity, to one that involves seeing the self as a representational system that is large, complex, dynamic, and subject to change. The self is now seen as a cognitively complex entity constituted of self-schemas, which are active, functional structures shaping a person’s emotional and behavioral responses to events (Aron, Aron, Tudor, and Nelson, 1991; Markus and Kitayama, 1991). In a recently completed study (Magai, 1999), we attempted to examine the stability of these two aspects of personality in the same people over an eight-year period. In effect, we were studying both implicit and explicit identity, though we referred to the two aspects of personality as dispositional tendencies (emotion traits) and characteristic adaptations (views of self). Stability coefficients for the emotion traits of anxiety, depression, interest, anger, anger-in, anger-out, and aggression ranged from .47 to .75, with an average of .62; only anger-out showed significant change over the eight years. These findings are in accord with other research, noted earlier, on the robust nature of individual differences in positive and negative affect. On the other hand, our respondents reported moderate changes in perspectives, goals, personality,

Commentary on Lewis and Ferrari

201

feelings, and ways of relating (an average score of 2.9 on a scale of 1 = not at all, to 5 = extremely). Moreover, ratings by family and friends of the informants were significantly correlated with self-reports of change for all variables except goals. Another interesting finding was that this kind of personality change was associated with positive and negative interpersonal life events of an intimate nature, such as marriage, divorce, and death of loved ones that took place during the eight years; moreover, it was not associated with other high and low points in peoples’ lives involving careers, changes in residence, and more distant social relationships. This suggests that interactions with others have a substantial impact on aspects of the self that relate to self-concept and motives, which would seem to support Lewis and Ferrari’s idea that explicit identity is dialogic in nature, and perhaps may even represent an interiorization of dialogues with others. In fact, elsewhere we (Magai and McFadden, 1995; Magai and Nusbaum, 1996) have argued that although intense emotion of a positive or negative nature is a necessary condition of personality change, change is unlikely to be sustained in the absence of interpersonal support and mental elaboration. Lewis and Ferrari suggest that the same three dynamic system principles may govern change in explicit and implicit identity. Given their apparent differential permeability to change it is possible that the underlying dynamic principles may be less similar than would at first appear to be the case. It might help to adopt an ideographic approach to the issue, say contrasting cases that exemplify dramatic, structural change involving change in emotion traits, as in the case of posttraumatic stress disorder, and cases of explicit identity change (as in the case of Richard Alpert), to begin to model these different kinds of identity change (Magai and Haviland-Jones, in press).

C HA P T E R 9

A self-organizational approach to identity and emotions: an overview and implications E. Saskia Kunnen, Harke A. Bosma, Cor P. M. Van Halen, and Matty Van der Meulen Introduction The perceptions, ideas, and feelings we hold of ourselves, whether consciously or not, are a natural, intrinsic part of our daily functioning. They seem a necessary ingredient of our attempts to sustain a basic sense of identity, or as Erikson (1950) described it, to protect a sense of personal continuity, unity, and social recognition. As such, they make us an individual in the eyes of others as well as ourselves. But they are also the orientating instruments by which we try to bring some coherence to our own life or to set out new directions. For psychologists, self-referential processes retain an intriguing, though somewhat elusive character, which makes it difficult to capture them in a single, neat conceptual and empirical framework. In a sense, the content of such processes seems to typify the person one has grown to be. Yet, in another sense, they constitute the foreshadowing of the person one could become. Moreover, in both cases, as James explained, these processes introduce an element of recursivity in the way we experience ourselves, without totally merging with ongoing mental, behavioral, and social processes. In chapter 1, we stated that in past decades psychologists have tried to deal with the indefinite status of such generic concepts as self and identity by exclusively focusing on the more tangible aspects. This generally meant that self and identity were equated with the selfconcept, which was seen as a relatively stable, generalized set of selfrepresentations that people have formed during their lives. This opened the way to the still predominant empirical practice of using self-descriptions and self-ratings as reliable indications of an underlying self-concept. Thus, the more static, decontextualized, and cognitive features were stressed at the expense of the more dynamic, contextual, and emotional qualities. A general dissatisfaction with the far from spectacular results of traditional self-concept research (e.g. Wylie, 1979) eventually forced 202

A self-organizational approach to identity and emotions

203

self-psychologists to venture in new directions. Van der Meulen’s review of recent developments in self-concept theory and research in chapter 2 provides an extensive discussion of the different attempts to compensate for the lack of attention to affect, context, and variability. She described how the various reconceptualizations and ensuing empirical research to date have shown that the beliefs persons have of themselves are closely associated with affect, and are a mixture of more generalized, enduring self-attributions and more situational and temporal ones. Unfortunately, when taken together, the empirical findings constitute a rather fragmentary picture of partial aspects, without the promise of an adequately integrated framework. Moreover, these new approaches remain rooted in a view of self and identity as a cognitive, internal structure that analytically and empirically stands apart from the accompanying emotional and contextual processes. The authors represented in this volume propose a radically new perspective on self and identity, theoretically as well as empirically. The central notion in this approach is the dynamics of self-organization, emerging from the reciprocity of cognitive, emotional, and contextual processes. More specifically, the following key issues are addressed. First, in what way(s) do emotional and cognitive processes implicate each other within the dynamics of self and identity? Second, how do these processes in interaction take shape within the contexts in which one is functioning? Third, how do seemingly contradictory characteristics such as stability and variability combine in self-organizing processes? While such issues can be posed in simple and straightforward terms, answering them is a totally different matter. This is reflected in the contributions to this volume. In dealing with the complexity of the dynamics involved in self and identity, each author has developed her or his own vantage point and explanatory devices. Hence, we can imagine that a first reading may have resulted in some confusion about the underlying divergences and similarities between the different contributions. In this final chapter, we will, therefore, start with a brief overview of the chapters in order to identity both the underlying assumptions and important future issues.

A concise overview Setting the stage for a dynamic perspective In disputing the exclusively cognitive connotation of self and identity at the outset of this volume, it seemed desirable to approach questions about the role of emotions from the standpoint of emotion theory. This was the task Frijda fulfilled in chapter 3. Basic principles in Frijda’s

204


(1986) emotion theory are that emotions indicate what matters to a person, and that they motivate human behavior. One only has to reverse this line of reasoning to come to the conclusion that if self and identity do matter and motivate human behavior they should logically include emotions. This subsequently raises the question in what ways, according to this theory, self and emotions may be related. In none of the emotion components that Frijda distinguishes is the self a necessary ingredient, at least not in its most common meaning as explicit selfawareness or self-representation. The only prerequisite is what Frijda calls the ‘‘transcendental self’’: the logical reference point of any experience (including emotions). The transcendental self has no mental content beyond being the locus of a vague sense of familiarity and initiative. In the remainder of his analysis, Frijda actually turns things around. Though aspects of self and identity may not be prerequisites of emotion, it can be shown that, if properly specified, different aspects of self and identity are implicated by emotions. First, there are many emotions in which the self figures as an object, either in becoming aware of undergoing a particular emotion (e.g. ‘‘I feel fear’’) or as the referent of an emotional appraisal (e.g. ‘‘I hate myself’’ or ‘‘I feel proud of myself’’). Such self-representations and self-evaluations are often involved in the more adult forms of shame, guilt, and pride. Besides, one’s own emotions can become objects of emotional appraisal, because of the personal and social implications of having or showing these emotions. This is the ‘‘significance’’ of an emotion, which motivates the further regulation of emotions. The significance often involves self-representations such as ‘‘a Friesian doesn’t cry,’’ or ‘‘I’m an anxious person.’’ Second, a sense of self may also be considered a possible factor in estimating one’s own coping resources. The perceived competence in dealing with the emotion-eliciting event is one of the determinants in the coming about of an emotion, in addition to the appraised event and the individual’s concerns. However, such a sense of competence is entirely implicit in the appraisal of the outcome of an emotional event, and no self-representation or conscious self-evaluation needs to be assumed here. The most pervasive sense in which the self is implicated in emotions is in the form of concerns, of ‘‘what and how the subject is’’ (Frijda, chapter 3, p. 54). ‘‘Concern’’ is a broad term for the individual’s motives, goals, attachments, aversions, etc., in other words all that matters to the person. Sometimes, the term ‘‘self’’ is used to denote these personal concerns. Persons can have concerns about themselves, for example the wish to achieve and maintain a certain identity. According to Erikson (1950), the need for a sense of identity is one of the most


205

fundamental human concerns. Frijda (1986) himself mentions selfesteem as one of the most basic concerns. These concerns, in turn, can lead, in the course of the individual’s development, to the more complete forms of self-conscious emotions (e.g. pride, shame). Summarizing, Frijda’s chapter provides several pertinent contributions to the central issues in this volume. Though self-representations may not be a necessary ingredient of emotions, emotions do play a major part in self and identity. Most importantly, they indicate what is essential to the person: his or her concerns and competencies. The need for a sense of identity and self-esteem are among the most fundamental human concerns. Moreover, by integrating emotions into the study of self and identity, a direct link between the individual and the context is guaranteed. Because emotions are fundamentally relational, fusing the subject and the object of an emotion, this integration can resolve the problems inherent in the decontextualized nature of selfrepresentations research. The conceptualization of emotion as a system of loosely coupled affective, cognitive, and behavioral components (see also Frijda, 1993) offers new opportunities for an understanding of how emotions and self may be intertwined. It also preludes a dynamic systems perspective, because the mutual interactions between the different emotion components and aspects of self are the defining elements in the eventual course of the emotion processes. In his critical commentary, Epstein elaborates on the relation between emotions and self-aspects in a way that to an important extent goes along with Frijda’s componential and dynamic view. However, he relegates much of Frijda’s emotion theory to the so-called ‘‘experiential system.’’ He describes the experiential system as the executive part of the self, and as something that has to be clearly separated from the self as the object of rational considerations. The system operates on the basis of an intuitive theory about oneself, about the world, and about their interrelationship. In this system, emotions indicate what is of significance to a person. Frijda and Epstein also agree that the dynamic interactions between implicit appraisals and emotions are largely nonconscious, but both describe the dynamics involved only in global terms. In this connection, the application of dynamic systems theory might be very useful. Chapter 4 by Van Geert is intended to be a general introduction to dynamic systems thinking. Since many contributors to this volume use or advocate a dynamic systems perspective on issues of self, identity, emotions, and development, this chapter offers the necessary background information for a preliminary evaluation of the usefulness of such a perspective. A dynamic systems perspective offers insight into processes, and thereby into questions concerning stability and change.

206


Van Geert describes a dynamic system as something that ‘‘consists of elements that exert specific influences or forces upon one another and by doing so, change each other’s and their own properties’’ (chapter 4, p. 64). In their commentary, Camras and Michel correctly point to one of the critical implications of this definition. In applying a dynamic systems perspective, one should define constructs in terms of systems, that is, in terms of a conglomerate of elements (e.g. prey and predator fish). In this volume, therefore, the question has to be posed as to what conglomerate of elements the generic terms ‘‘identity’’ or ‘‘self’’ refer to. It should also become clear what is part of the system and what is part of the system’s environment. Indeed, most authors in this volume have in one way or another, implicitly or explicitly, tried to deal with these implications. An important view of dynamic systems theory is that in order to explain variation over time, that is patterns of stability and change, it is not necessary to resort to external causal forces, or an inbuilt, deterministic ground plan. Instead, the most conspicuous characteristic of dynamic systems is self-organization. Complex, non-linear systems tend to self-organize into a state of increased order/complexity. Higherorder phenomena, for instance self-constructs or identity, can be seen as emerging from the dynamics of the system rather than being already present as internal structures. In dynamic systems terms, changes result from the time-dependent interactions between the constituent components of the system. These interactions – the set of relationships between the components of the system – form the core of a dynamic systems model, and not the physical instantiation of the components (e.g. molecules, fish, classroom reprimands, self-representations, identity choices). This means that the general process characteristics of dynamic systems, such as different types of attractors and bifurcations, can be applied to describe the dynamics in a specific domain such as self and identity. At a qualitative level, this can have an important heuristic value, but, as Van Geert has demonstrated, the relationships between the components can also be specified in mathematical functions which are surprisingly simple but which nevertheless capture the essence of the system’s behavior over time. Four elaborations on a theme The chapters by Frijda and by Van Geert have informed us about the role of self-phenomena in emotions, and about the application of a dynamic systems perspective in order to approach questions concerning stability and change. Chapters 5 through 8 provide in-depth discussions of emotions and context as formative conditions from which


207

self and identity emerge in a self-organizational process. However, each of these chapters has a different focus. Nevertheless, in certain ways, they also complement each other and build on the foundations laid by Frijda and Van Geert. Frijda makes it clear that one’s self can become the object of emotions. In order to experience oneself as an object of an emotion, some notion of oneself as enduring and stable over time must exist. But, when and how, in ontogeny, do these self-related phenomena emerge? These are the kinds of questions Fogel explored in chapter 5. Fogel based his approach on two premises: psychological experience (a) always implies a relationship (with the object-world, others, or oneself) and (b) is always dynamic and changing. Within this inherently relational and dynamic world, the crucial question then becomes how people can ever develop a sense of uniqueness and continuity.1 Fogel considers emotions as playing a central role in this process, because they are a way of discovering the meaning of a relationship for the self and, hence, the unique position of the self in a relationship. The perception of consistency in one’s emotional experience begins to yield a sense of permanence of the self over time. With these premises and his almost programmatic statements, Fogel occupies a distinctive position in the context of this volume: all experience is dynamic, self is rooted in relationships, and emotions (which in themselves are inherently contextual) play a constitutive role. Central to Fogel’s analysis of the dynamics of self and identity is his distinction between three time scales, and the observation that, depending on the time scale, different types of emotion go together with different experiences of uniqueness. A first and basic sense of self already comes into existence when the emotional experience that accompanies a currently ongoing relationship stabilizes into a primary orientation towards the relationship. This takes place on the time scale of (micro)seconds. In turn, orientations coalesce into identifiable events. On the time scale of minutes to days, sequences of events may selforganize into coherent and stable frames. Frames are similar to narrative scripts and enable individuals to experience a more explicit, stable sense of self. The third time scale, the scale of ‘‘development,’’ extends over years. The corresponding sense of self is identity: the experience of an enduring self. It is rooted in the relation with one’s past or future, but also in the surrounding cultural frames. Although identity may develop into a narrative account of oneself, Fogel considers it to be an emotional experience rather than a representation. Regarding the main questions considered in this volume, Frijda and Fogel have much in common. Both argue for a differentiated conceptualization of self and identity, and both agree that emotions are

208


relational experiences. In a certain sense, they also complement each other. Frijda makes clear that self-related phenomena are not a necessary part of the emotional experience. Fogel gives a similar primacy to emotions but seeks to understand how aspects of the sense of self emerge from emotional experience in relationships. In our opinion, Fogel’s description of the emergence of a sense of self on three different but interrelated time scales offers a significant and elegant contribution to our understanding of the constitutive role of emotions in the development of aspects of self. It is interesting to note that Fogel hardly refers to explicit representational content. Apparently, he, like Frijda, considers self-representations and self-awareness to be unnecessary in understanding the way in which the self becomes organized. In his commentary on Fogel, Jansz suggests an alternative explanation for the emergence of a sense of identity. In contrast to Fogel, he sees identity and its constituent feelings of continuity and consistency mainly as a social construction that people have learned to ascribe to themselves. In this they are guided by the prevailing notion of personhood, which in Western societies is characterized by an emphasis on personal consistency, uniqueness, autonomy, and rationality. In his reply to Jansz (not published) Fogel states that the difference between his and Jansz’s view stems from a difference in starting point. In Fogel’s view structures have to be explained, whereas relations need no explanation. Jansz starts from structures and has to explain relations. Thus, Jansz emphasizes the role of cultural scripts in formatting identity through dialogical exchange, and the cultural fact that emotions count as a violation of the aspired image of equanimity. Fogel, on the other hand, concentrates on the stabilizing role of emotions in the emergence of a clear sense of identity within dialogical relationships. Both ways of relating emotions to the dialogical nature of identity return in chapter 6. Hermans and Hermans-Jansen combine Fogel’s relational perspective with culturally determined patterns of meaningmaking by introducing a new element: a strong recognition of the developmental possibilities that are brought about by the human faculty for self-reflection. They start from the assumption that in the course of human development emotional responses are increasingly influenced by the relationship individuals have with themselves. In terms of the differentiation Frijda makes, Hermans and HermansJansen focus on experiences in which the subject becomes the object of the emotion, and in particular on the significance of the emotion. The significance plays an important role in emotion regulation, in the way individuals interpret and express, or do not express, their emotions. This is also the domain in which the culturally determined emotion scripts figure.


209

In their analysis of the dynamics and complexity of the relation between self and affect, Hermans and Hermans-Jansen elaborate on James’ distinction between ‘‘I,’’ i.e. the self-as-knower, and ‘‘Me,’’ i.e. the self-as-known. They see self-awareness as an essential feature of the I. Individuals can reflect on their feelings, appraise them, and develop feelings about these feelings. Thus, the self authors dynamic and complex compositions of personally significant, affect-laden meanings (i.e. valuations). Such narratives typically take the shape of an ongoing self-dialogue. Instead of being the omnipotent author of oneself, the self-reflecting individual continuously changes perspectives. Each perspective in fact implies a different I-position, entailing its own internalized voice and telling its own story. The different narrative positions usually revolve around recurring conflicts and obstacles one encounters in becoming the person one wants to be or should be. Thus, self-narratives have developmental implications. To serve as a fruitful point of departure for further identity development, the inner voices that speak for the own person should be brought into open dialogue with each other. The notion of the ‘‘dialogical self’’ returns in the valuation theory and the self-confrontation method, which can be employed to study the dialogical self empirically. Both in its conceptual connotations and in its methodology, the approach of the dialogical self has much in common with the phenomenological–dialectical personality model of Verhofstadt-Dene`ve, that is used as starting point in the application of psychodrama techniques. The authors of chapter 7, Haviland-Jones, Boulifard, and Magai, present a similar componential conceptualization of the self. By the self they mean self-concept, identity, or personality, terms which they use more or less synonymously. They take emotions as a starting point and study how emotions organize personality, in contrast to Hermans and Hermans-Jansen, who argue that I-positions organize patterns of emotions. Haviland-Jones and her colleagues give a lively account of their quest for tools that enable them to arrive at a better understanding of the dynamic interplay of cognitive and affective components of the self. They suggest that dynamic systems thinking and modeling offer exciting new possibilities. Their earlier studies failed to provide adequate insight into the role of emotions in processes of developmental change in self-concept and identity. It was their idea, therefore, that analyzing one aspect of these developmental processes might provide further insight. This brought Haviland-Jones and colleagues to the study of microprocesses of fear by means of a dynamic systems model. They base their model of emotion on the theory of Tomkins. In the model they combine the idea of continuous change in the system’s input with

210


various discrete emotions as its categorical output. Such a model also allows for continuous transitions from one emotion to another. The dynamic systems methodology developed by Van Geert makes it possible to run computer simulations of these continuous emotional processes and to study the complex interactions involved. The computer simulations of emotion processes carried out by Haviland-Jones, Boulifard, and Magai suggest that differences in the original state of the person, or in the stimulus itself, may make large differences in the emotional experience, with implications for immediate behavior as well as long-term identity scripts. Haviland-Jones and colleagues have demonstrated that information parameters are in themselves capable of producing shifts in emotion categories. Furthermore, conscious, goal-oriented knowledge does not seem to be essential for these shifts. The model and simulations also give room to interesting speculations about the interdependencies of emotional and cognitive processes: brief and unconscious emotional processes could evidence ‘‘phase shifts,’’ which could even lead to changes in stable patterns in the person’s self-concept. Beliefs and knowledge about emotions also play a role in the processing of self-relevant information. They may enable individuals to change themselves, but due to the intricacies of the mostly unconscious dynamics of microemotional processes, such changes can take unpredictable courses. These may or may not correspond with how individuals understand and describe themselves: A ‘‘good linear story’’ about how one came to be – call it one’s narrative or one’s meaning-making – does not necessarily reflect the workings of the microemotional processes in the change and stability of the self. Haviland-Jones and colleagues conclude by suggesting that the iteration of the phase shifts on the microlevel and the meaning-making level may lead to changes in one’s self-view. In his commentary on chapter 7 Heymans does not contest this suggestion, but he warns us that one should not forget the role of cultural constructions of the self in the individual’s developmental process, or the individual’s own active involvement. In line with this, he also observes a conceptual gap between the description of selforganizational processes occurring at the microlevel of ongoing, mostly unconscious emotion processes, and at the macrolevel of active, conscious identity construction. Heymans broaches an important and thorny issue that is still left unanswered. Throughout this volume, there seems to be a dichotomy in the kind of explanations that are used to describe the dynamic organization of self and identity. Self and identity are seen either as emergent qualities of emotion-driven self-organizational processes


211

(Fogel, Haviland-Jones et al.) or as a self-reflective construction of meanings (Hermans and Hermans-Jansen, Jansz, Heymans). In chapter 8, Lewis and Ferrari attempt to unite these two types of explanation under the header of ‘‘implicit’’ versus ‘‘explicit identity.’’ Unlike Epstein’s CEST, Lewis and Ferrari do not try to solve this issue by postulating two totally independent self-systems. Instead, they strictly adhere to the organizing principles of dynamic systems, principles that help to explain how complex, higher-order organization and stable states emerge from the interactions between lower-order components. As such, Lewis and Ferrari present an encompassing application of dynamic systems thinking in the domain of self and identity. According to Lewis and Ferrari, implicit identity – the sameness and uniqueness of one’s personality – is a naturally evolving characteristic that we have in common with most animals and that already exists in young infants. It emerges from recurrent, self-stabilizing patterning of appraisal–emotion interactions. The self-organizational processes involved build on each other along three different time scales. In real time, the appraisals and emotional states that accompany our actions are coupled into temporary ‘‘emotional interpretations’’ (EIs). These EIs dissipate as soon as the initial goals are achieved. When we are unable to achieve the goals, the goal state perseveres. The emotions that this situation generates have a consolidating effect on the current EIs, turning them into more or less lasting moods that may persist for minutes to weeks. Such moods have a facilitating effect on the interpretations and action-tendencies that are congruent with them. They also constrain the likelihood that alternative EIs develop into a mood. Through this habitual shaping of the mind, the configuration of EIs that fit in with a recurrent mood may start to function as a psychological attractor state, and become a more permanent ingredient of one’s personality. This is especially the case when long-lasting needs and wishes are involved that are difficult to accomplish, given the personal circumstances one lives in. In a nutshell, then, psychological continuity crystallizes from the mind’s preoccupation with ‘‘unfinished business,’’ rather than being the intrinsic property of an underlying, invariant self-structure. A fully developed personal identity relies on the sameness of the implicit identity, but also implies a coherent composition of selfappraisals. This is what Lewis and Ferrari call the ‘‘explicit identity.’’ It contains people’s ‘‘statements, stances, policies, or stories about who they are and how they live their lives’’ (p. 178), and stems from semantic, reflective, and often conscious activities of a more autobiographical and dialogical nature. Again, such self-reflective activities occur especially when people have to deal with unresolved needs and goals or

212


the conflicts that this arouses. The main function of self-narratives and self-dialogues is to account for the fact that one has not yet succeeded in becoming the person one wants to be. This is done in anticipation of real or imagined listeners. However, since the potential audience can differ widely in their appreciation of the narrator’s intentions, and the narrating individual can adopt a multitude of I-positions, there can be no conclusive understanding of oneself. Lewis and Ferrari also stress that from a dynamic systems perspective, identity change can be modeled through different kinds of change in self-organizing systems. Identity change across developmental time may stem from the emergence of new patterns, the strengthening of new habits, and dramatic events. However, in her commentary, Magai tries to qualify Lewis and Ferrari’s assertion that the same dynamic system principles govern change and stability in implicit and explicit identity. She refers to the empirical evidence that, precisely because of its dialogical nature, explicit identity appears to be much more susceptible to change than the more deeply entrenched implicit identity. She therefore wonders whether at a more specific level each type of identity may imply rather different types of organizing mechanisms and processes. This is one of the fundamental questions that will be discussed in the following section. In their reaction, Lewis and Ferrari (unpublished) agree that there are important differences between identity and personality: not only does identity change more easily, but probably it can also change more partially. They suggest however that the differences between both systems do not reside in the general dynamic systems principles, but in the more specific mechanisms and processes on which these general principles operate. In the course of Lewis and Ferrari’s comprehensive discussion, many of the conclusions of previous chapters reappear: the emphasis of Frijda, Fogel, and Haviland-Jones and her colleagues on the organizing role of emotions in self and identity; Van Geert’s outline of dynamic system principles; Fogel’s distinction between three time scales and levels of organization; and Hermans and Hermans-Jansen’s sketch of the dialogical self. However, as Magai has already observed, this comprehensiveness also carries the cost of addressing the dynamics of self and identity in a general and abstract way. In their study of these dynamics, researchers need to be aware that there are some decisive analytical implications that have to be taken into account when the general principles of self-organization are translated to the domain of self and identity. We will, therefore, conclude this closing chapter by discussing some of the most compelling of these implications.


213

Implications of a dynamic systems approach to self and identity In the first two chapters of this book three broad problem areas were identified in recent developments in the study of self and identity: the role of emotions in a predominantly cognitivistic approach, the question of how to take the context into account, and the issue of stability versus variability and change. Most of the authors in this book share a perspective that differs fundamentally from the approaches that have to contend with these three problems. They see emotions and context as formative conditions from which self and identity emerge in a selforganizational process. In such a dynamic systems perspective the various elements ‘‘cognitions, emotions, and context’’ are viewed as a complex, interacting network in which the embeddedness in the context and the changeability are inherently given. The limitations of the traditional cognitivistic approach can be overcome via this new perspective. The use of a dynamic systems perspective has brought us a wealth of new ideas, but it also raises new questions; the ensuing implications primarily have to do with how the concept under study is approached and defined. In his chapter Van Geert distinguishes four steps necessary for applying such a perspective. The first step, conceptual construction, concerns the definition of the components of the system. In addition, Camras and Michel stress the point that a dynamic system always consists of multiple layers, and that the phenomena emerging from these components should also be defined clearly. Further, it must also be clear what exactly is meant by the different concepts used in the self and identity literature (e.g. self-concept versus identity). A first implication therefore concerns an elaboration of self and identity in terms of multiple layers of components and emerging phenomena. As a further implication we will discuss the distinction between the various concepts that are central to this volume. The second step in Van Geert’s description concerns the formulation of the interactions between the components. Applying a selforganizational perspective to self and identity implies that the processes involved in the interactions between the components, within and between levels, have to be described. This is another implication to be elaborated here. Furthermore, as Camras and Michel point out, the relation between identity development and its context has also to be defined. Yet another implication is that there are alternative ways of conceptualizing this relation. The third step mentioned by Van Geert concerns a quantitative approach: the translation of the conceptual model into a mathematical

214


one. Elaboration of this step goes beyond the scope of the present chapter, although we will, by way of an example, briefly describe how different conceptualizations of the context could be translated into a mathematical model. Van Geert’s final step concerns the exploration of the model. This, of course, is needed both in a qualitative and in a quantitative approach: researchers have to formulate hypotheses and test these against empirical data. A final implication of a dynamic systems approach has to do with the kind of data and methods of analysis that are needed for testing such hypotheses. Components and emerging characteristics A dynamic system is – by definition – multilayered. This means that components and emerging characteristics should be defined separately for each level of aggregation: a phenomenon that is an emerging characteristic on one level can be a component on a higher level. Although the hierarchies and components discussed by the various authors are not always comparable, we will elaborate on how different levels of organization can be distinguished and ordered with regard to self and identity; we will then try to relate the concepts discussed in this book to these different levels. The lowest level (see table 9.1) discussed in this volume consists of components such as feelings, bodily sensations, cognitions, orientations, perceptions, appraisals, action tendencies, etc. This is a somewhat arbitrary choice of ‘‘lowest level,’’ because these components, for example bodily sensations, can also be seen as higher-order phenomena that emerge from components of an even lower order. Coupling these components results at the second level in the emergence of (higher-order) temporary emotional interpretations, which are probably comparable – in terms of level of aggregation – with what Fogel calls events. On the third level these events self-organize over time into enduring phenomena: recurrent emotional interpretations or EIs (Lewis and Ferrari), frames (Fogel), roles (Haviland-Jones et al.), and valuations (Hermans and Hermans-Jansen). Self-evaluations and selfrepresentations can also be situated on this level. On the fourth level, self-organizing EIs give rise to implicit and explicit identity (Lewis and Ferrari). ‘‘The self one is’’ (Frijda), sense of self, sense of identity (Fogel), and self-theory (Epstein) can also be situated on this fourth level of organization. (For reasons that will be explained below, self-concept is not included in the table.) Table 9.1 provides an overview of how – from a self-organizational perspective – different concepts can be related to each other. The table is


215

Table 9.1. Aggregation levels of self-organization 1. feelings, bodily sensations, cognitions, perceptions, appraisals, orientations, etc. 2. temporary emotional interpretations, events 3. Emotional Interpretations, roles, valuations, self-representations, self-evaluations, frames 4. self, self-theory, implicit identity, explicit identity, sense of identity, sense of self

not intended to be the basis of an exhaustive definition of the one and only system of self and identity. The definition and the defining components that are used depend on the concept on which a particular study focuses. Hermans and Hermans-Jansen, for example, focus on the narrative organization of valuations in their case study. Each valuation can be seen as a system, consisting of events, temporary emotional interpretations, emotions, and cognitions to begin with, but they become indicative of self and identity because they cohere into a meaningful whole. Haviland-Jones and colleagues, on the other hand, describe a system that consists of lower-order components and emerging patterns of emotions. The various levels are related to different time scales. Lowest-level components self-organize in events, that is, momentary emotional experiences, on a microscopic time scale; experiences subsequently organize in EIs on a medium time scale; and identity, finally, emerges on the highest time scale. The role of emotions and the relation with the context differs across the different time scales. It is essential, therefore, to be clear about the time scale one is talking about. From a systems perspective it is also important to realize that the levels of self-organization presented here are in turn part of a much more extensive system of systems. Fogel, for example, sees phenomena like history and culture as constructs of an even higher order. Conceptualizing self and identity as emerging from a multilayered system of interacting components has some major implications. First, the question ‘‘Which comes first, emotions or cognitions?’’ becomes a meaningless one. There is a continuous interaction between all components, EIs themselves being a conglomeration of emotions, perceptions, cognitions, etc. Second, this conceptualization implies that constructs like self-concept, self-esteem, self-evaluations, and all kinds of ideal, real, feared and other selves, are entities that do not pre-exist in a person’s mind. Stability does not mean that the construct is always there (in this sense the word construct is inadequate, because it suggests something that is always there), but that comparable phenomena

216


emerge in different situations over time. For example, a similar selfevaluation may emerge during many years in different situations in which an individual’s attention to his or her own person is drawn. What situations they are depends on the person, the moment, and the situation. Since self-related phenomena self-organize at specific moments, they will be comparable, but they will probably also vary systematically in different situations. A self-organizational perspective does not necessarily require the defining of different kinds of selves (‘‘self at school,’’ ‘‘self as I would like to be,’’ etc.) to account for these systematic differences. It is therefore much more economical than essentialistic theories, which do need such a priori selves. Moreover, since emotions and context are fully present in the system at the most basic level of self-organization, there is no need to add them as separate constructs later on. Sense of identity, implicit and explicit identity We have so far argued that in dealing with self and identity it is essential to make a distinction between different levels of aggregation. Processes on all levels play an important role but studies of self and identity usually focus on long-term developmental processes, that is, concepts on the third and fourth levels. In general, the concepts mentioned on the third level are clearly defined throughout the book. Concepts and definitions on the fourth level, however, cause considerable confusion. The concepts on this level are all global ones, they concern the person as a whole, and emerge from the same underlying layers. What they have in common, and what distinguishes them from other constructs in psychology, for example ‘‘the person’’ and ‘‘personality,’’ is that they refer to the person’s own perspective upon and experience of him- or herself. ‘‘Self’’ and ‘‘identity’’ are the most encompassing constructs in table 9.1. As stated in chapter 1, we treat them as interchangeable. We prefer to continue with the term ‘‘identity’’ only because of its more circumscribed meaning. The differentiation made by Lewis and Ferrari between ‘‘implicit’’ and ‘‘explicit’’ identity is especially helpful. Apart from ‘‘a sense of identity,’’ all fourth-order constructs in table 9.1 are covered by these two constructs. Thus, sense of identity, implicit identity and explicit identity will be the three central constructs used in the remainder of this chapter. Sense of identity is a general feeling, a sense of ‘‘sameness and continuity’’ (Erikson, 1968; see also Blasi and Glodis, 1995) that runs through the whole system. Because of its coordinating character, it should be situated on the highest level of abstraction (in table 9.1). The emergence of a sense of identity has nothing to do with the content of any of


217

the components, but everything to do with the way the different components fit together over time. Continuity of emotions, of frames, and of emotional interpretations gives rise to a feeling of being the same over time. Thus, a sense of identity may emerge from all three lower levels of aggregation: the underlying system of components, self-organized events, and EIs. Paradox is inherent in a sense of identity: if people have a sense of identity, they are (generally) not aware of it (Erikson, 1968). It manifests itself mainly as a sense of well-being; from a self-organizational perspective this means having a total absence of any positive feedback loops in the system. Positive feedback loops fixate, stabilize. In their absence, the system can be seen as a completely free-floating continuous relationship in which all kinds of change are possible, as in the state that Fogel describes as purely orientational. This conceptualization underlines the fact that a sense of identity refers to a fit and is fundamentally relational. The absence of a sense of identity, on the other hand, may be felt keenly and consciously, and may give rise to heightened and preoccupied attention to identity issues (Erikson, 1968). Implicit identity is the organization of especially non-conscious EIs a person has of him- or herself. It resembles Epstein’s concept of ‘‘implicit self-theory,’’ which – together with a ‘‘world theory’’ and notions about the relations between these two – makes up the experiential system. This experiential system is comparable to what Lewis and Ferrari call personality. As Epstein points out, there is no sharp boundary between a person’s self-theory and world theory. In fact, they are two sides of the same coin. For example, the notion that ‘‘the world is a dangerous place’’ implies that the person sees him- or herself as vulnerable. Nevertheless, the chance that mutual couplings will emerge is higher where both EIs concern oneself. This will result in the formation of a subsystem, the internal connections of which are much stronger than the connections with components that do not belong to the subsystem. Implicit identity can be distinguished, therefore, as a self-organized system within personality. This implies that specific EIs can be more or less categorized in terms of whether they are part of identity. Most components on the lower levels cannot in themselves be categorized as being or not being part of implicit identity. A feeling may be part of an EI concerning oneself, but also of some EI concerning the world. Frijda distinguishes several aspects of emotions that have to do with self: emotions that have oneself as object, the significance of the emotion, implicit estimates about one’s own competence, and the two basic concerns of self-esteem and sense of identity. We assume that at least one such aspect must be present before an EI concerning oneself can possibly emerge. However, all elements that are part of personality may become part of the implicit identity.

218


Explicit identity is an organization of conscious EIs about oneself. It is autobiographical and dialogical. It resembles the concept of narrative (Hermans and Hermans-Jansen), and explicit self-theory (Epstein). Explicit identity and related concepts have been clearly described elsewhere in this book, and need no further explanation here. As compared with implicit identity, explicit identity is organized in a more compartmentalized way, which means that parts may change, while the rest remains the same. Other concepts in the field differ from self and identity mainly with regard to their inclusiveness. Camras and Michel, for example, suggest that self-concept could be conceived as a component of identity viewed as a dynamic system. We prefer to view self-representations as components of identity. Self-concept could then be defined as the collection of descriptions of ‘‘who one is,’’ ‘‘who one likes to be,’’ ‘‘how one thinks others see oneself,’’ etc., and therefore as conscious self-representations that are part of one’s identity. In this view self-concept is not an emerging higher-order construct: as a collection it is no more than the sum of its components. Consequently, self-evaluations are defined as those self-representations that include emotions, values, and judgments about oneself. Apart from the differences in content described above and elsewhere in the book, differentiating the concept of identity in implicit identity, explicit identity, and sense of identity is also necessary because, as will be argued below, the processes involved and the assessment procedures differ according to which construct is entailed. Processes Now that identity has been described as a self-organizing system, we can examine the processes involved. Both implicit and explicit identity emerge from underlying levels. Lewis and Ferrari suggest that the mechanisms involved might be comparable. However, as Magai suggests, the processes involved in the emergence of implicit identity might well differ from those playing a role in conscious phenomena like explicit identity. She points out that explicit and implicit identity differ in susceptibility to external influences and in probability of change. This would mean that there are different processes involved in the two systems, as is also argued by Epstein and Pacini (1999). Camras and Michel’s distinction between verbal explanation versus formal mathematical models is relevant in this controversy. It may be that the differences are restricted to the verbal model. The formal causes and types of mechanism can be the same in systems that differ strongly with regard to the verbal causes and variables involved.


219

We will first consider the processes within the two identities on a verbal-psychological level. Various theories suggest that these are quite different. Epstein states that the processes in the experiential system are based on non-conscious, rapid association, and those in the rational system on deduction and reasoning. In a similar way, Dannefer and Perlmutter (1990) distinguish between two different processes that coordinate between inner-biological, individual-psychological, cultural-sociological, and outer-physical processes; such a distinction can also be related to the two types of identity. The first process, habituation, is a more-or-less automatic, self-regulatory process that is characterized by routine, non-reflective interaction with the environment. This process contributes to the non-conscious maintenance of a sense of identity. It works on the level of emotional interpretations and includes all kinds of biased types of information processing that prevent people from becoming aware of discrepancies with the desired perception of themselves. This is a discrepancy-reducing, equilibrium-seeking process. The second process, cognitive generativity, is an intentional, reflective form of coordination: the ability to take oneself as object of thought, to reflect upon the past and the present, to anticipate the future, and to speculate about the unknown. This resembles internal dialogues as described by Hermans and Hermans-Jansen, and explicit identity or narratives as described by Lewis and Ferrari. Cognitive generativity is a conscious attempt to modify and mobilize habitual processes that have become maladaptive. The process works on the level of the deliberate change of one’s narrative and is based on the rules of communication and narratives. One of the consequences of this may be that explicit identity can change in a more compartmentalized way (Lewis and Ferrari, unpublished). One of the commitments or voices in the dialogue may weaken, disappear, and be replaced without necessarily turning the whole system upside down. It is not clear how the processes in the two systems are related. Do they affect each other? Are they active alternately, or do they run in parallel? Epstein describes how the rational system may ‘‘correct’’ the experiential system if reflection and conscious reasoning ‘‘detect’’ maladaptive thoughts or actions. The experiential system, in turn, may affect or bias the rational system in a non-conscious way. People do not generally reflect upon themselves most of the time, however. Cognitive generativity might be activated following the inability of the habituation process to dissipate emotions (compare Jansz). This may disrupt one’s sense of identity and result in the emergence of strong emotions. Such strong emotions may trigger self-reflection, as Oosterwegel suggests. On a macroscopic time scale, we see that the developmental courses of the two types of identity differ strongly. According to Erikson (1968),

220


early characteristics of implicit identity are present from birth. One’s explicit identity becomes an imperative from adolescence onwards. The person is then confronted with a combination of biological, social, and cognitive changes that makes his or her implicit identity insufficient for safeguarding a sense of identity, resulting in feelings of identity confusion. People will then try to retrieve their sense of identity, and they do so by constructing a personal identity: a tentative, reflectively construed configuration of self-descriptions that allows the person to find his or her own place in the world. A new qualitative dimension is thereby added to the system. This new dimension does not differ from the sense of identity in terms of level of aggregation, but in terms of level of consciousness. Its emergence is the beginning of an ongoing dynamic process of identity maintenance: a more deliberate process of stabilization and change. Every time that – due to developmental growth, changes in situations or in underlying goals and needs – feelings of identity confusion occur, active and conscious changes in explicit identity may follow. Such changes in explicit identity probably have consequences for implicit identity. Identity development can thus be seen as an ongoing interaction between implicit and explicit identity, between non-conscious and conscious processes. The case studies discussed by Hermans and Hermans-Jansen, and by Verhofstadt-Dene`ve, and the intervention methods used in these cases provide some additional suggestions about the how and when of developmental switches and of the dynamics involved in such changes. Both of the clients involved sought help in a period of distress. Initially, the reason for this distress was not part of their conscious, explicit identity. During the therapy, different parts of the non-conscious system were brought into awareness. Especially in the case of Sharon, the client’s troubles seem to stem from the existence of two different organizations, or attractors, and from a lack of communication between them. The aim of the therapy is to bring the two attractors together, by making the person aware of their existence, and by stimulating the person to communicate between the two perspectives. From a self-organizational perspective, one can describe both therapies as attempts to lower the divisions between different ‘‘basins’’ of attraction, by coupling components of the two attractors in new and different ways, as attempts to build a new self-organizing attractor, new EIs. In Verhofstadt-Dene`ve’s psychodrama, one of these attractors or voices is as it were ‘‘moved’’ to another person. Distress, in general, may result in a focus on what is going on, in an increasing awareness of previously non-conscious EIs. This awareness enables a deliberate change of the couplings and thus the EIs. The non-conscious system is then changed via conscious intervention. This is not to say that all couplings can be changed at will.


221

Haviland-Jones and colleagues have argued that the construction and change of one’s own identity may also result from unconscious microlevel processes and have consequences other than those expected. Although the existence of different attractors in the self-system can be problematic, people in general have several EIs. This raises the question of when the existence of different attractors becomes troublesome. One could imagine that conflicts between EIs arise if they are both triggered in similar sorts of situations, in which they fulfil different needs, or if they threaten each other’s needs. For example, an interpretation of oneself as being strong and independent may threaten the need for belonging. Epstein (1998b) has described how different conflicts and dis-balances between needs can result in distress. From the foregoing, it may be obvious that on a verbal level of understanding the processes involved in implicit and explicit identity clearly differ. This does not tell us anything, however, about the differences from a formal, dynamic systems perspective. From this perspective, self and identity comprise multiple levels of aggregation, with two types of self-organizational process: within one level of aggregation (horizontal) and between different levels (vertical: bottom-up and top-down). These processes take place on and between all levels of aggregation. Lewis and Ferrari introduce two basic principles of self-organization, which together explain how lower-order components give rise to stable higher-order phenomena (bottom-up processes). First, components couple, and this coupling results in the emergence of higher-order forms; and second, subsequent recurrent coupling may lead to the stabilization and strengthening of these forms. Crucial in their approach is the idea that recurrence takes place when experiences cannot be brought to a completely satisfying completion, which leads to unresolved intentions and blocked-off actions. Depending on the strength of the accompanying emotions, such experiences may form the basis of recurrent and salient attractors. Mechanisms like these make it possible to explain why some experiences develop into recurrent, salient, characteristic patterns and others do not. Another elaboration of bottom-up processes can be found in Nowak, Vallacher, Tesser, and Borkowski (2000). They demonstrate, by means of a cellular automaton, how interacting elements give rise to higher-order phenomena such as the global evaluation of the self. The driving force in this model is the press towards integration, i.e. the striving for coherence between different components in the self-concept. In contrast to the emergence of higher-order phenomena from the interaction of lower-order components, other processes work topdown: higher-order phenomena may affect lower-order phenomena. In

222


his reply, Van Geert (unpublished) has discussed two possible forms of these top-down influences. First, newly emerging phenomena may affect streams of energy or information in the lower-order system. One’s narrative may affect the way one appraises new situations. For example, the appraisal of computer-related problems will change if one starts to see oneself as a computer specialist. People may in a conscious and deliberate way change their behavior and environment and try to adjust their interpretations, feelings, and thoughts to their own narrative. Second, a new phenomenon may be of the same order as the components from which it emerges, and become part of the system. A newly emerging role, for example becoming a mother, will lead to new couplings and may destroy old ones. In this way, the new role affects other roles. A process that occurs within one level (horizontal) is the so-called phase transition: a change in the organization of the higher-order phenomenon, which can be triggered by small changes in underlying, control variables. Sudden changes may depend on gradual changes in control parameters in a very non-linear way, and their occurrence often depends on all kinds of system characteristics. This means that a transition never has just one cause, but takes place as a consequence of the coincidence of many interacting variables: the state of the whole system and its context. Moreover, major and sudden changes do not need major external events for their occurrence, but can be caused by gradual, and possibly invisible, processes within the system itself. In physical systems, self-organization is assumed to be activated by a striving for the lowest level of energy. Lewis and Ferrari assume that psychological systems may be activated (or motivated) by the striving for a smooth stream of information or energy – flow in Csikszentmihalyi’s (1990) terms. In flow, the stream of information or energy is not hampered, does not get stuck in recurrent patterns, but is free floating. Several authors in this volume assume the existence of socalled basic needs that activate development. Frustration of these needs might block the stream of energy and disrupt the sense of identity. On a non-conscious level, this may result in the formation of new EIs. On a conscious level, in their narrative identity people aim at stability, continuity, and coherence. A stable, enduring and coherent narrative enables rapid processing and integration of information. A disruption of one’s narrative means that there is new information that cannot be processed and integrated. This calls for reflective forms of coordination to change the narrative in such a way that the new information can be integrated.


223

At this moment there is little to say about the difference between implicit and explicit identity as represented in formal (mathematical) models. Little is known about the formal characteristics of the mechanisms we described above in verbal terms. In their reaction to Magai, Lewis and Ferrari (unpublished) refer to Allan Schore who, in a personal communication (August, 1999), suggested that implicit and explicit identity may rely on altogether different memory systems, yet both develop in parallel and interact continually. For one thing, we have no idea about the developmental trajectories of either system, which leaves us with several questions. Do changes in the two systems always have the form of sudden transitions from one attractor to another, as Lewis and Ferrari describe? Probably, gradual changes are also possible. Are there comparable process parameters for the two systems? What does it mean on a formal level if underlying memory systems differ? Implicit identity aims at equilibrium, but is equilibrium also important for explicit identity? Our guess is that equilibrium is not especially important in the case of explicit identity. A narrative should be convincing, but does not need to be consistent. A good narrative even requires some degree of inconsistency and dramatic tension (Bruner, 1990). In Lewis and Ferrari’s opinion, retelling a narrative over and over again strengthens it in the same way as happens with repeated couplings on the implicit level. Are change processes within the two systems comparable? This is hard to determine. To be able to answer the question of differences on the formal level, much more needs to be known about the formal characteristics of the emergence and development of self and identity. Systems and their context Van Geert states that in a dynamic systems approach the context may refer to the whole universe, with the exception of the system itself. However, it is far from economical to include the universe in one’s model. Instead, it is the task of the researcher to define what aspects of the context are expected to be relevant for the system under study. In this regard, the authors in the present volume are not of one opinion. Fogel’s approach is fundamentally relational: the focus of his study is on the characteristics of relations rather than on those of the person or the context. Haviland-Jones and colleagues have a different approach. In their view, the context is important in another way. First, it offers specific situations, such as a prolonged, frightening war-time situation. Second, the influence of context is manifested in the different roles as perceived by the person him- or herself.

224


The way in which the context is brought into the system has nothing to do with the real characteristics of that context, but rather reflects those relations with the context that are considered essential in this specific model. An important insight from dynamic systems theory is that living dynamic systems have no clear and fixed boundaries. The context cannot simply be defined as ‘‘outside the skin’’ and the system as ‘‘inside the skin.’’ It is at least partly a characteristic of the perspective of the observer: an individual may perceive unwanted ‘‘inner voices’’ as being outside herself, outside her system, as ‘‘not me.’’ Thoughts and emotions that are felt as unacceptable by the person may also be seen as ‘‘context,’’ or at least as ‘‘not me.’’ A researcher, on the other hand, studying this person’s conflicts, may regard these opposites as – conflicting – parts of a single system (e.g. Kunnen and Klein Wassink, 1999). In a dynamic systems approach one should, therefore, define which aspects belong to the system, which contextual factors can be considered important, and how these contextual factors may affect the system under study. For the researcher who seeks to specify how the system is affected by the context, the three distinct conceptualizations of Dannefer and Perlmutter (1990) may be helpful. The influence of the context can be defined as (1) a random influence, (2) a factor having ordered effects, or (3) a system in itself. The differences between these conceptualizations become most explicit when the researcher wants to translate the relation with the context into a quantitative dynamic systems model. Therefore we will illustrate each conceptualization by describing such a ‘‘translation.’’ Context as random influence means that the self-organization of the system is affected in an unpredictable and non-systematic way from the outside. There is no systematic relation between subsequent disturbances or between the disturbances and the system. In a mathematical dynamic systems model, this could be expressed by adding or subtracting a small random number for each iteration. If context is seen as a factor that has ordered effects, there is a systematic relation between the contextual influences at different moments, and the context is more or less predictable. We already mentioned the example of living in a war situation providing a long sequence of potentially frightening experiences. Context as an ordered effect can be represented in a model as a parameter. A lower or higher value of this parameter, then, expresses whether the context is more or less frightening. Such a parameter may differ between individuals and situations, and change over time. If context is seen as a system in itself, it has its own internal dynamics, apart from the system under study. For example, the context could have its own stabilizing forces, which can counteract changes in the ‘‘identity’’


225

system. In this case, a mathematical model could consist of two interacting systems or represent the context in the form of a separate logistic equation that is related to the other equations. Note that the system and the environment could equally well be seen as two interacting systems, or as one system. This ambiguity illustrates the point that systems have no intrinsically defined borders and environments. Conceptualizing the context as a system has several advantages. First, it makes it clear that people are intrinsically related to their environment, while at the same time the environment is seen as being organized on its own terms, having its own dynamics and its own selfperpetuating tendencies and transitions. Second, the context can be meaningfully conceptualized at different levels of generality, in the same way that there are different levels of identity and self as shown in table 9.1. Society at a macrolevel concerns nations and institutions; on an intermediate level, schools, corporations, families etc.; and on a microlevel a face-to-face interaction between individuals (compare the discussion of Bronfenbrenner’s work in chapter 2). The different levels of generality offer possibilities for a much more detailed and specific focus on the relation between identity and context. The conception of context as a system in itself represents the most thorough account of the relation between system and context. This characterizes the work of Fogel. He integrates the infant-system and the mother-system into one system that consists of the interaction between mother and infant, thereby including the dynamics of both the infant and the mother. However, this mother-plus-infant system also has a context. Which components are part of the system and which belong to the context therefore depends on the researcher’s objectives. A consequence of a self-organizational perspective is that, regardless of whether the context is conceptualized as random factor, ordered factor, or system, the effects of the contextual factor on the development of the system depend on its state. Random influences, for example, often have no effect at all. In an unstable system, a small random disturbance may cause a transition, however. In the same way, an ordered factor may have a positive effect, no effect, or a negative effect, again depending on the state of the system. If the context is seen as a system in itself, the effects of the context not only depend on the target system, but the target system in turn affects the context. For example, whether an individual’s self-esteem is affected by a critical remark strongly depends on that individual’s appraisal. This appraisal results from the interaction between the identity-system and the context. The individual reaction, in turn, will affect the behavior of the person who made the remark, and thereby change the contextual system. Which may change the appraisals, and so on.

226


Assessing self and identity as self-organizing systems The new and promising approach towards identity presented here will only be successful if it generates and is supported by empirical research. A final implication we therefore want to address is the operationalization, measurement, and analysis of key constructs. The assessment of identity as a self-organizing process poses specific demands on the way information is measured and analyzed. Two notions of dynamic systems are especially relevant here. First, self and identity are not structures that are always present, but rather phenomena that emerge in a self-organizing system. Second, essential characteristics of self and identity as a self-organizing system manifest themselves in the way they change over time. A fundamental difference from previous conceptualizations is that stability of self and identity does not refer to structures that are always present. Stability means that comparable phenomena (EIs) emerge in different situations over time. We therefore have to find ways to ‘‘grasp’’ these phenomena at the moment at which they are present. These ways may differ in accordance with the various types and aspects of self and identity that have been distinguished in this chapter. Implicit identity is the emerging system of self-referent EIs. The phenomena we want to observe, for example the emotional reactions when something concerning the person is at stake, only emerge in specific situations. This implies that the system should be studied when it is ‘‘at work’’: in situations that matter with regard to self and identity. It cannot be assessed at any time and in any place, for example by administering a questionnaire. Several authors in this volume have developed different methods to assess the system ‘‘at work.’’ Fogel and Verhofstadt-Dene`ve observe their subjects directly in situations that matter to them. Fogel observes mothers and infants during their daily interactions, and Verhofstadt-Dene`ve creates situations in psychodrama in which self-referent EIs are expected to emerge. HavilandJones et al. and Hermans and Hermans-Jansen assess the working of the system more indirectly, by asking subjects about personally relevant roles and situations. By analyzing the relation between assessments, Haviland-Jones and colleagues and Hermans and Hermans-Jansen try to understand the structure of the higher-order system: the self or identity. Another possibility is to gather personal reports, such as diaries (Haviland-Jones et al.) and frequent open interviews (Kunnen and Klein Wassink, 1999), which yield relevant and emotional experiences in situations when the system was operating. In these methods, the researchers try to reveal non-conscious aspects of EIs by analyzing commonalities across situations.


227

Acquiring information about explicit identity is less difficult. Explicit identity is, by definition, the conscious and reflected narrative or identity. Although this identity need not always be present either, its emergence can be controlled by the person concerned. It can be assessed by asking people for their narratives, commitments, or simply ‘‘who they are.’’ This is closest to the open type of traditional assessment methods of self and identity. This openness is a prerequisite here: one’s stories and commitments are individual characteristics, and differ from person to person. A sense of identity has been described as an all-encompassing feeling of sameness, continuity, and uniqueness. This can be deducted from observations or assessed by direct questions concerning these feelings. Such questions have been developed, for example, by Blasi and Glodis (1995). The second relevant notion is that a dynamic system is a system of related components that changes over time. A major difference between a dynamic systems, self-organizational approach and more traditional approaches is that the former approach studies phenomena from a diachronous perspective instead of a synchronous perspective. That is, it focuses on relations over time, instead of relations between variables (in subjects) at the same time. In the preceding chapters identity has been described as a time-related process that can only be studied from a diachronous perspective. Fogel, therefore, stresses that research should focus on sequences, in order to track how a psychological process unfolds over time in the context of coregulated interactions. This implies that longitudinal observations are required for the study of self and identity. In addition, criteria should be formulated for the kinds of patterns of stability and change that are relevant, and methods are needed to determine whether or not these patterns are present. Most methods of analysis in the social sciences are directed towards assessing differences, rather than towards assessing change. Dynamic systems theory may help to identify the phenomena that are relevant in the search for change. The notion, for example, that a system often stabilizes in one of only a limited number of stable attractor states can be of help in the search for stable states. In a comparable way, dynamic systems theory describes specific types of behavioral characteristics that precede, accompany, and follow transitions in different types of systems. For example, an increased vulnerability to disruptions, and a longer return time (that means, a prolonged period of turbulence following disruptions) often precede a transition. These characteristics can be operationalized in observable behavior. Knowledge of bifurcation points – critical points at which development may follow either one pathway or the other – can also be helpful when observing the behavior

228


of a system over time. Important questions are then: what triggers a bifurcation, and in what situations is such a transition more likely to occur? Because sudden changes may depend on gradual changes in underlying process parameters, such parameters should be found, and conditions should be defined under which bifurcations are more or less plausible. Techniques that assess and analyze relevant dynamic systems behavior are still scarce. Lewis, Lamey, and Douglas (1999) show how attractors and state space can be operationalized and demonstrated by simply registering well-defined behavior of the system over a certain period. Fogel focuses on interactions that have to do with mutual handling of toys and he demonstrates qualitative changes in this interaction. More traditional methods for analyzing change can also be very useful, however. Hermans and Hermans-Jansen compare assessments of an individual’s valuations at different times, and Haviland-Jones and colleagues assess change by hierarchical cluster analysis to compare an adolescent’s old roles with his or her new ones. Heymans stresses that some ‘‘classical’’ methods of time-series analysis also offer possibilities to analyze change in systems. Often, these methods (ARIMA for example) appear to be more common in fields other than the social sciences. In addition, we point to a form of factor analysis that can be used to assess change in individual time-series, developed by Molenaar (1985). Another way to assess characteristics of a system is to build a quantitative model of the system, and to simulate its behavior. Such a model reflects the theoretical assumptions concerning the dynamics of the system. These assumptions can be tested by comparing the system’s behavior with empirical phenomena or with other theoretical assumptions. To build such a model requires defining the components in the system (this step is needed in a qualitative approach as well), choosing the time scale, and turning the process of development into a few essential principles; moreover, a translation of these components needs to be made into numbers, and of the process principles into equations. Haviland-Jones et al. give a clear example of this approach. They demonstrate how a few basic principles can be translated into a dynamic systems model of the emergence of different emotions in a real-time situation. Differences in personal and situational characteristics can be expressed in terms of the values of a few parameters. Relations between these characteristics and emerging outcomes can be used as a basis for the validation of the model. Other examples of the application of quantitative dynamic systems models in the field of self and identity are given in Kunnen, Bosma, and Van Geert (in press), and in Kunnen and Bosma (2000).


229

Mathematical models can also be used to model the relations between microlevel systems. Emerging behavior on a lower level can be included as a parameter or variable in a higher-level system. Microlevel processes, then, may result in changes in parameters on a higher level of aggregation. Repeated frightening experiences – to follow the example of Haviland-Jones and colleagues – could result in an increased sensitivity to such experiences, and an increased chance that new situations will be experienced in a comparable way. Most changes in identity will concern a macroscopic time scale, and can be assessed by comparing assessments over longer time periods. The duration of the therapies in the case studies of Hermans and Hermans-Jansen and Verhofstadt-Dene`ve and the time covered by the diaries in Haviland-Jones et al., and the interview assessments of Kunnen and Klein Wassink (1999) range from months to years. Fogel observes mother–infant dyads over a shorter time span but here, too, the changes that take place within ‘‘frames’’ cover a period of weeks or even months. Studying and assessing self and identity as self-organizing systems require a thorough thinking and planning of assessment procedures. At this moment suitable techniques and methods are not commonly available. However, more and more applications of dynamic systems views and techniques are appearing in many different fields. We mention just a few examples: social psychology (Nowak and Vallacher, 1999), language development (Van Geert, 1991; Ruhland, 1998), cognitive psychology (Van Geert, 1991; Eckstein, 1999; Van der Maas and Molenaar, 1992), infant research ( Fogel, Nwokah, and Karns, 1993), motor development (Thelen and Ulrich, 1991), psychopathology (Msihara, 1996), and personality (Svrakic, Svrakic, and Cloninger, 1996). Since (on a general level) the dynamic systems in these fields have much in common, the publications offer a wealth of inspiring and illustrative applications of dynamic systems. The authors in the present volume offer elaborate and specific views on self and identity as emotional, dynamic systems, rooted in context. Despite clear difference of views, there is sufficient agreement to be able to formulate general as well as more specific characteristics of this new approach. The componential approach and the attention to time scales and aggregation levels provide new opportunities for the analysis of stability and change in the development of self and identity. As mentioned in our preface, we see this book as a next step in a discussion and exploration of a new perspective in the field of self and identity. Not all the questions we posed at the beginning have been answered. Several still remain, some have disappeared, others have been reformulated. The question of how self and identity develop in adulthood, mentioned

230


in chapter 1, has, for instance, not been discussed at all. The conceptualization of self and identity as a self-organizing system, as worked out in this volume, however, does offer a framework, we hope, for a further exploration of these questions. Note 1 Fogel’s question is a direct consequence of his relational and dynamic perspective. Authors working from an individualistic perspective, on the other hand, tend to posit stability and uniqueness and have to go to great lengths to conceptualize change and connectedness (e.g. Grotevant and Cooper, 1986; Josselson, 1994).

References

Abraham, F. D. (1990). A visual introduction to dynamical systems theory for psychology. Santa Cruz: Aerial Press. Abraham, F. D. (1995). Dynamics, bifurcation, self-organization, chaos, mind, conflict, insensitivity to initial conditions, time, unification, diversity, free will, and social responsibility. In R. Robertson and A. Combs (eds.), Chaos theory in psychology and the life sciences, pp. 155–174. Mahwah, NJ: Erlbaum. Abram, D. (1997). The spell of the sensuous: perception and language in a more-thanhuman world. New York: Vintage Books. Allport, G. W. (1955). Becoming. New Haven: Yale University Press. Angyal, A. (1965). Neurosis and treatment: a holistic theory. New York: Wiley. Antonovsky, A. (1993). Complexity, conflict, chaos, coherence, coercion, and civility. Social Science and Medicine, 37, 969–981. Antonovsky, A., and Sagy, S. (1990). Confronting developmental tasks in the retirement transition. The Gerontologist, 30, 362–368. Aron, A., Aron, E., Tudor, M., and Nelson, G. (1991). Close relationships as including the other in the self. Journal of Personality and Social Psychology, 60, 241–253. Ashmore, R. D., and Ogilvie, D. M. (1992). He’s such a nice boy . . . when he’s with grandma: gender and evaluation in self-with-other representations. In T. M. Brinthaupt and R. P. Lipka (eds.), The self: definitional and methodological issues, pp. 236–290. Albany, NY: State University of New York Press. Atkins, P. W. (1984). The second law. New York/Oxford: Freeman. Ausubel, D. (1955). Relationships between shame and guilt in the socializing process. Psychological Review, 62, 378–390. Baddeley, A. D. (1992). What is autobiographical memory? In M. A. Conway, D. C. Rubin, H. Spinnler, and W. A. Wagenaar (eds.), Theoretical perspectives on autobiographical memory, pp. 13–29. Dordrecht: Kluwer. Baerveldt, C., and Voestermans, P. (1996). The body as a selfing device: the case of anorexia nervosa. Theory and Psychology, 6, 693–713. Bailey, S. T. (1970). Independence and factor structure of self-concept metadimensions. Journal of Counseling Psychology, 17, 425–430. Bakan, D. (1966). The duality of human existence. Chicago: Rand-McNally. Bakhtin, M. M. (1973). Problems of Dostoyevsky’s poetics, 2nd. edn. Trans. R. W. Rotsel. USA: Ardis. First published in 1929 under the title Problemy tvorchestva Dostoevskogo [Problems of Dostoyevsky’s Art]. Bakhtin, M. M. (1981). The dialogical imagination: four essays by M. M. Bakhtin. Austin: University of Texas Press.

231

232

References

Baltes, P. B., Lindenberger, U., and Staudinger, U. M. (1998). Life-span theory in developmental psychology. In R. M. Lerner (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn., vol. 1, Theoretical models of human development, pp. 1029–1143. New York: Wiley. Bambrough, R. (1963). The philosophy of Aristotle. Trans. J. L. Creed and A. E. Wardman. New York: New American Library. Bandura, A. (1986). Social foundations of thought and action: a social cognitive theory. Englewood Cliffs: Prentice Hall. Barclay, C. R. (1993). Remembering ourselves. In G. M. Davies and R. H. Logie (eds.), Memory in everday life, pp. 285–310. Amsterdam: Elsevier. Barclay, C. R. (1994). Composing protoselves through improvisation. In U. Neisser and R. Fivush (eds.), The remembering self: construction and accuracy in the self-narrative, pp. 55–77. Cambridge: Cambridge University Press. Barclay, C. R. (1996). Autobiographical remembering: narrative constraints. In D. C. Rubin (ed.), Remembering our past: studies in autobiographical memory, pp. 94–126. Cambridge: Cambridge University Press. Barrett, K. C. (1993). The development of non-verbal communication and emotion: a functionalist perspective. Journal of Nonverbal Behavior, 17, 145– 169. Barrett, K. C. and Campos, J. J. (1987). Perspectives on emotional development II: functionalist approach to emotion. In J. Osofsky (ed.), Handbook of infant development, 2nd. edn. New York: Wiley. Bateson, G. (1955). The message: ‘‘This is play.’’ In B. Schaffner (ed.), Group processes, vol. 2. Madison, NJ: Madison Printing Co. Baumeister, R. F. (1987). How the self became a problem: a psychological review of historical research. Journal of Personality and Social Psychology, 52, 163–176. Baumeister, R. F. (ed.) (1993). Self-esteem: the puzzle of low self-regard. New York: Plenum. Baumeister, R. F. (1998). The self. In D. T. Gilbert, S. T. Fiske, and G. Lindzey (eds.), The handbook of social psychology, vol. 1, pp. 680–740. New York: McGraw-Hill. Baumeister, R. F. and Newman, L. S. (1994). How stories make sense of personal experiences: motives that shape autobiographical narratives. Personality and Social Psychology Bulletin, 20, 676–690. Baumeister, R. F., Smart, L., and Boden, J. M. (1996). Relation of threatened egotism to violence and aggression: the dark side of high self-esteem. Psychological Review, 103, 5–33. Baumeister, R. F., Stillwell, A. M., and Heatherton, T. F. (1994). Guilt: an interpersonal approach. Psychological Bulletin, 115, 243–267. Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: International Universities Press. Becker, E. (1973). The denial of death. New York: Free Press. Beltrami, E. (1987). Mathematics for dynamic modeling. Boston: Academic Press. Bu¨tz, M. R., Duran, E., and Tong, B. R. (1995). Cross-cultural chaos. In R. Robertson and A. Combs (eds.), Chaos theory in psychology and the life sciences, pp. 319–329. Mahwah, NJ: Erlbaum. Blasi, A., and Glodis, K. (1995). The development of identity: a critical analysis from the perspective of the self as subject. Developmental Review, 15, 409–433. Blass, E. M., and Shah, A. (1995). Pain reducing properties of sucrose in newborns. Chemical Senses, 20, 29–35.

References

233

Block, J. H., and Block, J. (1980). The role of ego-control and ego-resiliency in the organization of behavior. In W. Andrew Collins (ed.), Development of cognition, affect, and social relations, Minnesota Symposia on Child Psychology, vol. 12, pp. 39–110. Hillsdale, NJ: Erlbaum. Block, J., and Robins, R. W. (1993). A longitudinal study of consistency and change in self-esteem from early adolescence to early adulthood. Child Development, 64, 909–923. Boesch, E. E. (1991). Symbolic action theory and cultural psychology. Berlin: Springer. Bosma, H. A. (1995). Identity and identity-processes: what are we talking about? In A. Oosterwegel and R. A. Wicklund (eds.), The self in European and North American culture: development and processes, pp. 5–17. Dordrecht: Kluwer. Bosma, H. A., Graafsma, T. L. G., Grotevant, H. D., and de Levita, D. J. (eds.) (1994). Identity and development: an interdisciplinary approach. London: Sage. Bourdieu, P. (1997). Me´ditations pascaliennes (Pascalian meditations). Paris: Seuil. Bracken, B. A. (ed.) (1996). Handbook of self-concept: developmental, social, and clinical considerations. New York: Wiley. Brandtsta¨dter, J. (1998). Action perspectives on human development. In R. M. Lerner (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn., vol. 1, Theoretical models of human development, pp. 807–863. New York: Wiley. Brewer, W. F. (1996). What is recollective memory? In D. C. Rubin (ed.), Remembering our past: studies in autobiographical memory, pp. 19–67. Cambridge: Cambridge University Press. Brinthaupt, T. M., and Lipka, R. P. (eds.) (1992). The self: definitional and methodological issues. Albany, NY: State University of New York Press. Bronfenbrenner, U., and Crouter, A. C. (1983). The evolution of environmental models in developmental research. In W. Kessen (ed.), Handbook of child psychology, vol. 1, pp. 357–414. New York: Wiley. Brown, B., Werner, C. M. and Altman, I. (1998). A dialectical-transactional perspective on close relationships. In B. M. Montgomery and L. A. Baxter (eds.), Dialectical approaches to studying personal relationships, pp. 137–183. Mahwah, NJ, London: Erlbaum. Bruner, J. S. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press. Bruner, J. S. (1990). Acts of meaning. Cambridge, MA: Harvard University Press. Bruner, J. S. (1996). The culture of education. Cambridge, MA: Harvard University Press. Bruner, J. S., and Kalmar D. A. (1998). Narrative and meta narrative in the construction of self. In M. D. Ferrari and R. J. Sternberg (eds.), Selfawareness: its nature and development, pp. 308–331. NewYork: Guilford Press. Buber, M. (1958). I and thou. Trans. R. G. Smith. New York: Scribners. Buck, R. (1985). Prime theory: an integrated view of motivation and emotion. Psychological Review, 92, 389–413. Bullock, M., and Lu¨tkenhaus, P. (1990). Who am I: selfunderstanding in toddlers. Merrill-Palmer Quarterly, 36, 217–238. Burghess, D. N. and Wood, A. D. (1985). Mathematical models in the social, management and life sciences. Chichester: Ellis Horwood. Butterworth, G. (1995). An ecological perspective on the origins of self. In J. L. Bermudez, A. Marcel, and N. Eilan (eds.), The body and the self, pp. 87–105. Cambridge, MA: Bradford.

234

References

Byrne, B. M. (1996). Measuring self-concept across the life span: issues and instrumentation. Washington, DC: American Psychological Association. Camras, L. A. (1992). Expressive development and basic emotions. Cognition and Emotion, 6, 269–283. Camras, L. A. (2000). Surprise!: facial expressions can be coordinative motor structures. In M. Lewis and I. Granic (eds.), Emotions, development, and self-organization, pp. 100–124. New York: Cambridge University Press. Camras, L. A., Lambrecht, L., and Michel, G. F. (1996). Infant ‘‘surprise’’ expressions as coordinative motor structures. Journal of Nonverbal Behavior, 20, 183–195. Cantor, N., and Harlow, R. (1994). Social intelligence and personality: flexible life-task pursuit. In R. J. Sternberg and P. Ruzgis (eds.), Personality and intelligence, pp. 137–168. New York: Cambridge University Press. Capps, D. (1989) The deadly sins and saving virtues: how they are viewed by laity. Pastoral Psychology, 37, 229–253. Carstensen, L. L., Isaacowitz, D. M., and Charles, S. T. Taking time seriously: a theory of emotional selectivity. American Psychologist, 54, 165–181. Carver, C. S., and Scheier, M. F. (1990). Origins and functions of positive and negative affect: a control-process view. Psychological Bulletin, 97, 19–35. Cassirer, E. (1955). The philosophy of symbolic forms, vol. 2, Mythical thought. New Haven, CT: Yale University Press. Caughey, J. L. (1984). Imaginary social worlds: a cultural approach. Lincoln: University of Nebraska Press. Chandler, M. J., and Ball, L. (1989). Continuity and commitment: a developmental analysis of identity formation process in suicidal and non-suicidal youth. In H. A. Bosma and S. Jackson (eds.), Coping and self-concept in adolescence, pp. 149–166. Berlin: Springer. Chandler, M. J., Lalonde, C., and Sokol, B. (in press). Continuities of selfhood in the face of radical developmental and cultural change. In L. Nucci, G. Saxe, and E. Turiel (eds.), Culture, thought, and development. Mahwah, NJ: Erlbaum. Christianson, S., and Safer, M. A. (1996). Emotional events and emotions in autobiographical memories. In D. C. Rubin (ed.), Remembering our past: studies in autobiographical memory, pp. 218–244. Cambridge: Cambridge University Press. Clausen, J. A. (1995). Gender, contexts, and turning points in adults’ lives. In P. Moen, G. H. Elder, and K. Lu¨scher (eds.), Examining lives in context: perspectives on the ecology of human development, pp. 365–389. Washington, DC: American Psychological Association. Cloninger, C. R., Svrakic, N. M., and Svrakic, D. M. (1997). Role of personality self-organization in development of mental order and disorder. Development and Psychopathology, 9, 881–906. Cole, M. (1996). Cultural psychology: a once and future discipline. Cambridge, MA: The Belknap Press. Conville, R. (1998). Telling stories: dialectics of relational transition. In B. M. Montgomery and L. A. Baxter (eds.), Dialectical approaches to studying personal relationships, pp. 17–40. Mahwah, NJ: Erlbaum. Cooley, C. H. (1902). Human nature and the social order. New York: Scribner. Cooper, C. R., Jackson, J. F., Azmitia, M., and Lopez, E. M. (1998). Multiple selves, multiple worlds: three useful strategies for research with ethnic minority youth on identity, relationships, and opportunity structures. In

References

235

V.C. McLoyd and L. Steinberg (eds.), Studying minority adolescents: conceptual, methodological and theoretical issues, pp. 111–125. Hillsdale, NJ: Erlbaum. Costa, P. T., Jr., and McCrae, R. R. (1996). Mood and personality in adulthood. In C. Magai and S. McFadden (eds.), Handbook of emotion, adult development, and aging, pp. 369–383. San Diego: Academic Press. Csikszentmihalyi, M. (1990). Flow: the psychology of optimal experience. New York: Harper and Row. Csikszentmihalyi, M. (1993). The evolving self. New York: Harper Collins. Csikszentmihalyi, M., and Rathunde, K. (1998). The development of the person: an experimental perspective on the ontogenesis of psychological complexity. In R. M. Lerner (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn., vol. 1, Theoretical models of human development, pp. 635– 684. New York: Wiley. Damasio, A. R. (1994). Descartes’ error. New York: Avon. Damon, W., and Hart, D. (1982). The development of self-understanding from infancy through adolescence. Child Development, 4, 841–864. Damon, W., and Hart, D. (1988). Self-understanding in childhood and adolescence. Cambridge: Cambridge University Press. Dannefer, D., and Perlmutter, M. (1990). Development as a multidimensional process: individual and social constraints. Human Develoment, 33, 108–137. Davitz, J. R. (1969). The language of emotion. New York, Academic Press. Day, J. M., and Tappan, M. B. (1996). The narrative approach to moral development: from the epistemic subject to dialogical selves. Human Development, 39, 67–82. De Boeck, P., and Rosenberg, S. (1988). Hierarchical classes: model and data analysis. Psychometrika, 53, 361–381. De Quincey, C. (1998). Engaging presence. Noetic Sciences Review, 45, 24–27. De Rivera, J. (1992). Emotional climate. Social structure and emotional dynamics. In K. T. Strongman (ed.), International review of studies on emotion, vol. 2. New York: Wiley. De Waal, F. B. M. (1982). Chimpanzee politics: power and sex among apes. London: Jonathan Cape. De Waal, F. B. M. (1996). Good natured. Cambridge, MA: Harvard University Press. De Weerth, C., and Van Geert, P. L. C. (2000). The dynamics of emotion-related behaviors in infancy. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 324–348. Cambridge: Cambridge University Press. Deiner, E., and Larsen, R. J. (1984). Temporal stability and cross-situational consistency of affective, behavioral, and cognitive responses. Journal of Personality and Social Psychology, 47, 871–883. Demo, D. H. (1992). The self-concept over time: research issues and directions. Annual Review of Sociology, 18, 303–326. Demo, D. H., and Savin-Williams, R. C. (1992). Self-concept stability and change during adolescence. In R. P. Lipka and T. M. Brinthaupt (eds.), Self-perspectives across the life span, pp. 116–148. Albany, NY: State University of New York Press. Demos, E. (1982). Facial expressions of infants and toddlers: a descriptive analysis. In T. Field and A. Fogel (eds.), Emotion and early interaction, pp. 127–160. Hillsdale, NJ: Erlbaum.

236

References

Dennett, D. C. (1978). Brainstorms: philosophical essays on mind and psychology. Montgomery: Bradford Books. Derryberry, D., and Rothbart, M. K. (1997). Reactive and effortful processes in the organization of temperament. Development and Psychopathology, 9, 633–652. Dewey, J. (1934). Art as experience. New York: Perigee. Dewey, J. (1963). Experience and education. New York: Macmillan. (Originally published in 1938). Dunne, J. (1996). Beyond sovereignty and deconstruction: the storied self. In P. Ricoeur, The hermeneutics of action, pp. 137–157. London: Sage. Duval, S., and Wicklund, R. A. (1972). A theory of objective self-awareness. New York: Academic Press. Eckstein, S. G. (1999). A dynamic model of cognitive growth in a population: spatial tasks and conservation. Journal of Mathematical Psychology, 43, 34–70. Eisenberg, N. (1998). Introduction. In N. Eisenberg (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn., vol. 3, Social, emotional, and personality development, pp. 1–24. New York: Wiley. Ekman, P. (1972). Universals and cultural differences in facial expressions of emotions. Nebraska symposium on motivation, vol. 19, pp. 207–283. Lincoln: University of Nebraska Press. Ekman, P. (ed). (1982). Emotion in the human face, 2nd edn. New York: Cambridge University Press. Elkind, D. (1974). Children and adolescents: interpretive essays on Jean Piaget. New York: Oxford. Ellis, A. (1972) Psychotherapy without tears. In A. Burton (ed.), Twelve therapists, pp. 103–126. San Francisco: Jossey-Bass Inc. Emde, R. N., and Oppenheim, D. (1995). Shame, guilt, and the Oedipal drama: developmental considerations concerning morality and referencing of critical others. In J. P. Tangney and K. W. Fischer (eds.), Self-conscious emotions: the psychology of shame, guilt, embarrassment, and pride, pp. 413–438. New York: Guilford Press. Epstein, S. (1973). The self-concept revisited or a theory of a theory. American Psychologist, 28, 404–416. Epstein, S. (1983). The unconscious, the preconscious, and the self-concept. In J. Suls and A. Greenwald (eds.), Psychological perspectives on the self, vol. 2, pp. 219–247. Hillsdale, NJ: Erlbaum. Epstein, S. (1984). Controversial issues in emotion theory. In P. Shaver (ed.), Annual review of research in personality and social psychology, pp. 64–87. Beverly Hills, CA: Sage. Epstein, S. (1985). The implications of cognitive-experiential self-theory for research in social psychology and personality. Journal for the Theory of Social Behavior, 15, 283–310. Epstein, S. (1989). Values from the perspective of cognitive-experiential selftheory. In N. Eisenberg, J. Reykowski, and E. Staub (eds.), Social and moral values, pp. 3–22. Hillsdale, NJ: Erlbaum. Epstein, S. (1990). Cognitive-experiential self-theory. In L. Pervin (ed.), Handbook of personality theory and research: theory and research, pp. 165–192. New York: Guilford Press. Epstein, S. (1991a). Cognitive-experiential self-theory: implications for developmental psychology. In M. Gunnar and L. A. Sroufe (eds.), Self-processes and development, Minnesota Symposia on Child Psychology, vol. 23, pp. 79–123. Hillsdale, NJ: Erlbaum.

References

237

Epstein, S. (1991b). The self-concept, the traumatic neurosis, and the structure of personality. In D. Ozer, J. M. Healy, Jr., and A. J. Stewart (eds.), Perspectives in personality, vol. 3A, pp. 63–98. London: Jessica Kingsley Publishers, Ltd. Epstein, S. (1992a). Coping ability, negative self-evaluation, and overgeneralization: experiment and theory. Journal of Personality and Social Psychology, 62, 826–836. Epstein, S. (1992b). The cognitive self, the psychoanalytic self, and the forgotten selves. Comment on Drew Westen’s The cognitive self and the psychoanalytic self: can we put our selves together? Psychological Inquiry, 3, 34–37. Epstein, S. (1993a). Implications of cognitive-experiential self-theory for personality and developmental psychology. In D. Funder. R. Parke, C. Tomlinson-Keasey, and K. Widaman (eds.), Studying lives through time: personality and development, pp. 399–438. Washington, DC: American Psychological Association. Epstein, S. (1993b). Bereavement from the perspective of cognitive-experiential self-theory. In M. S. Stroebe, W. Stroebe, and R. O. Hansson (eds.), Handbook of bereavement: theory, research, and intervention, pp. 112–125. New York: Cambridge University Press. Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious. American Psychologist, 49, 709–724. Epstein, S. (1998a). Cognitive-experiential self-theory: a dual process personality theory with implications for diagnosis and psychotherapy. In R. F. Bornstein and J. M. Masling (eds.), Empirical perspectives on the psychoanalytic unconscious, vol. 7, pp. 99–140. Washington, DC: American Psychological Association. Epstein, S. (1998b). Emotions and psychopathology from the perspective of cognitive-experiential self-theory. In W. F. Flack, Jr., and J. D. Laird (eds.), Emotions in psychopathology: theory and research, pp. 57–69. New York: Oxford University Press. Epstein, S. (1998c). Cognitive-experiential self-theory. In D. Barone, M. Hersen, and V. B. Van Hasselt (eds.), Advanced personality, pp. 211–238. New York: Plenum. Epstein, S., and Pacini, R. (1999). Some basic issues regarding dual-process theories from the perspective of cognitive-experiential self-theory. In S. Chaiken, and Y. Trope (eds.), Dual-process theories in social psychology, pp. 462–482. New York: Guilford Press. Erikson, E. H. (1950). Childhood and society. New York: Norton. Erikson, E. H. (1968). Identity: youth, and crisis. New York: Norton. Feiring, C., and Taska, L. S. (1996). Family self-concept: ideas on its meaning. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 317–373. New York: Wiley. Ferrari, M. (1998). Being and becoming self-aware. In M. Ferrari and R. J. Sternberg (eds.), Self-awareness: its nature and development, pp. 387–422. New York: Guilford Press. Ferrari, M., and Mahalingam, R. (1998). Personal cognitive development and its implications for teaching and learning. Educational Psychologist, 33, 35–44. Feyereisen, P. (1989). Theories of emotions and neuropsychological research. In F. Boller and J. Grafman (ed.), Handbook of neuropsychology, vol. 3, section 6, pp. 271–281. Amsterdam: Elsevier.

238

References

Fischer, K. W. (1980). A theory of cognitive development: the control and construction of hierarchies of skills. Psychological Review, 87, 477–531. Fischer, A. H., and Jansz, J. (1995). Reconciling emotions with Western personhood. Journal for the Theory of Social Behaviour, 25, 59–81. Fischer, K. W., Shaver, P. R., and Carnochan, P. (1990). How emotions develop and how they organise development. Cognition and Emotion, 4, 81–127. Fischer, K. W., and Tangney, J. P. (1995). Self-conscious emotions and the affect revolution: framework and overview. In J. P. Tangney and K. W. Fischer (eds.), Self-conscious emotions: the psychology of shame, guilt, embarrassment, and pride, pp. 3–22. New York: Guilford Press. Fogel, A. (1990). The process of developmental change in infant communicative action: using dynamic systems theory to study individual ontogenies. In J. Colombo and J. W. Fagen (eds.), Individual differences in infancy: reliability, stability, prediction, pp. 341–358. Hillsdale, NJ: Erlbaum. Fogel, A. (1993). Developing through relationships: origins of communication, self, and culture. New York: Harvester Wheatsheaf. Fogel, A. (1995). Relational narratives of the pre-linguistic self. In P. Rochat (ed.), The self in early infancy: theory and research, pp. 117–139. Amsterdam: Elsevier North Holland. Fogel, A. (2001). Infancy: Infant, family, and society, 4th edn. Belmont, CA: Wadsworth. Fogel, A. (unpublished). Reply to Jansz. Salt Lake City: University of Utah, Department of Psychology. Fogel, A., and Branco, A. U. (1997). Meta-communication as a source of indeterminism in relationship development. In A. Fogel, M. Lyra, and J. Valsiner (eds.), Dynamics and indeterminism in developmental and social processes, pp. 65–92. Hillsdale, NJ: Erlbaum. Fogel, A., Dickson, K. L., Hsu, H., Messinger, D., Nelson-Goens, G. C., and Nwokah, E. (1997). Communicative dynamics of emotion. In K. C. Barrett (ed.), New directions in child development: the communication of emotion, Current research from diverse perspectives, vol. 77, pp. 5–24. San Francisco: Jossey-Bass. Fogel, A., and Lyra, M. (1997). Dynamics of development in relationships. In F. Masterpasqua and P. Perna (eds.), The psychological meaning of chaos: selforganization in human development and psychotherapy, pp. 75–94. Washington, DC: American Psychological Association. Fogel, A., Nwokah, E., Dedo, J. Y., and Messinger, D. (1992). Social process theory of emotion: a dynamic systems approach. Social Development, 1, 122–142. Fogel, A., Nwokah, E., and Karns, J. (1993). Parent–infant games as dynamic social systems. In K. B. MacDonald (ed.), Parent–children play, pp. 43–70. Albany, NY: SUNY Press. Fogel, A., and Thelen. E. (1987). Development of early expressive and communicative action: reinterpreting the evidence from a dynamic systems perspective. Developmental Psychology, 23, 747–761. Freeman, W. J. (1995). Societies of brains: a study in the neuroscience of love and hate. Hillsdale, NJ: Erlbaum. Freeman, W. J. (2000). Emotion is essential to all intentional behaviors. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 209–235. New York: Cambridge University Press. Freeman, W. J., and Baird, B. (1987). Relation of olfactory EEG to behavior: spatial analysis. Behavioral Neuroscience, 101, 393–408.

References

239

Freeman, M. and Brockmeier, J. (in press). Narrative integrity: autobiographical identity and the meaning of the ‘‘good life.’’ In J. Brockmeier and D. Carbaugh (eds.), Narrative and identity: studies in autobiography, self, and culture. Fridlund, A. J. (1994). Human facial expression: an evolutionary view. New York: Academic Press. Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press. Frijda, N. H. (1993a). Moods, emotion episodes, and emotions. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 381–403. New York: Guilford Press. Frijda, N. H. (1993b). The place of appraisal in emotion. Cognition and Emotion, 7, 357–387. Frijda, N. H. (1994). The Lex Talionis: on vengeance. In S. H. M. Van Goozen, N. E. Van de Poll, and J. A. Sergeant (eds.), Emotions: essays on emotion theory, pp. 263–290. Hillsdale, NJ: Erlbaum. Frijda, N. H. (2000). The nature of pleasure. In J. Bargh (ed.), Unravelling the complexities of social life: a festschrift in honor of R. B. Zajonc, pp. 71–94. Washington, DC: American Psychological Association. Frijda, N. H. (unpublished). Reply to Epstein. Amsterdam: University of Amsterdam, Department of Psychology. Frijda, N. H., Kuipers, P., and Terschure, E. L. (1989). Relations between emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57, 212–228. Frijda, N. H., Mesquita, B., Sonnemans, J., and Van Goozen, J. (1991). The duration of affective phenomena or emotions, sentiments, and passions. In K. T. Strongman (ed.), International review of studies on emotion, vol. 1, pp. 187–225. Chichester: Wiley. Frijda, N. H., and Tcherkassof, A. (1997). Facial expression and modes of action readiness. In Russell, J. A. and Fernandez-Dols, J. M. (eds.), The psychology of facial expression, pp. 78–102. Cambridge: Cambridge University Press. Ganguly, S. N. (1976). Communication, identity, and human development. Communication, 2, 221–244. Gecas, V., and Mortimer, J. T. (1987). Stability and change in the self-concept from adolescence to adulthood. In T. Honess and K. Yardley (eds.), Self and identity: perspectives across the lifespan, pp. 265–286. London: Routledge and Kegan Paul. Gergen, K. (1984). Theory of the self: impasse and evolution. In L. Berkowitz (ed.), Advances in experimental social psychology, vol. 17, pp. 49–115. New York: Academic Press. Gergen, M. M. (1991). The emergence of the relational self. New York: Basic Books. Gergen, M. M., and Gergen, K. J. (1995). What is this thing called love? Emotional scenarios in historical perspective. Journal of Narrative and Life History, 5, 221–237. Gibson, J. J. (1966). The senses considered as perceptual systems. Boston, MA: Houghton Mifflin Company. Gibson, J. J. (1979). The ecological approach to visual perception systems. Boston, MA: Houghton Mifflin Company. Ginsburg, G. P. (1985). The analysis of human action: current status and future potential. In G. P. Ginsburg, M. L. Brenner and M. von Cranach (eds.),

240

References

Discovery strategies in the psychology of action, pp. 255–279. London: Academic Press. Goertzel, B. (1995). Belief systems as attractors. In R. Robertson and A. Combs (eds.), Chaos theory in psychology and the life sciences, pp. 123–134. Mahwah, NJ: Erlbaum. Goffman, E. (1974). Frame analysis: an essay on the organization of experience. Cambridge, MA: Harvard University Press. Goldsmith, H. H. (1994). Parsing the emotional domain from a developmental perspective. In P. Ekman and R. J. Davidson (eds.), The nature of emotion, pp. 68–73. New York: Oxford University Press. Goleman, D. (1995). Emotional intelligence. New York: Bantam Books. Goodnow, J. J. (1996). From household practices to parents’ ideas about work and interpersonal relationships. In S. Harkness and C. Super (eds.), Parents’ cultural belief systems: their origins, expressions, and consequences, pp. 313–344. New York: Guilford Press. Goodwin, B. C. (1987). Developing organisms as self-organizing fields. In F. E. Yates (ed.), Self-organizing systems: the emergence of order, pp. 167–180. New York: Plenum. Gordon, C., and Gergen, K. J. (eds.) (1968). The self in social interaction. New York: Wiley. Greenberg, J., Solomon, S., Pyszczynski, T., Rosenblatt, A., Burling, J., Lyon, D., Simon, L., and Pinel, E. (1992). Why do people need self-esteem? Converging evidence that self-esteem serves an anxiety-buffering function. Journal of Personality and Social Psychology, 6, 913–922. Greenwald, A. G., and Pratkanis, A. R. (1984). The self. In R. S. Wyer and T. K. Scrull (eds.), Handbook of social cognition, vol. 3, pp. 129–178. Hillsdale, NJ: Erlbaum. Grotevant, H. D., and Cooper, C. R. (1986). Individuation in family relationships. Human Development, 29, 82–100. Harkness, K. L., and Tucker, D. M. (2000). Motivation of neural plasticity: neural mechanisms in the self-organization of depression. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 186–208. New York: Cambridge University Press. Harre´, R. (1983). Personal being: a theory for individual psychology. Oxford: Blackwell. Harre´, R. (1987). The social construction of selves. In K. Yardley and T. Honess (eds.), Self and identity: psychosocial perspectives, pp. 41–52. New York: Wiley. Harre´, R. (1998). The singular self: an introduction to the psychology of personhood. London: Sage. Harre´, R., and Van Langenhove, L. (1991). Varieties of positioning. Journal for the Theory of Social Behaviour, 21, 393–407. Hart, D., and Edelstein, W. (1992). Self-understanding development in crosscultural perspective. In T. M. Brinthaupt, and R. P. Lipka (eds.), The self: definitional and methodological issues, pp. 291–322. Albany, NY: State University of New York Press. Harter, S. (1983). Developmental perspectives on the self-system. In P. H. Mussen (ed.), Handbook of child psychology, vol. 4, pp. 275–385. New York: Wiley. Harter, S. (1985). Manual for the self-perception profile for children. Denver, CO: University of Denver.

References

241

Harter, S. (1996). Historical roots of contemporary issues involving selfconcept. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 38–90. New York: Wiley. Harter, S. (1998). The development of self-representations. In N. Eisenberg (vol. ed.; W. Damon, general ed.). Handbook of child psychology, 5th edn., vol. 3, Social, emotional, and personality development, pp. 553–617. New York: Wiley. Harter, S. (1999). The construction of the self: a developmental perspective. New York: Guilford Press. Harter, S., and Marold, D. B. (1989). A model of the determinants and mediational role of self-worth: implications for adolescent depression and suicidal ideation. In J. Strauss and G. R. Goethals (eds.), The self: interdisciplinary approaches, pp. 66–92. New York: Springer. Hattie, J. (1992). Self-concept. Hillsdale, NJ: Erlbaum. Hattie, J., and Marsh, H. W. (1996). Future directions in self-concept research. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 421–462. New York: Wiley. Haviland, J. M. (1976). Looking Smart. In M. Lewis (ed.), Origins of Intelligence, pp. 353–378. New York: Plenum Press. Haviland, J. M., Davidson, R. B., Ruetsch, C., Gebelt, J. L., and Lancelot, C. (1994). The place of emotion in identity. Journal of research in adolescence, 4, 503–518. Haviland, J. M., and Kahlbaugh, P. (1993). Emotion and identity. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 327–339. New York: Guilford Press. Haviland, J. M., and Kramer, D. A. (1991). Affect-cognition relations in an adolescent diary: the case of Anne Frank. Human Development, 34, 143–159. Haviland, J. M., and Lelwica, M. (1987). The induced affect response: 10-weekold infants’ responses to three emotion expressions. Developmental Psychology, 23, 97–104. Haviland-Jones, J. M., Boulifard, D., and Magai, C. (unpublished manuscript). New models of emotion and identity development: towards dynamic systems. Held, R. (1965). Plasticity in sensory-motor systems. Scientific American, 213, 84–94. Hermans, H. J. M. (1987a). Self as an organized system of valuations: toward a dialogue with the person. Journal of Counseling Psychology, 34, 10–19. Hermans, H. J. M. (1987b). The dream in the process of valuation: a method of interpretation. Journal of Personality and Social Psychology, 53, 163–175. Hermans, H. J. M. (1996a). Voicing the self: from information processing to dialogical interchange. Psychological Bulletin, 119, 31–50. Hermans, H. J. M. (1996b). Opposites in a dialogical self: constructs as characters. The Journal of Constructivist Psychology, 9, 1–26. Hermans, H. J. M., and Hermans-Jansen, E. (1995). Self-narratives: the construction of meaning in psychotherapy. New York: Guilford Press. Hermans, H. J. M., and Kempen, H. J. G. (1993). The dialogical self: meaning as movement. San Diego, CA: Academic Press. Hermans, H. J. M., Kempen, H. J. G., and Van Loon, R. J. P. (1992). The dialogical self: beyond individualism and rationalism. American Psychologist, 47, 23–33. Heymans, P. (1992). Lifespan-learning: developmental tasks and their manage-

242

References

ment. In A. Tuynman and M. Van der Kamp (eds.), Learning across the lifespan: theories, research and policies, pp. 33–51. Oxford: Pergamon Press. Heymans, P. (1994). Developmental tasks: a cultural analysis of human development. In J. ter Laak, A. Podolski, and P. Heymans (eds.), Developmental tasks: towards a cultural analysis of human development, pp. 3–33. Dordrecht: Kluwer. Heymans, P. (2000). Talent for development: responding to contextual promises. In C. F. M. Van Lieshout and P. G. Heymans (eds.), Developing talent across the lifespan. Hove: Psychology Press. Higgins, E. T. (1987). Self-discepancy: a theory relating self and affect. Psychological Review, 94, 319–340. Higgins, E. T. (1996). The ‘‘self-digest’’: self-knowledge serving self-regulatory functions. Journal of Personality and Social Psychology, 71, 1062–1083. Higgins, E. T., Vookles, J., and Tykocinski, O. (1992). Self and health: how ‘‘patterns’’ of self-beliefs predict types of emotional and physical problems. Social Cognition, 10, 125–150. Hochschild, A. (1983). The managed heart. Berkeley: University of California Press. Hofbauer, J., and Sigmund, K. (1988). The theory of evolution and dynamical systems. Cambridge: Cambridge University Press. Holland, J. H. (1998). Emergence: from chaos to order. Reading, MA: Helix Books/ Addison-Wesley. Hoorens, V., and Nuttin, J. M. (1993). Overvaluation of own attributes: mere ownership or subjective frequency? Social Cognition, 11, 177–200. Hopkins, B., and Butterworth, G. (1997). Dynamical systems approaches to the development of action. In G. Bremner, A. Slater, and G. Butterworth (eds.), Infant development: recent advances, pp. 75–100. Hove: Psychology Press. Horowitz, M. J. (1992). Stress response syndromes, 3rd edn. Northvale, NJ: Jason Aronson. Horowitz, M. J. (1998). Cognitive psychodynamics: from conflict to character. New York: Wiley. Huntley, C. W. (1940). Judgments of self based upon records of expressive behavior. Journal of Abnormal Social Psychology, 35, 398–427. Isen, A. M. (1990). The influence of positive and negative affect on cognitive organization: some implications for development. In N. Stein, B. Leventhal, and T. Trabasso (eds.), Psychological and biological approaches to emotion, pp. 75–94. Hillsdale, NJ: Erlbaum. Izard, C. E. (1971). The face of emotion. New York: Appleton-Century-Crofts. Izard, C. E. (1984). Emotion-cognition relationships and human development. In C. E. Izard, J. Kagan, and R. B. Zajonc (eds.), Emotions, cognition and behavior, pp. 17–37. Cambridge: Cambridge University Press. Izard, C. E. (1992). Basic emotions, relations among emotions, and emotioncognition relations. Psychological Review, 99, 561–565. Izard, C. E. (1993). Four systems of emotion activation. Psychological Review, 100, 68–90. Izard, C. E., Ackerman, B., Schoff, K., and Fine, S. (2000). Self-organization of discrete emotions, emotion patterns, and emotion-cognition relations. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 15–36. New York: Cambridge University Press. Jackson, E. A. (1991a). Perspectives of nonlinear dynamics, vol. 1. Cambridge: Cambridge University Press.

References

243

Jackson, E. A. (1991b). Perspectives of nonlinear dynamics, vol. 2. Cambridge: Cambridge University Press. James, W. (1976). Essays in radical empiricism. Cambridge, MA, and London: Harvard University Press. James, W. (1890). The principles of psychology. New York: Holt. James, W. (1910). Psychology: the briefer course. New York: Holt. Jansz, J. (1995). Self-narratives as personal structures of meaning. In A. Oosterwegel and R. A. Wicklund (eds.), The self in European and North American culture: development and processes, pp. 65–72. Dordrecht: Kluwer. Jansz, J., and Timmers, M. (in press). Emotional dissonance: when emotions jeopardize identity. Theory and Psychology. Jaquish, G. A., and Savin-Williams, R. C. (1981). Biological and ecological factors in the expression of adolescent self-esteem. Journal of Youth and Adolescence, 10, 473–485. Jaynes, J. (1976). The origin of consciousness in the breakdown of the bicameral mind. Boston: Houghton Mifflin. Johnson, F. (1985). The Western concept of self. In A. J. Marsella, G. de Vos, and F. L. K. Hsu (eds.), Culture and self: Asian and Western perspectives, pp. 91– 138. New York: Tavistock Publications. Johnson, M. (1987). The body in the mind: the bodily basis of meaning, imagination, and reason. Chicago: University of Chicago Press. Jones, D. (1990). The matrix of narrative: family systems and the semiotics of story. New York: Mouton de Gruyter. Jopling, D. (1993). Cognitive science, other minds and the philosophy of dialogue. In U. Neisser (ed.), The perceived self, pp. 290–309. Cambridge: Cambridge University Press. Josephs, I. E. (1998). Constructing one’s self in the city of the silent: dialogue, symbols, and the role of ‘‘as-if’’ in self-development. Human Development, 41, 180–195. Josselson, R. (1994). Identity and relatedness in the life cycle. In H. A. Bosma, T. L. G. Graafsma, H. D. Grotevant, and D. J. de Levita (eds.), Identity and development: an interdisciplinary approach, pp. 81–102. Thousand Oaks, CA: Sage. Josselson, R. (1995). Imagining the real: empathy, narrative, and the dialogic self. The Narrative Study of Lives, 3, 27–44. Juhasz, A. (1992). Significant others in self-esteem development: methods and problems in measurement. In T. M. Brinthaupt and R. P. Lipka (eds.), The self: definitional and methodological issues, pp. 204–235. Albany, NY: State University of New York Press. Kagan, J. (1981). The second year: the emergence of self-awareness. Cambridge, MA: Harvard University Press. Kahlbaugh, P. E. and Haviland, J. M. (1994) Nonverbal communication between parents and adolescents: a study of approach and avoidance behaviors. Journal of Nonverbal Behavior, 18, 91–113. Kaplan, H. B. (1986). Social psychology of self-referent behavior. New York: Plenum Press. Kauffman, S. A. (1993). The origins of order: self-organization and selection in evolution. NewYork: Oxford University Press. Kaye, K. (1982). The mental and social life of babies. Chicago: University of Chicago Press. Kegan, R. (1982). The evolving self: problem and process in human development.

244

References

Cambridge, MA: Harvard University Press. Keith, L. K., and Bracken, B. A. (1996). Self-concept instrumentation: a historical and evaluative review. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 91–170. New York: Wiley. Kelly, G. A. (1955/1963). A theory of personality: the psychology of personal constructs. New York: Norton. Kelso, J. A. S. (1984). Phase transitions and critical behavior in human bimanual coordination. American Journal of Physiology, 15, A1000–A1004. Kendon, A. (1985). Behavioral foundations for the process of frame attunement in face-to-face interaction. In G. P. Ginsburg, M. Brenner, and M. von Cranach (eds.), Discovery strategies in the psychology of action, pp. 229–253. Orlando: Academic Press. Kernis, M. H. (1993). The roles of stability and level of self-esteem in psychological functioning. In R. F. Baumeister (ed.), Self-esteem: the puzzle of low self-regard, pp. 167–182. New York: Plenum Press. Kernis, M. H., Grannemann, B. D., and Barclay, L. C. (1989). Stability and level of self-esteem as predictors of anger arousal and hostility. Journal of Personality and Social Psychology, 56, 1013–1022. Kernis, M. H., and Waschull, S. B. (1995). The interactive roles of stability and level of self-esteem: research and theory. In M. P. Zanna (ed.), Advances in experimental social psychology, vol. 27, pp. 93–141. London: Academic Press. Kindermann, T. A., and Skinner, E. A. (1992). Modeling environmental development: Individual and contextual trajectories. In J. B. Asendorpf and J. Valsiner (eds.), Stability and change in development: a study of methodological reasoning, pp. 155–197. Newbury Park, CA: Sage. Kingsland, S. E. (1995). Modeling nature: episodes in the history of population ecology. Chicago: University of Chicago Press. Kitayama, S., Markus, H. R., and Matsumoto, H. (1995). Culture, self, and emotion: a cultural perspective on ‘‘self-conscious’’ emotions. In J. P. Tangney and K. W. Fischer (eds.), Self-conscious emotions, pp. 439–464. New York: Guilford Press. Klages, L. (1948). Charakterkunde (Characterology). Zu¨rich: Hirzel. Kohlberg, L. (1976). Moral stages and moralization: the cognitive-developmental approach. In T. Lickona (ed.), Moral development and behavior: theory, research and social issues, pp. 31–53. New York: Holt, Rinehart and Winston. Kuiper, N. A., and Rogers, T. B. (1979). Encoding of personal information: self-other differences. Journal of Personality and Social Psychology, 37, 499– 514. Kunnen, E. S., and Bosma, H. A. (1994, June). A growth model of commitments based on a dynamic systems perspective. Paper presented at the Meetings of the Society for the Study of Behavioral Development, Amsterdam, The Netherlands. Kunnen, E. S., and Bosma, H. A. (2000). Development of meaning making: a dynamic systems conceptualization. New Ideas in Psychology, 18, 57–82. Kunnen, E. S., Bosma, H. A., and Van Geert, P. L. C. (in press). A dynamic systems approach to identity formation: theoretical background and methodological possibilities. In J.-E. Nurmi (ed.), Navigating through adolescence: European perspectives. Michigan State University Series on Children, Youth and Families. New York: Garland Publishing Kunnen, E. S., and Klein Wassink, M. E. (1999). The transitional process in adult meaning making. Presentation at the XIth European Conference on Devel-

References

245

opmental Psychology, September 1–5. Spetses, Greece. l’Ecuyer, R. (1992). An experiential-developmental framework and methodology to study the transformations of the self-concept from infancy to old age. In T. M. Brinthaupt and R. P. Lipka (eds.), The self: definitional and methodological issues, pp. 96–134. Albany, NY: State University of New York Press. Labouvie-Vief, G., Orwell, L., and Manion, M. (1995). Narratives of mind, gender, and the life-course. Human Development, 38, 239–257. Lagerspetz, K. (1961). Genetic and social causes of aggressive behavior in mice. Scandinavian Journal of Psychology, 2, 167–173. Lakoff, G. and Johnson, M. (1999). Philosophy in the flesh. New York: Basic Books. Landauer, R. (1987). Role of relative stability in self-repair and self-maintenance. In F. E. Yates (ed.), Self-organizing systems: the emergence of order, pp. 435–443. New York: Plenum. Lazarus, R. S. (1991a). Emotion and adaptation. New York: Oxford University Press. Lazarus, R. S. (1991b). Cognition and motivation in emotion. American Psychologist, 46, 352–367. LeDoux, J. E. (1989). Cognitive-emotional interactions in the brain. Cognition and Emotion, 3, 267–289. Legerstee, M., Anderson, D., and Schaffer, A. (1998). Five- and eight-month-old infants recognize their faces and voices as familiar and social stimuli. Child Development, 69, 37–50. Leifer, M. (1980). Psychological effects of motherhood: a study of first pregnancy. New York: Praeger. Leont’ev, A. N. (1978). Activity, consciousness, and personality. Trans. M. J. Hall. Englewood Cliffs, NJ: Prentice-Hall. Levinas, E. (1969). Totality and infinity. Trans. A. Lingus. Pittsburgh: Duquesne University Press. Levine, R. L., and Fitzgerald, H. E. (1992a). Analysis of dynamic psychological systems: basic approaches to general systems, dynamic systems and cybernetics. New York: Plenum Press. Levine, R. L., and Fitzgerald, H. E. (1992b). Analysis of dynamic psychological systems: methods and applications. New York: Plenum Press. Lewis, C. S. (1967). Studies in words, 2nd edn. Cambridge: Cambridge University Press. Lewis, M. (1992). Shame, the exposed self. New York: Free Press. Lewis, M. (1993). The emergence of human emotions. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 223–235. New York: Guilford Press. Lewis, M., and Brooks-Gunn, J. (1979). Social cognition and the acquisition of self. New York: Plenum Press. Lewis, M., Sullivan, M. W., Stanger, C., and Weiss, M. (1989). Self development and self-conscious emotion. Child Development, 60, 146–156. Lewis, M. D. (1995). Cognition-emotion feedback and the self-organization of developmental paths. Human Development, 38, 71–102. Lewis, M. D. (1996). Self-organising cognitive appraisals. Cognition and Emotion, 10, 1–25. Lewis, M. D. (1997). Personality self-organization: cascading constraints on cognition-emotion interaction. In A. Fogel, M. C. Lyra, and J. Valsiner (eds.), Dynamics and indeterminism in developmental and social processes,

246

References

pp. 193–216. Mahwah, NJ: Erlbaum. Lewis, M. D. (2000a). Emotional self-organization at three time scales. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 23–69. New York: Cambridge University Press. Lewis, M. D. (2000b). The promise of dynamic systems approaches for an integrated account of human development. Child Development, 71, 36–43. Lewis, M. D., and Douglas, L. (1998). A dynamic systems approach to cognition-emotion interactions in development. In M. F. Mascolo and S. Griffin (eds.), What develops in emotional development?, pp. 159–188. New York: Plenum. Lewis, M. D., and Ferrari, M. D. (unpublished). Reply to Carol Magai. Toronto: OISE/University of Toronto, Department of Human Development and Applied Psychology. Lewis, M. D., and Granic, I. (1999a). Who put the self in self-organization? A clarification of terms and concepts for developmental psychopathology. Development and Psychopathology, 11, 365–374. Lewis, M. D., and Granic, I. (1999b). Self-organization of cognition-emotion interactions. In T. Dalgleish and M. J. Power (eds.), Handbook of cognition and emotion, pp. 683–701. Chichester, UK: John Wiley and Sons Ltd. Lewis, M. D., Lamey, A. V., and Douglas, L. (1999). A new dynamic systems method for analyzing early socio emotional development. Developmental Science, 2, 457–475. Lightfoot, C. (1997). Transforming the canonical cowboy: notes on the determinacy and indeterminacy of children’s play and cultural development. In A. Fogel, M. C. Lyra, and J. Valsiner (eds.), Dynamics and indeterminism in developmental and social processes, pp. 165–174. Mahwah, NJ: Erlbaum. Locke, J. (1690). An essay concerning human understanding. New York: Penguin, 1998. Loewald, H. W. (1980). Papers on psychoanalysis. New Haven and London: Yale University Press. Logan, R. D. (1987). Historical change in prevailing sense of self. In K. Yardley and T. Honess (eds.), Self and identity: psychosocial perspectives, pp. 13–26. New York: Wiley. Lucariello, J. (1995). Mind, culture, person: elements in a cultural psychology. Human Development, 38, 2–18. Lyra, M., and Winegar, L. T. (1997). Processual dynamics of interation through time: adult–child interactions and process of development. In A. Fogel, M. Lyra, and J. Valsiner (eds.), Dynamics and indeterminism in developmental and social processes, pp. 93–109. Mahwah, NJ: Erlbaum. Maclean, P. (1990). The triune brain in evolution: role in palaeocerebral functions. New York: Plenum Press. Magai, C. (1999). Personality change in adulthood: loci of change and the role of interpersonal process. International Journal of Aging and Human Development, 49, 339–352. Magai, C., and Haviland-Jones, J. M. (in press). Emotion and personality: dynamic trajectories of lives. New York: Cambridge University Press. Magai, C., and Hunziker, J. (1993). Tolstoy and the riddle of developmental transformation: a lifespan analysis of the role of emotions in personality development. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 247–259. New York: Guilford Press.

References

247

Magai, C., and McFadden, S. H. (1995). The role of emotions in social and personality development: history, theory, and research. New York: Plenum Press. Magai, C., and Nusbaum, B. (1996). Personality change in adulthood: dynamic systems, emotions, and the transformed self. In C. Magai and S. H. McFadden (eds.), Handbook of emotion, adult development, and aging, pp. 403–420. San Diego: Academic Press. Magnusson, D., and Stattin, H. (1998). Person-context interaction theories. In R. M. Lerner (vol. ed.; general ed. W. Damon), Handbook of child psychology, 5th edn., vol. 1, Theoretical models of human development, pp. 685–759. New York: Wiley. Malatesta, C. (1990). The role of emotion in the development and organization of personality. In R. Dienstbier and R. Thompson (eds.), Nebraska Symposium on Motivation, vol. 36, pp. 1–56. Lincoln: University of Nebraska Press. Malatesta, C., Culver, C., Tesman, J., and Shepard, B. (1989). The development of emotion expression during the first two years of life. Monographs of the Society for Research on Child Development, 54, nos. 1 and 2, pp. 1–104. Chicago: University of Chicago Press. Malatesta, C., and Haviland, J. M. (1982). Learning display rules: the socialization of emotion expression in infancy. Child Development, 53, 991–1003. Malatesta C. Z., and Wilson, A. (1988). Emotion cognition interaction in personality development: a discrete emotions, functionalist analysis. British Journal of Social Psychology, 27, 91–112. Malatesta-Magai, C., and Culver, C. (1991). Identification of emotions, moods, and personality via diary data. In K. T. Strongman (ed.), International Review of Studies on Emotion, pp. 163–186. New York: John Wiley and Sons. Malatesta-Magai, C., Jonas, R., Shepard, B., and Culver, C. (1992). Type A personality and emotional expressivity in younger and older adults. Psychology and Aging, 7, 551–561. Marcia, J. E., (1980). Identity in adolescence. In J. Adelson (ed.), Handbook of adolescent psychology, pp. 159–187. New York: Wiley. Marcia, J. E., Waterman, A. S., Matteson, D. R., Archer, S. L., and Orlofsky, J. L. (1993). Ego identity: a handbook for psychosocial research. New York: Springer. Markus, H. R. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63–78. Markus, H. R., and Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253. Markus, H. R., and Kunda, Z. (1986). Stability and malleability of the selfconcept. Journal of Personality and Social Psychology, 51, 858–866. Markus, H. R., and Nurius, P. (1986). Possible selves. American Psychologist, 41, 954–969. Markus, H. R., and Sentis, K. (1982). The self in social information processing. In J. Suls (ed.), Psychological perspectives on the self, vol. 1, pp. 41–70. Hillsdale, NJ: Erlbaum. Markus, H. R., and Wurf, E. (1987). The dynamic self-concept: a social psychological perspective. Annual Review of Psychology, 38, 299–337. Marsh, H. W., Byrne, B. M., and Shavelson, R. J. (1992). A multidimensional, hierarchical self-concept. In T. M. Brinthaupt and R. P. Lipka (eds.), The self: definitional and methodological issues, pp. 44–95. Albany, NY: State University of New York Press. Marsh, H. W., and Hattie, J. (1996). Theoretical perspectives on the structure of

248

References

self-concept. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 38–90. New York: Wiley. Mascolo, M. F., and Fischer, K. W. (1998). The development of self through the coordination of component systems. In M. D. Ferrari and R. J. Sternberg (eds.), Self-awareness: its nature and development, pp. 332–384. New York: Guilford Press. Mascolo, M. F., Harkins, D., and Harakal, T. (2000). The dynamic construction of emotion: varieties in anger. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 125–152. NewYork: Cambridge University Press. Maturana, H. R., and Varela, F. J. (1987). The tree of knowledge: the biological roots of human understanding. Boston: New Science Library. McAdams, D. P. (1985). Power, intimacy, and the life story: personological inquiries into identity. Chicago: Dorsey Press (reprinted by Guilford Press). McAdams, D. P. (1983). The stories we live by: personal myths and the making of the self. New York: Morrow. McAdams, D. P., Diamond, A., de St. Aubin, E., and Mansfield, E. ( 1997). Stories of commitment: the psychosocial construction of generative lives. Journal of Personality and Social Psychology, 72, 678–694. McClelland, D. C. (1984). Motives, personality, and society. New York: Praeger. Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press. Meltzoff, A. N., and Moore, M.K. (1979). Interpreting ‘‘imitiative’’ responses in early infancy. Science, 205, 217–219. Merleau-Ponty, M. (1945). Pheńomeńologie de la perception [Phenomenology of perception]. Paris: Gallimard; trans. Colin Smith (1962), Phenomenology of perception. London: Routledge and Kegan Paul. Merleau-Ponty, M. (1968). The visible and the invisible. Trans. A. Lingis. Evanston, IL: Northwestern University Press. Michaels, C. F., and Carello, C. (1981). Direct perception. Englewood Cliffs, NJ: Prentice-Hall. Michel, G. F., Camras, L. A., and Sullivan, J. (1992). Infant interest expressions as coordinative motor structures. Infant Behavior and Development, 15, 347–358. Miller, P. J., Mintz, J., Hoogstra, L., Fung, H., and Potts, R. (1992). The narrated self: young children’s construction of self in relation to others in conversational stories of personal experience. Merrill-Palmer Quarterly, 38, 45–67. Miller, W. I. (1997). The anatomy of disgust. Cambridge, MA: Harvard University Press. Mischel, W., and Shoda, Y. (1995). A cognitive-affective system theory of personality: reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246–268. Mistry, J. (1993). Cultural context in the development of children’s narratives. In J. Altarriba (ed.), Cognition and culture: a cross-cultural approach to psychology, pp. 207–228. Amsterdam: Elsevier Science Publishers. Moen, P., Elder, G. H., and Lu¨scher, K. (eds.) (1995). Examining lives in context: perspectives on the ecology of human development. Washington, DC: American Psychological Association. Molenaar, P. C. M. (1985). A dynamic factor model for the analysis of multivariate time series. Psychometrika, 50, 181–202. Molenaar, P. C. M., and Newell, K. M. (1998). Applications of non-linear dynamics to developmental process modeling. Mahwah, NJ: Erlbaum. Moreno, J. L. (1946). Psychodrama, vol. 1. New York: Beacon House.

References

249

Moreno, J. L. (1959). Psychodrama, vol. 2, Foundations of psychotherapy. New York: Beacon House. Moreno, J. L., and Elefthery, D. G. (1982). An introduction to group psychodrama. In G. M. Gazda (ed.), Basic approaches to group psychotherapy and group counseling, 3rd edn., pp. 101–131. Springfield, IL: Charles C. Thomas. Moreno, J. L., and Moreno, Z. T. (1969). Psychodrama, vol. 3, Action therapy and principles of practice. New York: Beacon House. Morris, W. N. (1989). Mood: the frame of mind. New York: Springer. Msihara, B. L. (1996). A dynamic developmental model of suicide. Human Development, 39, 181–194. Mummendey, H. D. (ed.) (1988). Verhalten und Einstellung (Behavior and attitude). Berlin: Springer. Murray, J. D. (1989). Mathematical biology. Berlin: Springer. Murray, K. (1985). Life as fiction. Journal for the Theory of Social Behaviour, 15, 173–188. Neisser, U. (1991). Five kinds of self-knowledge. In D. Kolak and R. Martin (eds.), Self and identity: contemporary philosophical issues, pp. 386–406. Toronto: Cassell and Collier Macmillan. Neisser, U. (ed.) (1993). The perceived self: ecological and interpersonal sources of self-knowledge. Cambridge: Cambridge University Press. Nelson, K. (1992). Emergence of autobiographical memory at age 4. Human Development, 35, 172–177. Nesselroade, J. R. (1991). Interindividual differences in intraindividual change. In L. M. Collins and J. L. Horn (eds.), Best methods for the analysis of change, pp. 92–105. Washington DC: American Psychological Association. Niedenthal, P. M., Tangney, J. P., and Gavanski, I. (1994). ‘‘If only I weren’t’’ versus ‘‘if only I hadn’t’’: distinguishing shame and guilt in counterfactual thinking. Journal of Personality and Social Psychology, 67, 585–595. Nolen-Hoeksema, S., and Morrow, J. (1991). A prospective study of depression and distress following a natural disaster: the 1989 Loma Prieta earthquake. Journal of Personality and Social Psychology, 61, 115–121. Novak, G. (1996). Developmental psychology: dynamic systems and behavior analysis. Reno, NV: Context Press. Nowak, A., and Vallacher, R. R. (1999). Dynamical social psychology. London: Guilford Press. Nowak, A., Vallacher, R. R., Tesser, A., and Borkowski, W. (2000). Society of self: the emergence of collective properties in self-structure. Psychological Review, 107, 39–61. Oatley, K. (1992). Best laid schemes: the psychology of emotions. Cambridge: Cambridge University Press. Oatley, K., and Johnson-Laird, P. N. (1987). Towards a cognitive theory of emotions. Cognition and Emotion, 1, 29–50. Ohman, A. (1993). Fear and anxiety as emotional phenomenon. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 511–536. New York: Guilford. Oosterwegel, A. (1992). The organization of the self-system: developmental changes in childhood and adolescence. University of Amsterdam, unpublished doctoral dissertation. Oosterwegel, A. (September 1996). Away from a rational self in psychology: how re-inventing the wheel can provide clarity. Paper presented at a Meeting of the Frankfurt/Southampton Link, Frankfurt.

250

References

Oosterwegel, A., and Wicklund, R. A. (eds.) (1995). The self in European and North American culture: development and processes. Dordrecht: Kluwer. Ormel, J., and Schaufeli, W. B. (1991). Stability and change in psychological distress and their relationship with self-esteem and locus of control: a dynamic equilibrium model. Journal of Personality and Social Psychology, 60, 288–299. Ortony, A., Clore, G., and Collins, A. (1988). The cognitive structure of emotions. Cambridge: Cambridge University Press. Overton, W. F. (1998). Developmental psychology: philosophy, concepts, and methodology. In R. M. Lerner (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn., vol. 1, Theoretical models of human development, pp. 107–188. New York: Wiley. Pacini, R., Muir, F., and Epstein, S. (1998). Depressive realism from the perspective of cognitive-experiential self-theory. Journal of Personality and Social Psychology, 74, 1056–1068. Panksepp, J. (1998). Affective neuroscience: the foundations of human and animal emotions. New York: Oxford University Press. Panksepp, J. (2000a) Emotions as natural kinds within the mammalian brain. In M. Lewis and J. M. Haviland-Jones (eds.), The Handbook of Emotion, 2nd edn., pp. 137–156. New York: Guilford Press. Panksepp, J. (2000b). The neuro dynamics of emotions: an evolutionary-neuro developmental view. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 236–264. New York: Cambridge University Press. Pedrosa, M. I., Carvalho, A. M. A., and Imperio-Hamburger, A. (1997). From disordered to ordered movement: attractor configuration and development. In A. Fogel, M. C. D. P. Lyra, and J. Valsiner (eds.), Dynamics and indeterminism in developmental and social processes, pp. 135–152. Hillsdale, NJ: Erlbaum. Piaget, J. (1935). La naissance de l’intelligence chez l’enfant. Neuchatel: Delacheaux et Niestl. Piaget, J . (1949). La psychologie de l’intelligence. Paris: Colin. Piaget, J. (1968). Le structuralisme. Paris: Presses Universitaires de France. Pickering, J. (1999). The self as a semiotic process. Journal of Consciousness Studies, 6, 31–47. Pipp, S. (1993). Infants’ knowledge of self, other, and relationship. In U. Neisser (ed.), The perceived self: ecological and interpersonal sources of self-knowledge, pp. 185–204. Cambridge: Cambridge University Press. Plato (1961). The collected dialogues, ed. E. Hamilton and H. Cairns. Princeton, NJ: Princeton University Press. (Original work written 390 BC) Polivy, J. (1998). The effects of behavioral inhibition: integrating internal cues, cognition, behavior, and affect. Psychological Inquiry, 9, 181–204. Port, R. F., and Van Gelder, T. (1995). Mind as motion: explorations in the dynamics of cognition. Cambridge, MA: Bradford Books/MIT Press. Prigogine, I., and Nicolis, G. (1977). Self-organization in non-equilibrium systems. New York: Wiley-Interscience. Prigogine, I., and Stengers, E. (1984). Order out of chaos: man’s new dialogue with nature. Toronto: Bantam Books. Prout, H. T., and Prout, S. M. (1996). Global self-concept and its relationship to stressful life conditions. In B. A. Bracken (ed.), Handbook of self-concept: developmental, social, and clinical considerations, pp. 259–286. New York:

References

251

Wiley. Provinelli, D. J., and Simon, B. B. (1998). Young children’s understanding of briefly versus extremely delayed images of self: emergences of the autobiographical stance. Developmental Psychology, 34, 188–194. Reddy, V. (1991). Playing with others’ expectations: teasing and mucking about in the first year. In A. Whiten (ed.), Natural theories of mind: evolution, development and simulation of everyday mindreading, pp. 143–158. Oxford: Basil Blackwell. Reed, E. S. (1987). Why do things look as they do? The implications of J. J. Gibson’s: The ecological approach to visual perception. In A. Costall and A. Still (eds.), Cognitive psychology in question, pp. 90–114. New York: St. Martin’s Press. Rhodewalt, F., and Morf, C. C. (1998). On self-aggrandizement and anger: a temporal analysis of narcissism and affective reactions to success and failure. Journal of Personality and Social Psychology, 74, 672–685. Ricoeur, P. (1983). Temps et recit, vol. 1, L’ordre philosophique. Paris: Seuil. Trans. K. McLaughlin and D. Pellauer, Time and narrative, vol. 1. Chicago: University of Chicago Press, 1984. Ricoeur, P. (1996). The hermeneutics of action. Thousand Oaks, CA: Sage. Robinson, M. D. (1998). Running from William James’ bear: a review of preattentive mechanisms and their contributions to emotional experience. Cognition and Emotion, 12, 667–696. Rochat, P. (1995). Early objectification of the self. In P. Rochat (ed.), The self in early infancy: theory and research, pp. 53–72. Amsterdam: Elsevier Science. Rochat, P., and Hespos, S. J. (1997). Differential rooting response by neonates: evidence for an early sense of self. Early Development and Parenting, 6, 105–112. Rogers, C. R. (1951). Client-centered therapy: its current practice, implications, and theory. Boston: Houghton Mifflin. Rogers, C. R. (1961). On becoming a person: a therapist’s view of psychotherapy. Boston: Houghton Mifflin. Rogers, T. B., Kuiper, N. A., and Kirker, W. S. (1977). Self-reference and the encoding of personal information. Journal of Personality and Social Psychology, 35, 677–688. Rosenberg, M. (1979). Conceiving the self. New York: Basic Books. Rosenberg, M. (1986). Self-concept from middle childhood through adolescence. In J. Suls and A. G. Greenwald (eds.), Psychological perspective on the self, vol. 3, pp. 107–135. Hillsdale, NJ: Erlbaum. Rozin, P., and Fallon, A. E. (1987). A perspective on disgust. Psychological Review, 94, 23–41. Rubin, D. C. (ed.) (1996). Remembering our past: studies in autobiographical memory. Cambridge: Cambridge University Press. Ruhland, R. (1998). Going the distance: a non-linear approach to change in language development. University of Groningen: unpublished doctoral dissertation. Rychlak, J. F. (1988). The psychology of rigorous humanism, 2nd edn. New York: New York University Press. Saarni, C. (1997). Emotional competence and self-regulation in childhood. In P. Salovey and D. J. Sluyter (eds.), Emotional development and emotional intelligence: educational implications, pp. 35–66. New York: Basic Books.

252

References

Saarni, C., Mumme, D. L., and Campos, J. J. (1998). Emotional development: action, communication, and understanding. In N. Eisenberg (vol. ed.; W. Damon, general ed.), Handbook of child psychology, (5th edn. vol. 3, Social, emotional and personality development, pp. 237–309. New York: Wiley. Salovey, P. (1992). Mood-induced self-focused attention. Journal of Personality and Social Psychology, 62, 699–707. Sameroff, A. J., and Fiese, B. H. (1990). Transactional regulation and early intervention. In S. J. Meissels and J. P. Shonkoff (eds.), Handbook of early childhood intervention, pp. 119–149. Cambridge: Cambridge University Press. Sampson, E. E. (1989). The deconstruction of the self. In J. Shotter and K. J. Gergen (eds.), Texts of identity, pp. 1–19. London: Sage. Sampson, E. E. (1993). Celebrating the other: a dialogic account of human nature. San Francisco: Westview Press. Sander, L. W. (1962). Issues in early mother–child interaction. Journal of the American Academy of Child Psychiatry, 1, 141–166. Sansone, C., and Berg, C. A. (1993). Adapting to the environment across the life span: different process or different inputs? International Journal of Behavioural Development, 16, 215–241. Sarbin, T. R. (1986a). Narrative psychology: the storied nature of human conduct. New York: Praeger. Sarbin, T. R. (1986b). The narrative as a root metaphor for psychology. In T. R. Sarbin (ed.), Narrative psychology: the storied nature of human conduct, pp. 3– 21. New York: Praeger. Sarbin, T. R. (1993). The narrative as the root metaphor for contextualism. In S. C. Hayes and L. J. Hayes (eds.), Varieties of scientific contextualism, pp. 51–65. Reno, NV: Context Press. Sartre, J. P. (1939). Esquisse d’une theórie des e´motions. Paris: Hermann. Trans. B. Frechtman, The emotions: sketch of a theory. New York: Philosophical Library, 1984. Savin-Williams, R. C., and Demo, D. H. (1983). Situational and transituational determinants of adolescent self-feelings. Journal of Personality and Social Psychology, 44, 824–833. Schank, R. C., and Abelson, R. P. (1995). Knowledge and memory: the real story. In R. S. Wyer (ed.), Advances in social cognition, vol. 8, pp. 1–87. Hillsdale, NJ: Erlbaum. Scheff, T. J. (1987). Creativity and repetition: a theory of the coarse emotions. In J. Rabow, G. Platt, and M. Goldman (eds.), Advances in psychoanalytic sociology, pp. 70–100. Malabor, FL: Robert E. Krieger. Scherer, K. R. (1984). Emotion as a multicomponent process: a model and some cultural data. In P. Shaver (ed.), Review of personality and social psychology, vol. 5, pp. 37–63. Berverly Hills: Sage. Scherer, K. R. (1999). Appraisal theory. In T. Dalgleish and M. Power (eds.), Handbook of cognition and emotion, pp. 637–663. London: Wiley. Scherer, K. R. (2000). Emotion as episodes of subsystem synchronization driven by nonlinear appraisal processes. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 70–99. New York: Cambridge University Press. Schlenker, B. R., and Weigold, M. F. (1989). Goals and the self-identification process: constructing desired identities. In L. A. Pervin (ed.), Goal concepts in personality and social psychology, pp. 243–290. Hillsdale, NJ: Erlbaum.

References

253

Schneider-Rosen, K., and Cicchetti, D. (1991). Early self-knowledge and emotional development: visual self-recognition and affective reactions to mirror self-images in maltreated and non-maltreated toddlers. Developmental Psychology, 27, 471–478. Schore, A. N. (1994). Affect regulation and the origin of the self: the neurobiology of emotional development. Hillsdale, NJ: Erlbaum. Schore, A. N. (1997). Early organization of the nonlinear right brain and development of a predisposition to psychiatric disorders. Development and Psychopathology, 9, 595–631. Schore, A. N. (2000). Contributions from the neurobiology of emotional development to models of self-organization. In M. D. Lewis and I. Granic (eds.), Emotion, development, and self-organization, pp. 155–185. New York: Cambridge University Press. Schwartz, G. E., and Russek, L. G. (1997). Do all dynamical systems have memory? Implications of the systemic memory hypothesis for science and society. In K. H. Pribram and J. S. King (eds.), Brain and values: behavioral neurodynamics, pp. 249–273. Hillsdale, NJ: Erlbaum. Schweder, R. A., Goodnow, J., Hatano, G., LeVine, R. A., Markus, H., and Miller, P. (1998). The cultural psychology of development: one mind, many mentalities. In R. M. Lerner (vol. ed.; W. Damon, general ed.), Handbook of child psychology, 5th edn. vol. 1, Theoretical models of human development, pp. 865–937. New York: Wiley. Sedikides, C., and Strube, M. J. (1997). Self-evaluation: to thine own self be good, to thine own self be sure, to thine own self be true, and to thine own self be better. In M. P. Zanna (ed.), Advances in experimental social psychology, vol. 29, pp. 209–269. New York: Academic Press. Selye, H. (1955). Stress and disease. Science, 122, 625–631. Selye, H. (1956). The stress of life. New York: McGraw-Hill. Shotter, J. (1981). Vico, moral worlds, accountability, and personhood. In P. Heelas and A. Lock (eds.), Indigenous psychologies: the anthropology of the self, pp. 266–284. New York: Academic Press. Shotter, J. (1989). Social accountability and the social construction of ‘‘you.’’ In J. Shotter and K. Gergen (eds.), Texts of identity, pp. 133–151. London: Sage. Skarda, C. A., and Freeman, W. J. (1987). How brains make chaos in order to make sense of the world. Behavioral and Brain Sciences, 10, 161–195. Smirnov, S. D. (1994). Intelligence and personality in the psychological theory of activity. In R. J. Sternberg and P. Ruzgis (eds.), Personality and intelligence, pp. 221–247. Cambridge University Press. Smith, L. B. (1995). Self-organizing processes in learning to learn words: development is not induction. In C. A. Nelson (eds.), New perspectives on learning and development, Minnesota Symposia on Child Psychology, vol. 28, pp. 1– 32. New York: Academic Press. Smith, L. B., and Thelen, E. (1993). A dynamic systems approach to development: applications. Cambridge, MA: Bradford Books/MIT Press. Sonnemans, J., and Frijda, N. H. (1995). The determinants of subjective emotional intensity. Cognition and Emotion, 9, 483–507. Spencer, S. (1971). Space, time, and structure in the modern novel. New York: New York University Press. Sroufe, L. A. (1979). Socio emotional development. In J. Osofsky (eds.), Handbook of infant development, pp. 462–518. New York: Wiley. Sroufe, L. A., and Waters, E. (1976). The ontogenesis of smiling and laughter: a

254

References

perspective on the organization of development in infancy. Psychological Review, 83, 173–189. Stapley, J., and Haviland, J. M. (1989). Beyond depression: gender differences in normal adolescents’ emotional experiences. Sex Roles, 20, 295–308. Stein, N. L. and Trabasso, T. (1992). The organization of emotional experience: creating links between emotion, thinking, and intentional action. Cognition and emotion, 6, 225–244. Stein, N. L., Trabasso, T., and Liwag, M. (1993). The representation and organization of emotional experience: unfolding the emotion episode. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 279–300. New York: Guilford Press. Stern, D. (1985). The interpersonal world of the infant. New York: Basic Books. Stern, D. (1993). The role of feelings for an interpersonal self. In U. Neisser (ed.), The perceived self, pp. 205–215. Cambridge: Cambridge University Press. Stewart, I. (1989). Does God play dice? Oxford: Basil Blackwell. Stipek, D. J., Recchia, S., and McClintic, S. (1992). Self-evaluation in young children. Monographs of the Society for Research in Child Development, 57 (serial no. 226). Chicago: University of Chicago Press. Stratton, P. (1988). Parents’ conceptualization of children as the organizer of culturally structured environments. In J. Valsiner (ed.), Child development within culturally structured enviroments, vol. 1, Parental cognition and adultchild interaction, pp. 5–29. Norwood, NJ: Ablex. Strauman, T. J. (1996). Stability within the self: a longitudinal study of the structural implications of self-discrepancy theory. Journal of Personality and Social Psychology, 71, 1142–1153. Svrakic, N. M., Svrakic, D. M., and Cloninger, C. R. (1996). A general quantitative theory of personality development: fundamentals of a self-organizing complex. Development and Psychopathology, 8, 247–272. Tangney, J. P., and Fischer, K. W. (eds.) (1995). Self-conscious emotions: the psychology of shame, guilt, embarrassment, and pride. New York: Guilford Press. Teasdale, J. D., and Barnard, P. J. (1993). Affect, cognition, and change: re-modelling depressive thought. Hillsdale, NJ: Erlbaum. Tesser, A. (1988). Toward a self-evaluation maintenance model of social behavior. In L. Berkowitz (ed.), Advances in experimental social psychology, Vol. 21, pp. 181–227. San Diego, CA: Academic Press. Thatcher, R. W. (1994). Psychopathology of early frontal lobe damage: dependence on cycles of development. Development and Psychopathology, 6, 565– 596. Thelen, E., and Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action. Cambridge, MA: Bradford/ MIT Press. Thelen, E., and Ulrich, B. D. (1991). Hidden skills: a dynamic systems analysis of treadmill stepping during the first year. Monographs of the Society for Research in Child Development, 56, 1 (serial no. 223). Chicago: University of Chicago Press. Tomkins, S. S. (1962). Affect, imagery, consciousness, vol. 1: The positive affects. New York: Springer. Tomkins, S. S. (1978). Script theory: differential magnification of affects. In H. E. Howe, Jr., and R. A. Dunstbier (eds.), Nebraska symposium on motivation, pp. 201–236. Lincoln: University of Nebraska Press.

References

255

Trevarthen, C. (1993). The self born in intersubjectivity: the psychology of an infant communicating. In U. Neisser (ed.), The perceived self: ecological and interpersonal sources of self-knowledge, pp. 121–173. Cambridge: Cambridge University Press. Trevarthen, C., and Hubley, P. (1978). Secondary intersubjectivity: confidence, confiding and acts of meaning in the first year. In A. Lock (ed.), Action, gesture and symbol: the emergence of language, pp. 183–227. New York: Academic Press. Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological Review, 96, 506–520. Tucker, D. M. (1992). Developing emotions and cortical networks. In M. R. Gunnar and C. Nelson (eds.), Developmental behavioral neuroscience, Minnesota symposia on child psychology, vol. 24, pp. 75–128. Hillsdale, NJ: Erlbaum. Van der Maas, H. L. J., and Molenaar, P. C. M. (1992). Stagewise cognitive development: an application of catastrophe theory. Psychological Review, 99, 395–417. Van der Meulen, M. (1987). Self-references in young children: content, metadimensions, and puzzlement. Groningen: Kinderstudies. Van Geert, P. L. C. (1989). Cognitive development: a process of organized self-organization. In G. J. Dalenoort (ed.), The paradigm of self, pp. 146–166. London: Gordon and Breach Science Publishers. Van Geert, P. L. C. (1991). A dynamic systems model of cognitive and language growth. Psychological Review, 98, 3–53. Van Geert, P. L. C. (1994). Dynamic systems of development: change between complexity and chaos. New York: Prentice Hall/Harvester Wheatsheaf. Van Geert, P. L. C. (in press). Time, change and intentional action in a dynamic framework: on the nature of psychological causes and measurement. In N. Granott and J. Parziale (eds.), Microdevelopmental clues: transition processes in development and learning, pp. 00–00. Mahwah, NJ: Erlbaum. Van Geert, P. L. C. (unpublished). Reply to Camras and Michel. Groningen: University of Groningen, Department of Psychology. Van Gelder, T., and Port, R. F. (1995). It’s about time: an overview of the dynamical approach to cognition. In R. F. Port and T. van Gelder (eds.), Mind as motion: explorations in the dynamics of cognition, pp. 1–43. Cambridge, MA: MIT Press. Van Halen, C. P. M. (in preparation). Experiencing self-definition problems over the life span. Groningen: University of Groningen, doctoral dissertation in preparation. Van Hooff, J. A. R. A. M. (1972). A structural analysis of the social behavior of a semi-captive group of chimpanzees. In M. Van Cranach and J. Vine (eds.), Social communication and movement, pp. 75–162. New York: Academic Press. Van der Werff, J. J. (1967). ‘‘Onbewuste’’ zelfbeoordeling: een experiment in de stijl van Werner Wolff (‘‘Unconscious’’ self-evaluation: an experiment after Werner Wolff). Nederlands Tijdschrift voor de Psychologie, 22, 150–161. Van der Werff, J. J. (1990). The problem of self-conceiving. In H. Bosma and S. Jackson (eds.), Coping and self-concept in adolescence, pp. 14–33. Berlin: Springer. Varela, F. J., Thompson, E., and Rosch, E. (1991). The embodied mind: cognitive science and human experience. Cambridge, MA: MIT Press. Vasil’eva, I. I. (1988). The importance of M. M. Bakhtin’s idea of dialogue and

256

References

dialogic relations for the psychology of communication. Soviet Psychology, 26, 17–31. Vedeler, D. (1991). Infant intentionality as object directedness: an alternative to representation. Journal for the Theory of Social Behavior, 21, 431–448. Vedeler, D. (1994). Infant intentionality as object directedness: a method for observation. Scandinavian Journal of Psychology, 35, 343–366. Verhofstadt-Dene`ve, L. (1985). Crises in adolescence and psycho-social development in young adulthood: a seven-year follow-up study from a dialectical viewpoint. In C. J. Brainerd and V. F. Reyna (eds.), Developmental psychology, pp. 509–522. Amsterdam, New York, Oxford: North Holland. Verhofstadt-Dene`ve, L. (1988). The phenomenal-dialectic personality model: a frame of reference for the psychodramatist. Journal of Group Psychotherapy, Psychodrama and Sociometry, 41, 3–20. Verhofstadt-Dene`ve, L. (1995). How to work with dreams in psychodrama: developmental therapy from an existential-dialectic viewpoint. International Journal of Group Psychotherapy, 45, 405–435. Verhofstadt-Dene`ve, L. (1997). Using conflict in a developmental therapeutic model. International Journal of Adolescent Medicine and Health, 9, 151–164. Verhofstadt-Dene`ve, L. (1999). Action- and drama-techniques with adolescent victims of violence: a developmental therapeutic model. International Journal of Adolescent Medicine and Health, 11, 351–367. Verhofstadt-Dene`ve, L. (2000). Theory and practice of action and drama techniques: developmental psychotherapy from an existential-dialectical viewpoint. London, Philadelphia, Sydney: Jessica Kingsley Publishers. Verhofstadt-Dene`ve, L. and Schittekatte, M. (1996). Adolescents have become adults: a 15-year follow-up. In Verhofstadt-Dene`ve, L., Kienhorst, I., and Braet, C. (eds.), Conflict and development in adolescence, pp. 47–60. Leiden: DSWO Press, University of Leiden. Voloshinov, V. N. (1929/1986). Marxism and the philosophy of language. Cambridge, MA: Harvard University Press. Vygotsky, L. S. (1979). Mind in society: the development of higher psychological processes, 2nd edn. Cambridge, MA: Harvard University Press. Walker, L. J. (1983). Sources of cognitive conflict for stage transition in moral development. Developmental Psychology, 19, 103–110. Watkins, M. (1986). Invisible guests: the development of imaginal dialogues. Hillsdale, NJ: Erlbaum. Wells, A. J. (1992). Variations in self-esteem in daily life: methodological and developmental issues. In R. P. Lipka and T. M. Brinthaupt (eds.), Selfperspectives across the life span, pp. 151–185. Albany, NY: State University of New York Press. Werner, H., and Wapner, S. (1953), Sensory-tonic field theory of perception. In J. S. Bruner and D. Krech (eds.), Perception and personality, pp. 83–107. Durham, NC: Duke University Press. White, G. M. (1993). Emotions inside out: the anthropology of affect. In M. Lewis and J. M. Haviland (eds.), Handbook of emotions, pp. 29–41. New York: Guilford Press. Whitehead, A. N. (1978). Process and reality, an essay in cosmology. New York: The Free Press. Wilber, K. (1979). No boundary: eastern and western approaches to personal growth. Boston: Shambala.

References

257

Wolff, W. (1943). The expression of personality. New York: Harper. Wylie, R. (1979). The self concept, vol. 2. Lincoln: University of Nebraska Press. Wylie, R. C. (1989). Measures of self-concept. Lincoln: University of Nebraska Press. Yasuo, Y. (1987). The body: toward an eastern mind-body theory. Albany: State University of New York Press. Zajonc, R. B. (1980). Thinking and feeling: preferences need no inferences. American Psychologist, 35, 151–175. Zender, M. A., and Zender, B. F. (1974). Vygotsky’s view about age periodization of child development. Human Development, 17, 24–40.

MMMM

Author index

Abelson, R. P., 118, 119 Abraham, F. D., 64, 185 Ackerman, B., 183 Allport, G. W., 17, 58 Alpert, R., 200, 201 Altman, I., 143 d’Ancona, H., 66 Anderson, D., 97 Angyal, A., 129 Anne–Albert, 164, 166 Antonovsky, A., 112 Archer, S. L., 1 Aron, A., 200 Aron, E., 200 Ashmore, R. D., 17, 21 Atkins, P. W., 78 Ausubel, D., 51 Baddeley, A. D., 118 Bailey, S. T., 25 Baird, B., 186 Bakan, D., 129 Bakhtin, M. M., 97, 104, 124, 125, 127, 138, 149, 193 Ball, L., 195 Baltes, P. B., 18, 23–26, 30–32 Bambrough, R., 91 Bandura, A., 56 Barclay, C. R., 97, 118, 119 Barclay, L. C., 27 Barnard, P. J., 185, 186 Barrett, K. C., 93 Bateson, G., 102 Baumeister, R. F., 2, 3, 10, 11, 14, 16, 17, 27, 28, 31, 53, 119 Beck, A. T., 58 Becker, E., 122 Beltrami, E., 64 Berg, C. A., 19, 21, 30 Blasi, A., 216, 227 Blass, E. M., 44 Block, J., 112, 177

Block, J. H., 112 Boden, J. M., 2, 14 Boesch, E. E., 97, 121 Borkowski, W., 221 Bosma, H. A., 3, 6, 97, 107, 109, 117, 169, 177, 178, 228 Boulifard, D., 8, 35, 164, 172, 209, 210 Bourdieu, P., 191 Bracken, B. A., 11, 13, 14, 26 Branco, A. U., 98 Brandtsta¨dter, J., 19 Brewer, W. F., 118, 119 Brinthaupt, T. M., 11, 24, 29 Brockmeier, J., 191, 197 Bronfenbrenner, U., 19, 20, 225 Brooks-Gunn, J., 12, 13, 20 Brown, B., 143 Bruner, J. S., 2, 109, 123, 140, 145, 190, 194, 197, 223 Buber, M., 98, 100 Buck, R., 183 Bullock, M., 12, 13 Burghess, D. N., 66 Butterworth, G., 97, 185 Bu¨tz, M. R., 112 Byrne, B. M., 5, 10–12, 26 Campos, J. J., 12, 93 Camras, L. A., 85, 92, 206, 213, 218 Capps, D., 175 Captain Picard, 65 Carello, C., 97 Carnochan, P., 93 Carstensen, L. L., 111 Carvalho, A. M. A., 98 Carver, C. S., 56 Cassirer, E., 126 Caughey, J. L., 126 Chandler, M. J., 191, 192, 195 Charles, S. T., 111 Chomsky, N., 92 Christianson, S., 119

259

260

Author index

Cicchetti, D., 12 Clausen, J. A., 20 Cloninger, C. R., 182, 229 Clore, G., 56 Cole, M., 110, 191 Collins, A., 56 Conville, R., 143 Cooley, C. H., 6, 10–12, 17, 18 Cooper, C. R., 194, 230 Costa, P. T. Jr., 199 Crouter, A. C., 19, 20 Csikszentmihalyi, M., 21, 31, 37, 98, 112, 222 Culver, C., 152, 163 Damasio, A. R., 184 Damon, W., 10, 14, 21, 24–26, 29, 121 Dannefer, D., 219, 224 Davidson, R. B., 4 Davitz, J. R., 42 Day, J. M., 104 De Boeck, P., 21, 154 De Levita, D. J., 3 De Quincey, C., 98 De Rivera, J., 93 De Waal, F. B. M., 55, 57 De Weerth, C., 85 Dedo, J. Y., 87 Deiner, E., 199 Demo, D. H., 3, 20, 22, 24, 26, 27, 30, 31 Demos, E., 103 Dennett, D. C., 46 Derryberry, D., 182 Descartes, R., 138 Dewey, J., 94, 98, 99 Dostoyevski, F. M., 124, 125, 149 Douglas, L., 181, 187, 228 Dunne, J., 109 Duran, E., 112 Duval, S., 37 Eckstein, S. G., 229 Edelstein, W., 20 Eisenberg, N., 11, 18 Ekman, P., 43, 54, 152, 163 Elder, G. H., 19 Elefthery, D. G., 146 Elkind, D., 197 Ellis, A., 158, 160 Ellis, Albert, 151, 157–161, 166, 169, 170 Emde, R. N., 109 Epstein, S., 10, 12, 15–18, 21, 29–31, 36, 59, 62, 205, 214, 217, 218, 221 Erikson, E. H., 3, 6, 107, 109, 110, 143, 174, 175, 202, 204, 216, 217, 219 Fallon, A. E., 45

Feiring, C., 20 Ferrari, M. D., 8, 35, 109, 191, 192, 194, 199, 200, 201, 211, 212, 214, 216, 218, 219, 221–223 Feyereisen, P., 162 Fiese, B. H., 22 Fine, S., 183 Fischer, A. H., 112, 116 Fischer, K. W., 14, 18, 21, 30, 35, 93, 196 Fitzgerald, H. E., 64 Fogel, A., 4, 8, 20–22, 35, 87, 94, 97–99, 102–104, 106, 111–113, 115, 117, 181–183, 185, 194, 196, 207, 208, 211, 212, 214, 215, 225–230 Frank, Anne, 151, 155–157, 159, 161, 167, 169, 170 Freeman, M., 191, 197 Freeman, W. J., 184, 186, 188 Freud, S., 6, 186 Fridlund, A. J., 43 Frijda, N. H., 7, 16, 17, 34, 35, 41–44, 49, 52–55, 58, 59, 62, 93, 96, 101, 103, 116, 163, 176, 183, 186, 203–208, 212, 214, 217 Ganguly, S. N., 97 Gavanski, I., 51 Gebelt, J. L., 4 Gecas, V., 20 Gergen, K. J., 2, 21, 29, 94, 109, 110 Gergen, M. M., 109, 110 Gibson, J. J., 97 Ginsburg, G. P., 102 Glodis, K., 216, 227 Goertzel, B., 185 Goffman, E., 102 Goldsmith, H. H., 199 Goleman, D., 11 Goodnow, J. J., 20 Goodwin, B. C., 179, 181 Gordon, C., 29 Graafsma, T. L.G., 3 Granic, I., 5, 87, 182 Grannemann, B. D., 27 Greenwald, A. G., 10–12, 15–17, 29 Grotevant, H. D., 3, 230 Harakal, T., 183 Harkins, D., 183 Harkness, K. L., 184, 188 Harre´, R., 11, 31, 116, 118, 139 Hart, D., 10, 11, 14, 20, 21, 24–26, 29, 121 Harter, S., 2, 3, 11, 14, 18, 20, 24, 26, 27, 31, 32, 196 Hatano, G., 20, Hattie, J., 1, 11–13, 16, 21, 22, 24–26, 30

Author index Haviland, J. M., 4, 8, 102, 109, 151–155, 172–174, 178, 190, 192, 197 Haviland-Jones, J.M., 8, 35, 158, 164, 201, 209–212, 214, 221, 226, 228, 229 Heatherton, T.F., 53 Held, R., 48 Hermans, H. J. M., 4, 8, 21–23, 30, 35, 96–99, 102, 107, 127, 128, 130, 131, 138, 141, 142, 145, 147, 149, 193–195, 208, 209, 211, 212, 214, 215, 218–220, 226, 228, 229 Hermans-Jansen, E., 8, 35, 127, 128, 130, 141, 142, 145, 147, 149, 208, 209, 211, 212, 214, 215, 218–220, 226, 228, 229 Hespos, S. J., 97 Heymans, P., 175, 176, 210, 228 Higgins, E. T., 4, 15, 18, 32 Hochschild, A., 116 Hofbauer, J., 66 Holland, J. H., 78 Hoorens, V., 12 Hopkins, B., 184 Horowitz, M. J., 48, 186 Hubley, P., 108 Huntley, C. W., 12 Hunziker, J., 192 Imperio-Hamburger, A., 98 Isaacowitz, D. M., 111 Isen, A. M., 112 Izard, C. E., 31, 44, 152, 182–184, 188 Jackson, E. A., 64, 65 James, W., 2, 3, 6, 7, 10–12, 16–18, 23, 50, 58, 99, 111, 120–122, 127, 138, 140, 141, 202 Jansz, J., 97, 112, 116, 208, 211, 219 Jaquish, G. A., 22, 31 Jaynes, J., 123 Johnson, F., 139 Johnson, M., 97 Johnson-Laird, P. N., 185 Jonas, R., 152 Jones, D., 102 Jopling, D., 98 Josephs, I. E., 97 Josselson, R., 137– 230 Juhasz, A., 20 Kagan, J., 12, 13 Kahlbaugh, P., 102, 109, 151, 153, 178, 190, 192, 197 Kalmar, D. A., 194, 197 Kaplan, H. B., 17 Karns, J., 229 Kauffman, S. A., 78, 179 Kegan, R., 98

261 Keith, L. K., 14, 26 Kelly, G. A., 36, 138, 143 Kelso, J. A. S., 179 Kempen, H. J. G., 21–23, 30, 96–99, 102, 107, 127, 193 Kendon, A., 102 Kernis, M. H., 27, 31 Kindermann, T. A., 19 Kingsland, S. E., 66 Kirker, W. S., 16 Kitayama, S., 20, 109, 200 Klages, L., 129 Klein Wassink, M. E., 224, 226, 229 Kohlberg, L., 143 Kramer, D. A., 154, 155 Kuiper, N. A., 16 Kuipers, P., 42 Kunda, Z., 31 Kunnen, E. S., 169, 178, 224, 226, 228, 229 Labouvie-Vief, G., 191 Lakoff, G., 97 Lalonde, C., 191 Lambrecht, L., 85, 92 Lamey, A. V., 181, 228 Lancelot, C., 4 Landauer, R., 181 Larsen, R. J., 199 Lazarus, R. S., 44, 45, 48, 93, 116 l’Ecuyer, R., 14, 25, 26 LeDoux, J. E., 162 Legerstee, M., 97 Leifer, M., 20 Lelwica, M., 152 Levinas, E., 94, 98 LeVine, R. A., 20 Levine, R. L., 64 Lewis, C. S., 193 Lewis, M., 12, 13, 20, 39 Lewis, M. D., 4, 5, 8, 35, 87, 103, 107, 109, 110, 112, 169, 181–183, 185, 187, 199–212, 214, 216, 218, 219, 221–223, 228 Lightfoot, C., 178 Lindenberger, U., 18 Lipka, R. P., 11, 24, 29 Liwag, M., 111 Locke, J., 177 Loewald, H. W., 100, 101 Logan, R. D., 10 Lotka, 76 Lucariello, J., 191 Lu¨scher, K., 19 Lu¨tkenhaus, P., 12, 13 Lyra, M., 112, 113, 196 Maclean, P., 55

262

Author index

Magai, C., 4, 8, 35, 151, 158, 159, 164, 169, 172, 183, 192, 197, 200, 201, 209, 210, 212, 223 Magnusson, D., 19, 22, 23, 30 Mahalingam, R., 191, 192, 194 Malatesta, C. Z., 112, 152, 183, 188, 199 Malatesta-Magai, C. 152, 163 Manion, M., 191 Marcia, J. E., 1, 3, 5, 89, 177 Markus, H. R., 1, 4, 10, 16, 20–22, 24, 29, 31, 51, 200 Marold, D. B., 14 Marsh, H. W., 11–13, 22, 24, 25, 30 Mascolo, M. F., 183, 196 Matteson, D. R., 1 Maturana, H. R., 180 McAdams, D. P., 129, 178, 191, 194, 195 McClelland, D. C., 200 McClintic, S., 12 McCrae, R. R., 199 McFadden, S. H., 4, 151, 159, 169, 201 Mead, G. H., 6, 10, 11, 18, 120, 125, 138, 147 Meltzoff, A. N., 50 Merleau-Ponty, M., 96, 97, 99, 103, 127 Mesquita, B., 163 Messinger, D., 87 Michaels, C. F., 97 Michel, G. F., 85, 92, 206, 213, 218 Miller, P., 20 Miller, P. J., 109 Miller, W. I., 45 Mischel, W., 178, 182, 185 Mistry, J., 109 Moen, P., 19 Molenaar, P. C. M., 64, 84, 182, 228, 229 Moore, M. K., 50 Moreno, J. L., 146 Moreno, Z. T., 146 Morf, C. C., 14, 28 Morris, W. N., 43 Morrow, J., 186 Mortimer, J. T., 20 Msihara, B. L., 229 Muir, F., 15 Mumme, D. L., 12 Mummendey, H. D., 20 Murray, J. D., 66 Murray, K., 175 Neisser, U., 22, 108 Nelson, G., 200 Nelson, K., 108 Nesselroade, J. R., 24 Newell, K. M., 64 Newman, L. S., 119 Nicolis, G., 143

Niedenthal, P. M., 51 Nolen-Hoeksema, S., 186 Novak, G., 70 Nowak, A., 221, 229 Nurius, P., 21 Nusbaum, B., 182, 183, 197, 201 Nuttin, J. M., 12 Nwokah, E., 87, 229 Oatley, K., 54, 185 Ogilvie, D. M., 17, 21 Ohman, A., 161 Oosterwegel, A., 1, 11, 12, 16, 22, 23, 25, 31, 37, 219 Oppenheim, D., 109 Orlofsky, J. L., 1 Ormel, J., 199 Ortony, A., 56 Orwell, L., 191 Overton, W. F., 138 Pacini, R., 15, 62, 218 Panksepp, J., 162, 187, 188 Pedrosa, M. I., 98 Perlmutter, M., 219, 224 Piaget, J., 48, 108, 125, 143 Pickering, J., 97, 100 Pipp, S., 20 Plato, 177, 193 Poincare´, 65 Polivy, J., 186 Port, R. F. 64, 182 Pratkanis, A. R., 10–12, 15–17, 29 Prigogine, I., 78, 143, 179 Prout, H. T., 20 Prout, S. M., 20 Provinelli, D. J., 108 Rathunde, K., 98, 112 Recchia, S., 12 Reddy, V., 108, 109 Reed, E. S., 97 Rhodewalt, F., 14, 28 Ricoeur, P., 102, 109 Robins, R. W., 177 Robinson, M. D., 17, 21 30, 31 Rochat, P., 97 Rogers, C. R., 37, 112 Rogers, T. B., 16 Rosch, E., 100 Rosenberg, M., 27, 120 Rosenberg, S., 21, 154 Rothbart, M. K., 182 Rozin, P., 45 Rubin, D. C., 118, 119 Ruetsch, C., 4 Ruhland, R., 229

Author index Russek, L. G., 198 Rychlak, J. F., 139 Saarni, C., 12, 13, 18, 21, 30 Safer, M. A., 119 Sagy, S., 112 Salovey, P., 35 Sameroff, A. J., 22 Sampson, E. E., 98, 138 Sander, L. W., 101 Sansone, C., 19, 21, 30 Sarbin, Th. R., 103, 123, 141, 193 Sartre, J.-P., 46 Savin-Williams, R. C., 20, 22, 27, 30, 31 Schaffer, A., 97 Schank, R. C., 118, 119 Schaufeli, W. B., 199 Scheff, T. J., 185 Scheier, M. F., 56 Scherer, K. R., 41, 45, 185 Schittekatte, M., 143 Schlenker, B. R., 177 Schneider-Rosen, K., 12 Schoff, K., 183 Schore, A. N., 182, 184, 185, 187, 188, 213 Schwartz, G. E., 198 Schweder, R. A., 20 Sedikides, C., 17 Selye, H., 49, 158 Sentis, K., 16, 31 Shah, A., 44 Shavelson, R. J., 11, 12 Shaver, P. R., 93 Shepard, B., 152 Shoda, Y., 177, 178, 182, 185 Shotter, J., 97, 118 Sigmund, K., 66 Simon, B. B., 108 Skarda, C. A., 186 Skinner, E. A., 19 Smart, L., 2, 14 Smith, L. B., 64, 98, 179, 181, 182, 184 Sokol, B., 191 Sonnemans, J., 54, 163 Spencer, S., 124 Sroufe, L. A., 56, 183 Stanger, C., 12 Stapley, J., 152 Stattin, H., 19, 22, 23, 30 Staudinger, U. M., 18 Stein, N. L., 54, 111, 185 Stengers, E., 78, 179 Stern, D., 94, 97, 99, 103, 104, 108, 112 Stewart, I., 65 Stillwell, A. M., 53 Stipek, D. J., 12, 14

263 Stratton, P., 22 Strauman, T. J., 25 Strube, M. J., 17 Sullivan, J., 92 Sullivan, M. W., 12 Svrakic, D. M., 182, 229 Svrakic, N. M., 182, 229 Tangney, J. P. 12, 14, 18, 21, 30, 51 Tappan, M. B., 104 Taska, L. S., 20 Tcherkassof, A., 43 Teasdale, J. D., 185, 186 Terschure, E., 42 Tesman, J., 152 Tesser, A., 2, 221 Thatcher, R. W., 85 Thelen, E., 64, 98, 179, 181, 182, 184, 229 Thompson, E., 100 Timmers, M., 117 Tomkins, S. S., 103, 161, 182, 183 Tong, B. R., 112 Trabasso, T., 54, 111, 185 Trevarthen, C., 20, 96, 108 Triandis, H. C., 20 Tucker, D. M., 184, 186, 188 Tudor, M., 200 Tykocinski, O., 15 Ulrich, B. D., 181, 182, 229 Vallacher, R. R. 221, 229 Van der Maas, H. L. J., 84, 182, 229 Van der Meulen, M., 6, 7, 26, 33–37, 203 Van der Werff, J. J., 6, 12 Van Geert, P. L. C., 4, 7, 64, 78, 85, 86, 89, 90–92, 181, 182, 185, 197, 205–207, 210, 212–214, 222, 223, 228, 229 Van Gelder, T., 64, 182 Van Goozen, J., 163 Van Halen, C. P. M., 6 Van Hooff, J. A. R. A. M, 42 Van Langenhove, L., 139 Van Loon, R. J. P., 127 Varela, F. J., 100, 180 Vasil’eva, I. I., 124 Vedeler, D., 96, 97 Verhofstadt-Dene`ve, L., 123, 143–146, 209, 220, 226, 229 Voloshinov, V. N., 137 Volterra, V., 66 Vookles, J., 15 Vygotsky, L. S., 125, 143 Walker, L. J., 143 Wapner, S., 48

264

Author index

Waschull, S. B., 27, 31 Waterman, A. S., 1 Waters, E., 56 Watkins, M., 125, 126, 149 Watson, J. B., 49 Weigold, M. F., 177 Weiss, M., 12 Wells, A. J., 21, 27, 31 Werner, C. M., 143 Werner, H., 48 White, G. M., 116 Whitehead, A. N., 97–99 Wicklund, R. A., 1, 37

Wilber, K., 100, 112 Wilson, A., 112, 183, 188 Winegar, L. T., 113 Wolff, W., 12 Wood, A. D., 66 Wurf, E., 1, 4, 10, 20–22, 24, 29 Wylie, R., 10, 26, 202 Yasuo, Y., 97 Zajonc, R. B., 44 Zender, B. F., 143 Zender, M. A., 143

Subject index

action readiness, 41–43 affect, discrete classes of activators, 161–162 Albert Ellis analysis of autobiography, 157–160 development of, 158–160 Anne Frank analysis of diary, 155–157 development of, 155–157 appraisal, 44, 58 levels of, 45 primary and secondary, 44 self as object of, 52–53 assessment of change over time, 227–228 of self and identity as self-organizing system, 226–230 of the system ‘‘at work,’’ 226–227 attractors, 82–83 in identity development, 220–221 authorship, 102–107 a definition of, 95 in parent–infant communication, 104 autobiographical memories, 118–119 becoming and being, 99–102 bifurcations, 84 Cardinal Sins, 174–175 Cartesian split, 138 case study research versus averaging across cases, 154, 173–174 change-based emotions, 162, 164 change emotions, 111–112 chaos, 83 cognition–emotion interactions, 182–190 cognition–emotion recursions, 182–190 cognitions, implicit versus explicit, 62 Cognitive Experiential Self-Theory (CEST), 58–63 experiential and rational system, 59, 60

cognitive generativity, 219 color model of emotion, 162 competence, 56 concerns, 54–56, 176 and self, 55 conflict between I-positions, 144 in development, 169 in personality development, 142–143 conscious versus non-conscious processing, 62 consistency and blocked goals, 188–189, 194–195 context as system, 224–225 as ordered effect, 224 as random influence, 224 in a dynamic systems approach, 223–224 coupling of elements, 179 development, a dynamic systems perspective, 181–182 developmental processes levels of analysis, 169–170 time scales, 173 diachronous perspective, 227 dialectical developmental psychology, 143 dialogical relationships, 124–125 dialogues, 125 imaginal, 125 dynamic systems a definition of, 64 a mathematical non-linear model, 72 a qualitative approach, 86–88 a quantitative approach, 85–86 and environment, 68 and geometric representations, 67–68 and self-organization, 78–81 as structure of interactions, 69 change across time, 81–82

265

266

Subject index

dynamic systems (cont.) mathematical analysis versus modeling, 75–78 dynamic systems model of emotional processes, 163–168 emergence of continuity, 179–181 emotional dissonance, 117 emotional experience, 45–47, 93 and sense of self, 45–47 as non-linear process, 153 as iterative process, 153 in adolescence, 152–153 emotional expression in mother–child interactions, 152 emotional interpretation (EI), 220–221 neurobiology of, 186–187 real time, 184–186 recurrent, 184–186 emotional processes computer models of, 163–168 microlevel versus semantic meaning-making level, 171 emotional traits, 153–154 emotions a definition of, 41 a functional perspective, 41–43 and awareness of self, 40, 45–47 and cognitions, 60 and competence, 56 and concerns, 54–56 and facial expressions, 92 and self-representations, 43–44 and subjectivity, 40 and the self as subject, 47–51 appraisal of (significance), 53 as microprocesses, 161–162 as organizer of identity, 155, 160–161 as subject–object relationship, 41 conceptual confusions, 39 duration of, 163–164 modeling of individual differences, 166–168 self as object of, 51–53 standards of evaluation, 116 the evaluation of, 116 emotions narratives in western culture, 112 events, a definition of, 98 explicit identity, 190 a definition of, 218 consistency of, 192–195 feedback loops, 179 frames definition of, 102

nonverbal, 104 Gedanken experiment, 164 habituation, 219 identity a definition of, 177 and autobiographical consistency, 118–119 as autobiography, 191 as beginning in infancy, 108 as cognitive structure, 177 as dialogue, 193 as embodied being, 117 implicit and explicit, 178 stability and change, 177–178 identity change, 195–197, 220 self-organizing processes, 196–197 time scales, 196–197 identity crisis, 175 identity development, as dialectical process, 145–149 implicit identity, a definition of, 217 I-positions, 127, 132–133, 137–139 level-based emotions, 162, 164 logistic map, 83 Lotka–Volterra equations, 76 methods of analysis, 175–176 moods, 188–189 multivoicedness of the self, 126 narrative, 102–103, 123, 191–192 narrative emotions, 103–111 narrative identity, a definition of, 109 O-motive (contact and union with others), 129, 132–133 orientation, as most basic form of self, 96 orientational emotions, 111 personal identity, a definition of, 178 personality, a definition of, 178 phase shift, 170 phase transitions, 179, 220 Phenomenological–Dialectical Personality Model, 144–145 polyphonic novel, 124 principles of self-organization recursion, coupling and complementarities, 181 processes, 218–223 bottom-up and top-down, 221–222

Subject index in implicit and explicit identity, 218–223 verbal psychological models, 218–223 formal mathematical models, 218–223 psychodrama, 145 psychological theories as verbal narratives, 91 as formal mathematical models, 91 reference emotions, 108–109, 111 relational pespective, 94–95 roles, development of, 154 self a definition of, 34 and emotions, 60–61, 139–140 as object in emotions, 51–53 as self-reflecting agency, 121 as set of concerns, 55 as subject in emotions, 47, 51 cultural construction of, 174–175 development in infancy and childhood, 108 executive, 58–60 meta-affective knowledge, 170 multivoiced, 127 stability and change, 61–62 terminological issues, 31, 34 self and emotion, a relational perspective, 94–95 self and identity and self-organization, 4–5 as cognitive structures, 1–2 as multilayered system, 214–216 components and emerging phenomena, 90, 214–216 definition as a system, 89–90 from a dynamic systems perspective, 206 fundamental questions, 3, 59–63 history of thinking about, 3 implications of a dynamic systems approach, 213 new perspectives, 4 terminological issues, 6 the issue of quantification, 91 self-as-knower, 121 self-as-known, 121 self-awareness, the spectrum of, 100–101 self-concept a definition of, 11 history of the concept, 10–11 measurement methods, 30–31

267 short-term variability, 25–26 life-span developmental change, 24–25 stability and variability, 23–29 terminological issues, 31 the dynamic interaction with context, 22–23 the role of context, 18–23, 35–36 the role of emotions, 11–18, 35 self-confrontation method, 130–131 self-esteem a definition of, 11 stability and fluctuations, 26–28 self-experience, and time scales, 95, 110–112 self-evaluation and emotions, 2 self-investigation, 131, 136–137 self-organization, 5, 78–81, 178–181 a definition of, 79 and continuity, 178–181 self-organization of personality the principle of accrual and crystallization, 187–189 self-perpetuation, 178–181 self-related concepts, a framework, 36–38 sense of identity, 107 a definition of, 216–217 S-motive (self-enhancement), 129, 132–133 stability, differences between implicit and explicit identity, 199–201 systems, lower-order elements and higher-order forms, 179 time scales, 215, 219–220 and levels of organization, 215 in emotional self-organization, 184–187 in the modeling of developmental processes, 173 time series analysis, 175–176 transcendental self, 47 empirical contents, 47–48 uniqueness of self, 97 valuation theory, 127–129 generalization (G), 129 idealization (I), 129 valuations, 128 organization and reorganization, 130–131 working context, 21 working self-concept, 20–21

Studies in Emotion and Social Interaction First Series Editors: Paul Ekman and Klaus R. Scherer Handbook of methods in noverbal behavioral research Edited by Klaus R. Scherer and Paul Ekman Structures of social action: studies in conversational analysis Edited by Max Atkinson and John Heritage Interaction structure and strategy Starkey Duncan, Jr., and Donald W. Fiske Body movement and speech in medical interaction Christian Heath The Emotions Nico Frijda Conversations of friends: speculations on affective development Edited by John M. Gottman and Jeffrey G. Parker Judgment studies: design, analysis, and meta-analysis Robert Rosenthal The individual, communication, and society: essays in memory of Gregory Bateson Edited by Robert W. Rieber Language and the politics of emotion Edited by Catherine Lutz and Lila Abu-Lughod Fundamentals in nonverbal behavior Edited by Robert Feldman and Bernard Rime´ Gestures and speech Pierre J. M. Feyereisen and Jacques-Dominique de Lannoy Landscapes of emotion: mapping three cultures of emotion in Indonesia Karl G. Heider Contexts of accommodation: developments in applied sociolinguistics Howard Giles, Justine Coupland, and Nikolas Coupland Best laid schemes: the psychology of emotions Keith Oatley Interpersonal expectations: theory, research, and applications Edited by Peter David Blanck Emotional contagion Elaine Hatfield, John T. Cacioppo, and Richard L. Rapson Exploring affect: the selected writings of Silvan S. Tomkins Edited by E. Virginia Demos

Identity and Emotion: Development Through Self-Organization (Studies in Emotion and Social Interaction)

Structures of Social Action (Studies in Emotion and Social Interaction)

Emotions and Multilingualism (Studies in Emotion and Social Interaction)

Metaphor and Emotion: Language, Culture, and Body in Human Feeling (Studies in Emotion and Social Interaction)

Gender and Emotion: Social Psychological Perspectives (Studies in Emotion and Social Interaction)

Emotions and Multilingualism (Studies in Emotion and Social Interaction)

Whole Analysis (Studies in Emotion and Social Interaction)

Social Structure and Emotion

The Hidden Genius of Emotion: Lifespan Transformations of Personality (Studies in Emotion and Social Interaction)

Emotion in Memory and Development: Biological, Cognitive, and Social Considerations

Emotion, Social Relationships, and Health

Judgment Studies: Design, Analysis, and Meta-Analysis (Studies in Emotion and Social Interaction)

Judgment Studies: Design, Analysis, and Meta-Analysis (Studies in Emotion and Social Interaction)

Emotion

Emotions across Languages and Cultures: Diversity and Universals (Studies in Emotion and Social Interaction)

Group Dynamics and Emotional Expression (Studies in Emotion and Social Interaction)

Action, Emotion and Will

Emotion and Adaptation

Emotion, Place and Culture

Emotion and Consciousness

Emotion (Cambridge Studies in Philosophy)

Causes and Consequences of Feelings (Studies in Emotion and Social Interaction)

The Self and Its Emotions (Studies in Emotion and Social Interaction)

Interpersonal Expectations: Theory, Research and Applications (Studies in Emotion and Social Interaction)

Communicating Emotion: Social, Moral, and Cultural Processes

Fact and Value in Emotion

Emotion and the Arts

Emotion and Adaptation

Social Justice and the Experience of Emotion

Identity and Emotion: Development Through Self-Organization (Studies in Emotion and Social Interaction)

Structures of Social Action (Studies in Emotion and Social Interaction)

Emotions and Multilingualism (Studies in Emotion and Social Interaction)

Metaphor and Emotion: Language, Culture, and Body in Human Feeling (Studies in Emotion and Social Interaction)

Gender and Emotion: Social Psychological Perspectives (Studies in Emotion and Social Interaction)

Emotions and Multilingualism (Studies in Emotion and Social Interaction)

Whole Analysis (Studies in Emotion and Social Interaction)

Social Structure and Emotion

The Hidden Genius of Emotion: Lifespan Transformations of Personality (Studies in Emotion and Social Interaction)

Emotion in Memory and Development: Biological, Cognitive, and Social Considerations

Emotion, Social Relationships, and Health

Judgment Studies: Design, Analysis, and Meta-Analysis (Studies in Emotion and Social Interaction)

Judgment Studies: Design, Analysis, and Meta-Analysis (Studies in Emotion and Social Interaction)

Emotion

Emotions across Languages and Cultures: Diversity and Universals (Studies in Emotion and Social Interaction)

Group Dynamics and Emotional Expression (Studies in Emotion and Social Interaction)

Action, Emotion and Will

Emotion and Adaptation

Emotion, Place and Culture

Emotion and Consciousness

Emotion (Cambridge Studies in Philosophy)

Causes and Consequences of Feelings (Studies in Emotion and Social Interaction)

The Self and Its Emotions (Studies in Emotion and Social Interaction)

Interpersonal Expectations: Theory, Research and Applications (Studies in Emotion and Social Interaction)

Communicating Emotion: Social, Moral, and Cultural Processes

Fact and Value in Emotion

Emotion and the Arts

Emotion and Adaptation

Social Justice and the Experience of Emotion

Recommend Documents