Self, Ego, and Identity

Self, Ego, and Identity
Daniel K. Lapsley F. Clark Power

Editors
Self, Ego, and Identity

Integrative Approaches
Springer-Verlag
New York Berlin Heidelberg
London Paris Tokyo
Daniel K. Lapsley
Department of Psychology
University of Notre Dame
Notre Dame, Indiana 46556, USA
F. Clark Power
Program of Liberal Studies
University of Notre Dame
Notre Dame, Indiana 46556, USA
Library of Congress Cataloging-in-Publication Data

Self, ego, and identity.
Bibliography: p.
Includes index.
1. Self. 2. Ego (Psychology) 3. Identity
(Psychology) 4. Psychology, Pathological.
I. Lapsley, Daniel K. II. Power, F. Clark.
RC455.4.S42S44 1988 155.2 87-28480
© 1988 by Springer-Verlag New York Inc.

Softcover reprint of the hardcover 1st edition 1988
All rights reserved. This work may not be translated or copied in whole or in part without the
written permission of the publisher. (Springer-Verlag, 175 Fifth Avenue, New York, New
York 10010, USA), except for brief excerpts in connection with reviews or scholarly analysis.
Use in connection with any form of information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter devel-
oped is forbidden.
The use of general descriptive names, trade names, trademarks, etc. in this publication, even
if the former are not especially identified, is not to be taken as a sign that such names, as
understood by the Trade Marks and Merchandise Marks Act, may accordingly be used freely
by anyone.
Typeset by Asco Trade Typesetting Ltd., Hong Kong.
ISBN 978-1-4615-7836-9 ISBN 978-1-4615-7834-5 (eBook)

DOl 10.1007/978-1-4615-7834-5
Preface
In the midst of the "cognitive revolution," there has been a veritable ex-
plosion of interest in topics that have been long banished from academic
consideration under the intellectual hegemony of behaviorism. Most
notably, notions of self, ego, and identity are reasserting themselves as
fundamental problems in a variety of research traditions within psychol-
ogy and the social sciences. Theoretical models, review articles, edited vol-
umes, and empirical work devoted to these constructs are proliferating at
a dizzying rate. This clearly attests to the renascent interest in these
topics, the vitality of these research paradigms, and the promise that these
constructs hold for explaining fundamental aspects of human development
and behavior.
Although the renewed academic interest in self, ego, and identity is
obviously an exciting and healthy development, there is always the tenden-
cy for research to take on a parochial character. When boundaries are
erected among different theoretical perspectives, when empirical findings
are viewed in isolation, when theories are too sharply delimited and segre-
gated from other domains of behavior, then what may seem like progres-
sive, healthy, and content-increasing tendencies in a research paradigm
may turn out to be, on closer inspection, merely an inchoate thrashing
about. Fortunately there is an internal dynamic to scientific investigation
that tends to combat this degenerating tendency. There is something about
the rhythm of science that bids us to transcend parochial theoretical in-
terests and seek the most general theory. We believe that we are at such a
juncture in the study of self, ego, and identity. A number of scholars are
now either proposing novel integrative frameworks or are at least consider-
ing the integrative potentials of their work. The purpose of this edited
volume is to contribute to the integrative push that is now evident in the
field by bringing together in one volume those writers whose work holds
great promise for further integrative study of self, ego, and identity.
The book is divided into three sections, one devoted each to self, ego,
and identity. This division is only an organizational device, and does not
represent any rigid commitment to a particular demarcation of the field or
vi Preface
definitional preference. This is in keeping with the integrative spirit of the

volume. Indeed, some of our authors use the terms self and ego inter-
changeably, and others attempt to account for self and identity within a
common framework.
The first section is devoted to integrative approaches to the self. In the
lead-off chapter, Gil Noam articulates a clinical-developmental theory of
"biography and transformation." This powerful lifespan theory attempts
to reconcile neo-Piagetian theories of social cognition with neo-
psychoanalytic accounts of object relations in order to effect a comprehen-
sive model of the self. The aim of this project is to account for patterns of
psychopathology in light of structural integration and cognitive transforma-
tion. One intriguing conclusion that is drawn is that "development" can
have untoward consequences for healthy personality functioning. In other
words, mental health is not necessarily advanced by structural develop-
ment. To explicate his theory, Noam develops several richly suggestive
concepts, such as vertical and horizontal encapsulation, overassimilation,
internal perspective, and core themes. In the confrontation between
Piagetian and psychoanalytic models, Noam forges a new understanding of
the relation between normal and impaired development, and of equilibra-
tion, integration, and the rhythm of stage development. Noam illustrates
the workings of his theory with reference to the life of Franz Kafka, a
strategy that recalls Erikson's case study analysis of Gandhi and Martin
Luther. The theory of biography and transformation is an innovative and
important step towards the merging of cognitive development and
psychoanalysis that was once envisioned by Piaget.
Anthony Greenwald has made numerous important contributions to the
study of the self in adult information-processing. In Chapter 2, he takes on
quite a different assignment by reflecting on the ontogenesis of the
"organization-of-knowledge" conception of the self. That is, given the con-
ception of the adult self that devolves from the cognitive and social psycho-
logical literatures, Greenwald attempts to describe the developmental
trajectory, the intermediate levels, that would bring the developing person
to a mature self. Hence, the self is viewed as a knowledge structure that
can be described in terms of ordered levels of representational complexity,
such as features, objects, categories, propositions, and schemata. Accord-
ing to Greenwald, the self develops through these levels of representation,
which are further explicated with reference to the cognitive developmental
theories of Bruner and Piaget. Greenwald also discusses variations in the
adult's sense of self-worth. These variations are seen to reflect different
types of ego involvement which, in turn, reflects diffuse, public, private,
and collective facets of the self. There is speculation that these facets may
also have developmental features. This integrative framework would seem
to suggest productive new lines of research, and it underscores the impor-
tance of rapprochement between developmental and social psychological
theories of the self.
The topic of self-worth is taken up again by Susan Harter in Chapter 3.
Preface vii
Harter argues that a comprehensive model of self-worth must incorporate

the insights of James and Cooley. In a manner akin to the previous chap-
ter, Harter effectively pursues the developmental underpinnings ofthe ma-
ture self. This yields an excellent review of how the insights of James and
Cooley can be translated into developmental research strategies on how
the ability to conserve the self can have an impact on self-worth, on how
self-worth can be protected and enhanced (see also Greenwald's chapter),
and on whether self-enhancement is "worth the energy." Harter concludes
her chapter by considering Eastern perspectives on the self and the in-
teresting possibility of describing individuals as I-people and Me-people.
Numerous integrative possibilities are encountered along the way. Not
only is there a fresh developmental look at the theories of James and
Cooley, and an attempt to transcend Western biases concerning the self,
there are also developmental insights regarding beneffectance, the identity
process in adolescence, attribution theory, and certain individual differ-
ence constructs (self-monitoring).
Chapter 4 by Daniel Hart further extends the Jamesian approach to the
self developed in the previous chapter, and anticipates the theme in the
next chapter by Ruthellen Josselson regarding the "embedded self." The
focus of Hart's chapter is on the "social self" aspect of James' theory. He
defines the social self in terms of a general model of self-understanding that
includes the self as subject and as object, charted along four developmental
levels of organization. Hart convincingly argues that self-understanding
varies by social context, that the self-concept should be considered in light
of a network of relationships, and that the self-concept exists not as a
monolith but rather as a "confederacy" of social selves. After reviewing
the contextual themes in self-concept research, Hart demonstrates the
power of this perspective with an empirical demonstration that includes a
novel methodological approach. Not only do Hart's findings provide sup-
port for his general model of self-understanding, they also have intriguing
implications as well for our understanding of the psychological separation
process in adolescence and for adolescent peer relations.
In the final chapter of this section, Ruthellen Josselson presses the theme
of Dan Hart's chapter. Josselson eloquently argues that we must re-think
our conception of what mature selfhood entails. In her view, adolescent
and female development have vitiated the traditional view that selfhood is
achieved through autonomy, separation, and independence. With particu-
lar reference to separation-individuation theory, Josselson points out that
separation-individuation and connectedness are two sides of the same ma-
trix and that, as yet, we do not have an adequate account of human bonded-
ness that recognizes this fact. However, Josselson does seek out the possi-
bilities for such an account in a variety of theoretical outlooks, including
object relations and attachment theory, Kegan's conception of the evolving
self, Kohut's account of selfobjects, and existential writings. Josselson
makes a strong plea to redefine identity in terms that make it possible to
conceive of the mature self as one intertwined with others, noting that: "A
viii Preface
theory of the self must be interwoven with a theory of relationship." This

chapter takes us a considerable distance toward that end.
The second section of the volume includes four chapters on the "ego."
In Chapter 6, Dan Lapsley and Kenneth Rice attempt to phrase an integra-
tive model of adolescent ego development. They argue that the study of
recapitulated patterns of development has considerable heuristic value for
understanding ego development in adolescence. To demonstrate the power
of this heuristic, they first review three recapitulation themes involv-
ing separation-individuation, the emergence of self-understanding, and
the transmutation of narcissism, and then show how these processes are
anchored in adolescence by the imaginary audience and personal fable
constructs. This framework also attempts to account for impaired ego
development, and sheds new light on individual differences constructs.
In Chapter 7, Clark Power, Ann Power, and John Snarey address the
question of ego integrity in the elderly. Ego integrity is a long-neglected
research topic, and this empirical chapter represents an original and impor-
tant advance in our attempts to come to grips with it. The authors define
ego integrity in terms of four concepts: detachment, acceptance of limits,
holistic awareness, and aretaic reflection. After first distinguishing between
structural stage and age-bound functional phase accounts of development,
they conclude that ego integrity is adequately captured by the four con-
cepts, and that integrity must be described in terms of both stage and phase.
Integrative stage models, then, appear to be necessary to capture the com-
plexity of ego functioning in the elderly.
In Chapter 8, Loren Lee and John Snarey take on a daunting task. By
means of an intensive secondary analysis of nine data sets, they attempt to
delineate the relation between Loevinger's ego development construct and
Kohlberg's moral development sequence. After a useful comparision of
the claims made by the two theories and an exploration of their rela-
tionship, the authors empirically assess the nature of the correspondence
and the influence of moderator variables. The results are complex and illu-
minating. Although no strong and consistent claims can be made regarding
the question of ego-moral primacy, the authors are able to describe a mod-
al developmental trajectory regarding the ego-moral relationship. This
modal trajectory is then interpreted within the context of Erikson's life-
span account of psychosocial development. This strong chapter represents
a significant advance in our understanding of the relation between moral
and ego development, and points the way for conceptualizing these con-
structs in terms of psychosocial development.
John Broughton and Marta Zahayekevich also discuss Loevinger's
theory of ego development in Chapter 9. The authors critically examine
every facet of the ego paradigm, including the nature of the ego concept
itself, the coherence of theory and assessment, and metatheoretical issues
revolving around assumptions regarding self, mind, and society. Brought-
on and Zahaykevich are particularly interested in uncovering the ideologi-
Preface ix
cal biases immanent in the theory. As such, the authors intend this analysis
to be a case study assessment of trends that afflict the general study of ego
development and of the possibility of constructing meaningful explanations
of personality development.
Integrative approaches to identity are considered in the final four chap-
ters. In Chapter 10, James Marcia examines the relation between ego
identity status and two major theoretical streams in developmental
psychology, object relational accounts of individuation, and the cognitive-
moral development tradition. Marcia argues not only that there are empir-
ical relations among these constructs, but that, in addition, they also share
common developmental goals and processes and have similar child-rearing
implications. Marcia concludes by suggesting that ego identity might serve
as an integrative concept when viewing psychosocial development, cogni-
tive and moral development, and individuation. This suggestion recalls the
conclusions reached by Lee and Snarey in Chapter 8, who argued that
Erikson's lifespan psychosocial theory has sufficient integrative power to
account for the evolving ego-moral relation across the lifespan. Certainly
these two chapters should provide a stimulating impetus for continued
work along these lines.
In Chapter 11, Gus Blasi provides a trenchant critique of contemporary
assessment strategies for measuring Erikson's ego identity construct. He
argues that these approaches distort and trivialize the concepts of crisis and
commitment, and fragment the unity of the self among various identity
domains. To recover the unity of the self and to reestablish its role in iden-
tity, Blasi articulates a notion of the self as subject. After describing four
aspects that characterize the subjectivity of experience, Blasi then shows
how the subjective self is related to identity and how patterns of identity
are reflected in certain of Loevinger's stages of ego development. In this
remarkable chapter, Blasi extends his view on the responsible and subjec-
tive self, and points the way clearly for an adequate conception of the role
of the self in psychological functioning.
Mike Berzonsky, in Chapter 12, presents the most thorough integration
of the adult social cognition and self and identity literatures yet available.
Berzonsky proposes a process view of identity formation, with particular
focus on how self-relevant information is encoded, elaborated, and struc-
tured. Identity is viewed as a self-theory, and this model is creatively expli-
cated with reference to recent developments in the philosophy of science.
He argues, for example, that the self-theory serves the same functions and
is composed of the same elements as a scientific theory. Berzonsky also
shows how cognitive schemas, scripted behavioral strategies, and self-
presentations relate to the self-theory and the identity statuses. Berzonsky
concludes by describing a unified view of identity that includes process,
content, and structure. This chapter is theoretically informed and empirical-
ly driven, and should provoke a plethora of integrative studies on identity.
The final chapter, by Andrew Weigert, presents a sociological perspec-
x Preface
tive on self-understanding that is motivated by "pragmatic social construc-

tivism," the view that personal and social meanings are produced and sus-
tained in the process of symbolic interaction. In this chapter, Weigert links
self and identity with concepts suggested by the sociology of religion, such
as authenticity, ambivalence, and eschatology. According to Weigert, the
social form and content of the self and identity are derived socio-historical
productions and, in the modern era, the form and content can be described
in terms of crisis. Crisis is seen to take the form of ambivalence and exis-
tential anxiety, and as a search for authenticity. Weigert then examines
existential and eschatological responses to the modern crisis. This is an
appropriate chapter with which to close the volume, inasmuch as Weigert
picks up on numerous themes that have resonated throughout many of the
previous chapters. For example, his focus on the socio-historical character
of self and identity recalls a similar theme evident in the chapters by Harter
and by Broughton and Zahayekevich. His definition of identity as a
typified self situated in a network of social relationships would find broad
agreement by Hart and Josselson. His account of the subjective self and
authenticity reveals similarities and differences with the account given by
Blasi. Weigert reminds us that the challenge facing modern selfhood has
deep sociological and historical connections, and that the goal of identity
formation is to transcend ambivalence so as to creatively construct social
reality and courses of moral action in community.
What is unique about this book is its integrative thrust and the willing-
ness of our authors to confront the challenge of seeking points of contact
with other literatures. It is our hope that these chapters will contribute to
increasing dialogue among researchers from different theoretical perspec-
tives and thereby encourage innovative and integrative lines of research.
We hope that this volume will have a galvanizing effect to this end.
Daniel K. Lapsley
F. Clark Power
Contents
Preface.......................................................... v
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Part I: Self
1. The Self, Adult Development, and the Theory of Biography

and Transformation
Gil G. Noam ............................................... 3
2. A Social-Cognitive Account of the Self's Development

Anthony G. Greenwald..................................... 30
3. The Construction and Conservation of the Self: James and

Cooley Revisited
Susan Harter ............................................... 43
4. The Adolescent Self-Concept in Social Context

Daniel Hart ................................................ 71
5. The Embedded Self: I and Thou Revisited

Ruthellen fosse/son ......................................... 91
Part II: Ego
6. The "New Look" at the Imaginary Audience

and Personal Fable: Toward a General Model of Adolescent
Ego Development
Daniel K. Lapsley and Kenneth Rice . . . . . . . . . . . . . . . . . . . . . . . . . 109
xii Contents
7. Integrity and Aging: Ethical, Religious, and Psychosocial

Perspectives
F. Clark Power, Ann R. Power, and John Snarey . . . . . . . . . . . . . 130
8. The Relationship Between Ego and Moral Development: A

Theoretical Review and Empirical Analysis
Loren Lee and John Snarey ................................. 151
9. Ego and Ideology: A Critical Review of Loevinger's Theory

John M. Broughton and Marta K. Zahaykevich .............. 179
Part III: Identity
10. Common Processes Underlying Ego Identity,

Cognitive/Moral Development, and Individuation
James E. Marcia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
11. Identity and the Development ofthe Self

Augusto Blasi .............................................. 226
12. Self-Theorists, Identity Status, and Social Cognition

Michael D. Berzonsky ...................................... 243
13. To Be or Not: Self and Authenticity, Identity, and

Ambivalence
AndrewJ. Weigert.......................................... 263
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Subject Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

Contributors
Michael D. Berzonsky, Department of Psychology, State University of

New York at Cortland, Cortland, NY 13045
Augusto Blasi, Department of Psychology, University of Massachusetts-

Boston, Boston, MA 02125
John M. Broughton, Box 33, Teacher's College, Columbia University,

New York, NY 10027
Anthony G. Greenwald, Department of Psychology, NI-25, University of

Washington, Seattle, WA 98195
Daniel Hart, Department of Psychology, Rutgers University, Camden, NJ

08102
Susan Harter, Department of Psychology, University of Denver, Denver,

CO 80208
Ruthellen Josselson, Department of Psychology, Towson State University,

Baltimore, MD 21204
Daniel K. Lapsley, Department of Psychology, University of Notre Dame,

Notre Dame, IN 46556
Loren Lee, School of Education, Harvard University, Appian Way, Cam-

bridge, MA 02138
James E. Marcia, Department of Psychology, Simon Fraser University,

Burnaby, B.C., Canada V5A 1S6
Gil G. Noam, Harvard Medical School, McLean Hospital, 115 Mill St.,
Belmont, MA 02178
xiv Contributors
Ann R. Power, South Bend Community School Corporation, 635 South

Main Street, South Bend, IN 46601
F. Clark Power, Program of Liberal Studies, 343 Decio Hall, University of

Notre Dame, Notre Dame, IN 46556
Kenneth Rice, Department of Psychology, University of Notre Dame,

John Snarey, Candler School of Theology, Bishops Hall, Emory Universi-

ty, Atlanta, GA 30322
Andrew J. Weigert, Department of Sociology, University of Notre Dame,

Marta K. Zahaykevich, Center for the Study of Psychological Develop-

ment, University of Rochester, Lattimore 519, Rochester, NY 14627
Part I Self
1
The Self, Adult Development, and
the Theory of Biography and
Transformation
GILG. NOAM
During the past decade, we have witnessed a great deal of interest in un-
covering the developmental paths of adulthood. Knowledge about these
avenues has found a rich source in the work of the great novelists of the
19th Century to whom it was no secret that the longest era of the life-
span was full of great psychological, social, and spiritual transformations.
But psychoanalysts and psychologists have found it difficult to translate
this knowledge into theory and research paradigms. For psychoanalysts,
the formative years had passed by the age of six. And though very differ-
ent in its general outlook, Piaget's cognitive psychology explored develop-
ment no further than adolescence. Earlier theories that did trace develop-
ments in adulthood (e.g., Buhler, 1933; lung, 1923/1971) did not receive
the attention they deserved.
One important exception is Erikson (1950) who, well adhead of his time,
painted an inspired picture of the eight ages of man. More than any other,
his work has created the momentum towards a field of adult development.
All present day theories of adult development are influenced by his vision,
even when there is disagreement about model and method. Vaillant's (1978)
study of adaptive processes, for example, and Levinson's (1978) functional
model of adult development are direct extensions of Erikson's work.
Piagetian and lifespan theorists, too, such as Gilligan (1982), Kohlberg
(1969), and Baltes, Reese, and Nesselroade (1977), have in general been
critical of psychoanalytic assumptions and yet are indebted to Erikson's
epigenetic framework. Supported by longitudinal research, Erikson and
the recent theoretical developments have nurtured the exciting prospect of
uncovering the forms and rhythms of adult development.
Although there now exists a growing concensus among psychologists
that the lifelong interactions between the individual and the environment
produce many possibilities for movement in the adult years, it is also
widely noted that the field of adult development is in great need of new
theories and systemic research. One important issue that continues to puz-
zle theorists is the relationship between developmental continuities and
discontinuities in a person's life. While a variety of longitudinal studies
4 GiIG. Noam
have uncovered personality continuities over time (e.g., Block, 1971;

Chess, Thomas, & Birch, 1959), research also has shown important discon-
tinuous and qualitative shifts at different stages of development (e.g.,
Kagan & Moss, 1962; Loevinger, 1976). And even though many longitu-
dinal studies have revealed the coexistence of continuous and discon-
tinuous variables (for a discussion, see Thomae, 1979), it has been difficult
to integrate these findings into explanatory models. Furthermore, the crea-
tion of new theory is stifled by questions about the nature of adult develop-
ment. Is adulthood characterized mostly by reworkings of old experiences,
by changes due to environmental conditions, or by the development of
qualitatively new psychological organizations?
Rather than try to explore all these issues in this chapter, I will instead
address a few key themes that guide my approach to lifespan development
and the adult years. I will introduce a general theory that places bio-
graphical continuities and adult self-transformations together in one model,
a model that is steeped in a social-development tradition. Although I am
influenced by Piagetian developmental principles, I have found it necessary
to introduce new conceptual tools to support a broader vision than tradition-
al cognitive or social-cognitive theories have put forth. Eriksonian and
lifespan-developmental theories have fostered my understanding of adult
development. But they also have failed to describe the complex rela-
tionship between a recurrent grammar of self, which is connected to a per-
son's biography, and the continued transformational activities of the self in
adulthood. The framework presented here may be used to study all of the
phases of the lifespan, but it is especially useful for an analysis of the
adolescent and adult years. These eras of life encompass periods in which
the tensions between an awareness of biographical themes and their con-
tribution to self-consistency and self-discrepancy become central.
In the first part of the chapter, I will discuss the background for the
developmental perspective and introduce the general principles of the
theory of biography and transformation. To clarify these constructs, in the
last section I will discuss a literary document, Franz Kafka's letter to his
father.
Theory in Context
The theory presented here was first introduced in a series of papers propos-
ing new views on the self (Noam, 1985; Noam, Kohlberg, & Snarey, 1983),
psychopathology, (Noam, 1986a, 1986b; Noam, 1987), and psychotherapy
(Noam, in press). It was argued that psychoanalytically informed develop-
mental psychology tends to overemphasize replays of early unresolved con-
flict. Piagetians (e.g., Kohlberg, 1969), on the other hand, emphasize con-
tinued evolution of psychological functions without offering an explanatory
1. The Theory of Biography and Transformation 5
model for what might be called "carry-overs" from earlier eras. The person
is rightly viewed as an active, integrating, and synthesizing organism,
considered by some the "producer of development" (Lerner & Busch-
Rossnagel, 1981). The strengths ofthis view, however, are based on princi-
ples that also contribute to important limitations. The Piagetian notion of
structural wholeness and synthesis screens out a view of developmental
discrepancies that have been dealt with in other traditions under the terms
"divided self" (Laing, 1962) and "false self" (Winnicott, 1958). These con-
cepts originally stem from observations of patients in psychiatric treat-
ment. Their importance in less extreme forms can also be traced through
the normal crises and everyday pains of adult life. Thus, it becomes neces-
sary to develop a language that will accommodate these phenomena in
light of our advancing knowledge about integration and cognitive trans-
formation.
Through a number of theoretical deductions and careful clinical observa-
tions, I have found that the relationships between integration and disin-
tegration, between progression and regression, and between the unified
and the divided self can be understood in a systematic fashion. Exploring
the structure of these relationships breaks new ground in the study of con-
structivist activities of both psychological recapitulation and developmen-
tal transformation in adulthood.
This line of work, then, is a constructive-developmental theory that
establishes systematic relationships between the biographical and the
transformational activities of the self. The main focus of constructive-
developmental theory is on the continued development between self and
others. The evolution of the self throughout life brings about new under-
standings of the self's relationships and provides opportunities for recon-
structing the past. This more advanced developmental organization is not,
as it is in psychoanalysis, interpreted as a set of separate ego functions
(such as stress tolerance or defensive style), but as a structure that the self
attempts to impose on social reality. According to this model, the overall
self-structures (and other subdomains such as cognition, affectivity, moral-
ity, etc.) can continue to develop in childhood, adolescence, and adult-
hood, even when aspects of the self remain under the governance of earlier
constructions. Thus, as distinguished from other structural theories in the
Piagetian tradition (e.g., Kegan, 1982; Selman, 1980), the framework of
biography and transformation places these mature structural positions into
a larger context, one in which the earlier self-other grammar is related to a
system of later development. This broad view, which requires a new under-
standing of psychological structure and its developmental course, is appli-
cable to both psychopathology and normal development. Indeed, the study
of psychopathology has served as a lens to magnify typical processes
relevant to the development of the self and personality.
The second, biographical component of constructive-developmental
theory refers not only to the content of life experience, but also to the
6 Gil G. Noam
status of structural integration. Each developmental transformation pro-

vides the opportunity for a new synthesis, but can occur while the person
remains under the influence of parts of the old structure. Under ideal cir-
cumstances, the earlier structures become transformed into content and
are cherished and sustaining images of a history of important interpersonal
relationships. But frequently this integration does not occur and leads to
what I have termed encapsulations, a form of living biography through
which earlier self-systems coexist with later ones. In either event, the ear-
lier forms remain part of a person's frame of reference, creating what I
have called elsewhere a "grammar of self-other relationships" (Noam,
1985). In the positive outcome, we talk of styles, biographical continuities
that interact with the transforming activities of the self; in the more nega-
tive outcome, we talk of encapsulations or discrepancies in the self-system.
Expanding Structural Perspectives for a Theory

of Adult Development
Although I make use of development stages in describing the self and have
posited a stage model of internal and interpersonal aspects of the self, I
differ from other structural-developmentalists in the status that I ascribe
to the stages. Kohlberg (1984), for example, has made the distinction
between "hard stages" and "soft stages." The "hard stage" criteria include
Piaget's sequence of logical development and Kohlberg's justice structures.
Hard stages require clear distinctions between structure and content as
well as competence and performance. The structures form an invariant
sequence, independent of cultural influences; the stages are hierarchical
integrations that are increasingly differentiated and integrated.
In contrast to hard stages, soft stages refer to theories that deal more
with the content and function of personality than with the structure of
cognitive operations. They refer to "the individual's reflections upon the
self's psychology" (Kohlberg, 1984, p. 243) and are self-constructed indi-
vidual theories rather than structural forms of reasoning. The problem with
these theories (e.g., Broughton, 1978; Fowler, 1982; Kegan, 1982; etc.) is
that they have adopted hard-stage principles of structural wholeness,
sequentiality, and hierarchical integration as if they were dealing with the
development of cognitive operations. In other words, soft-stage theorists
directly applied the principles of logical operations to self and life by way of
a method of analogy. I had at first taken a similar approach (Noam, 1985;
Noam & Kegan, 1983) and have only gradually discovered the problems
that emerge. The ever broader radius of structural interpretations (from
intellectual functions to moral judgment, to epistemology, to self and faith)
places a shadow over clear structural categories. More importantly, since
the analogical nature of the models was not taken into consideration, it was
quite literally assumed that the stage transformation was reorganizing all
the categories of self, personality, or faith into a new and structured whole.
But since more and more "soft" contents have been included in the
models, it is impossible to assume that a stage change will reorganize and
transform all the aspects of self and personality as these theorists claim.
Kegan, for example, has introduced stages of the self that parallel the
descriptions of Loevinger, but that take the important step of defining an
underlying structural system at each position. This step has provided im-
portant information regarding the organization and reorganization of the
self. At present, Kegan and his colleagues (Lahey, Souvaine, Kegan,
Goodman, & Felix, 1987) are testing a scoring system called the subject-
object coding manual, which derives an ego stage by coding statements
from a research interview. Kegan's interest in clinical phenomena has
led him and his group to comprehensively and creatively describe and
code developmental transitions. However, like earlier coding systems
(e.g., Fowler, 1982; Oser, Power, Gmeunder, Fritzsche, & Widmer, 1980;
Selman, 1980), the approach again subsumes all scorable statement into a
single-stage structure (or a transition between two stages), systematically
erasing discrepancies in the self's activities through a method of averaging.
This measurement problem has its theoretical source in the model itself.
Using Piaget's principles for the cognitive realm, Kegan proposes their ap-
plication to personality development. Each stage is considered in terms of
the relationship between a current subjective experience and that which
had been subjective in the previous stage and has become objective. The
"object" refers to those feelings, thoughts, and relationships that are held
consciously, that are observed and thus available for coordination and
mediation. The "subject" side of this balance refers to those aspects of the
ego of which a self-observing person has no current awareness-that in
which the person is embedded and from which he has no distance. Dis-
embedding from the subject side and making it objective is the transition
from "being" to "having" (i.e., from being one's relationships to having
one's relationships in the transition from Stage 3 to Stage 4).
This idea, however, assumes that the process of "disembedding" from
the earlier stage (e.g., from impulses at Stage 1, from needs at Stage 2) has
occurred, since the subject-object relationship always makes up the pre-
sent stage or a given transition. How are we to understand the many cases
where the "disembedding" has not fully occurred and the person continues
to remain subjected to earlier self systems? The architecture of the theory
does not account for those frequent occurrences, and one is forced to
interpret the data as a function of the mature self-descriptions. Such an
interpretation identifies what I view as only the most mature aspect of self-
development as the overall self-structure.
Another issue emerges here, relevant to any theory of adult develop-
ment. Most social-cognitive theories view the development of the self as
identical to the strengthening of the self in that the higher stages are more
8 Gil G. Noam
adequate and more comprehensive. In Kohlberg's scheme, for example,

Stage 5 is the morally stronger position about issues of justice and fairness
than all preceding stages. In his and Piaget's models, development of
cognitive structures always means moving toward greater equilibrium in
the organism-environment interaction, which provides the individual with
an expanded capacity for knowledge and adaptation. For this reason,
Kohlberg could claim that "higher stages are better stages." Piagetian
theorists have largely adapted this orientation to personality development.
Again, we have to review whether such a general transfer of principles to
this psychological domain is appropriate. Are higher self stages better than
earlier ones?
In my developmental research with clinical populations (e.g., Noam,
1984), I have frequently observed serious weaknesses of self (e.g., frag-
mentation under stress, severe symptoms, inability to act willfully, etc.)
associated not only with the lower stages of ego and self-development, but
also with the more complex ones. The latter cases show us that high stages
are not necessarily better stages if the criteria are adaptation and a greater
equilibrium in the environment-organism interaction. One could argue
that the domain of psychopathology is responsible for this "noise," but I
would disagree given that the phenomenon can also be observed in people
without clinical problems. I found it necessary to introduce the distinction
between self-complexity and self-strength. Self-complexity refers to the
most mature structural self stage achieved, while self-strength addresses
the status of structural integration. A person can have a great deal of self-
complexity and continuously fluctuate between different developmental
levels (low integration). These concepts will be described in greater detail
in a later section of this chapter.
Piaget himself seems to have been receptive to a transformation of some
of his principles for an understanding of personality development. For ex-
ample, in a rarely quoted section of his book, Insight and Responsibility,
Erikson (1968b) described Piaget's own skepticism about whether structural
stages can encompass the general domain of personality. Erikson states:
He [Piaget] took for granted that, in the healthy child, there is, at any given time, a
high degree of functional unity, that is, an ability to reconcile and co-ordinate the
growth patterns of all the physiological, as well as the mental and emotional func-
tions. This he called the unity of personality. But he discarded as unproven any
claim finding the principles governing the functional unity at a given stage in such a
way that a structural unity [logical relationship between elements of personality to
form a structure, G.N.] for that particular stage could be demonstrated. (p. 135)
Instead, Piaget interpreted personality as "multiple, divisee, et contradic-
toire" and thus not governed by one structural principle. This critique,
found also in other parts of Piaget's work (Piaget, 1973), has provided
additional impetus to the question of how we can more appropriately inte-
grate Piaget's important developmental principles into the study of adult
development.
I will now address the transformational self, which forms the basis of the
most mature aspect of the self.
The Self in Transformation

In a tradition linked with Piaget's psychology and built primarily on Kohl-
berg (1969), Mead (1934), James (1980), and Baldwin (1902), there are
now a number of social-cognitive theories that address areas relevant to a
theory of the self and self concept (Blasi, 1983; Edelstein & Noam, 1982;
Damon & Hart, 1982; Kegan, 1982; Lapsley, in this book; Noam, Powers,
Kilkenny, & Beedy, in press). These theorists examine the self in terms of
an underlying logic of self-understanding or self-other differentiation. The
theories differ in their definitions of the self and in their approach to the
question of whether the self is a unitary process or is comprised of a num-
ber of subdomains (e.g., physical, psychological, social self, etc.).
The theory of self presented here builds on an underlying logic of social
perspective-taking, a theoretical construct that was first introduced in the
self theories of the symbolic-interactionist school of Chicago during the
early part of this century (Baldwin, 1902; Mead, 1934). The symbolic-
interactionists also influenced Piaget (even though he did not ork directly
on a theory of self) and later Kohlberg, as evidenced in his important essay
on identification and the logic of the moral stages (Kohlberg, 1969). Sel-
man has empirically operationalized this construct and conducted a variety
of important developmental studies of the evolution of perspective-taking
in childhood and adolescence (Selman, 1980). However, he, as well as
others who have employed a role-taking perspective (e.g., Chandler, 1973;
Flavell, 1977) have not taken the necessary step of developing an explicit
theory of self, based on the "deep structure" of social perspective taking.
My self-theory is based on the assumption that the activities of the self
always involve role-taking (that is, an understanding that the other is in
some way both like and unlike the self, and that the other knows or is
responsive to the self in a system of complementary expectations). This
notion makes the interpersonal nature of the self a fundamental notion,
also found in the important psychological theories of Sullivan (1953) and
Vygotsky (1986). Vygotsky states: "We are aware of ourselves, for we are
aware of others, and in the same way as we know others; and this is as it is
because in relation to ourselves we are in the same position as others are to
us" (Vygotsky, 1979). Vygotsky viewed the intrapersonal processes as
internalized interpersonal relationships: "Each function in the child's
cultural development appears twice: first on the social level, and later on
the individual level, first between people (interpsychological), and then
inside the child (intrapsychological)" (Vygotsky, 1978). These ideas place
him in close proximity to symbolic interactionists such as Mead and Bald-
win. For Baldwin (1902), what once was attributed to others becomes part
of the self: " ... by imitating the other he (the child) has brought them
10 Gil G. Noam
(significant others) over to the opposite pole, and found them applicable,
with a richer meaning and a modified value, as true predicates of himself
also" (p. 8).
That the genesis of the interpersonal and intrapersonal worlds have the
same origins does not mean, however, that they can be viewed as being
identical. The process of internalization produces many opportunities for
modifications, distortions, and reorganizations. For this reason, I have
found it necessary to develop the category of internal perspective, the
intrapsychic counterpart to the interpersonal perspective of social de-
velopment. The concept refers to the internalized life of interpersonal
relationships with its manifestations in self-reflection, the distinction
between conscious and unconscious processes, and the experience of inter-
nal dialogue.
Table 1.1 summarizes my stages of the developing self-other relation-
ships that are based on social and internal perspectives.
TABLE 1.1. Self-other stages.
1. The Subjective-Physical Self Stage
At the subjective-physical stage, there is no consideration of the other's interests and desires
as different from the self's. There is an emerging awareness of the distinction of physical and
psychological characteristics in people, but mostly, actions are evaluated in terms of physical
consequences. Impulsive responses are typical and feelings are expressed in action language.
Strength is the ability to distinguish between fantasy and reality, to show strong will and to
demonstrate an independent curiosity. These strengths are in part based on at least a partial
achievement of object constancy. The weakness is an emphasis on wish fulfillment, seeing
others as suppliers and being very dependent on them. The concrete perspective on the self
leads to a dichotomous view of being good or bad. In the process, the self hides from or
submits to powerful authority figures that can inflict physical harm. This stage is extremely
rare in adulthood.
2. The Reciprocal-Instrumental Self Stage
At the reciprocal-instrumental stage, there is the possibility of understanding self-interests

and goals as separate from the intent of others. Conflicting interests between self and others
are usually resolved through instrumental exchange. The person can step out of the concrete
bounds of the self, thus creating "two-way" reciprocity. This perspective also changes the
internal perspective-taking ability. There is now the conceptual distinction between the outer
appearance of the "public self" and the "inner hidden self." This creates the possibility of
planned deception through which the self can impose its boundaries. Conflict usually does not
lead to submission or impulsive action, but to self-protective assertion of control. The nega-
tive outcome is opportunism, exploitation, and manipulation. The positive outcome is the
mastery of the tool world and the ability to control feelings and to concentrate on tasks. The
limitation of the reciprocal-instrumental stage is the isolation of two exchange-partners whose
relationship is not primarily guided by trust and altruism, but rather by interest.
3. The Mutual-Inclusive Self Stage
At the mutual-inclusive self-other perspective, the person understands others in a relationship

coordinated through a generalized perspective. The person experiences different points of
TABLE 1.1. (Continued).
view through the "Golden Rule" of seeing reality through the eyes of another person. This
perspective creates the context for altruistic actions and for surpassing the bounds of self-
interest. Attitudes and values are seen as persisting over time, often leading to stereotypes
like "I am that kind of person ... " These "self-traits" in addition to the new internal per-
spective lead to more complex self-observational capacities. The limitations of this stage,
however, are an overidentification with the views of the other and the dangers of conformist
social behavior. It is crucial for the self to be liked and appreciated in order to feel a sense of
esteem. Typical feelings of low self-esteem and a proneness to experienced depression and
anxiety are linked to a sense of abandonment and feeling "lost in the world."
4. The Systemic-Organizational Self State
At the systemic-organizational stage, the societal point of view is distinguished from the inter-
personal one. Multiple mutual perspectives can be integrated into a broader systems views.
When the self takes a systemic perspective on relationships, the communication between
people is seen as existing on a number of levels simultaneously. Individual relations are inter-
preted in terms of their place within a larger system of consciously defined roles and rules.
System maintenance of the self becomes the hallmark of stage.
The person views the self as having control over his or her destiny. It is also the point,
however, at which the person realizes the existence of parts of the self not easily managed by
the system's control (i.e., the discovery of unconscious motivations). The social perspective
also brings out strong motivations of achievement, duty, and competition. The limitations of
the systemic self-other perspective is the attempt to overcontrol self and other, to reflect on
social relations too much in terms of power, role, and status, and to take so many perspectives
on self and other that obsessive-compulsive indecision can result. These contradictions are
reintegrated into a new whole at the Integrated (5) and Universal (6) self-other perspectives.
Developmental Processes of the Self: Assimilation,

Accomodation, and Encapsulation
As is widely known, a schema is a cognitive or mental structure by which
individuals intellectually adapt to and organize the environment. By means
of these structures, events are organized into groups according to common
characteristics. Perhaps Piaget's greatest contribution was to describe the
developmental nature of schemas from birth to adulthood.
Each new life situation creates the opportunity for an organizational
change, or a developmental spurt. This change is what Piaget refers to as
accommodation, that is, the organism's ability to adapt to an inanimate
object or person by changing its internal structure or schema. Important
events in a person's life hold the potential for such critical, qualitative
changes. When the system is very stable, however, it is possible for dra-
matic experiences to be integrated into the existing developmental schemes.
Piaget describes this integration as assimilation, the incorporation of an
object into the existing structure.
The nature of the relationship between assimilation and accommoda-
tion, what Piaget terms equilibration, is the description of process in de-
12 Gil G. Noam
velopment. Equilibrium is the state of balance between assimilation and

accommodation; disequilibrium is the state of imbalance between assimila-
tion and accommodation. Piaget states that equilibrium is a condition to-
ward which the organism strives. When disequilibrium occurs, it provides
an individual the motivation to further assimilation and accommodation
and achieve equilibrium. These processes are based on the idea that the
organism finds new equilibrium after a state of disequilibrium (transition)
in which all earlier self structures are synthesized into a new structure
(stage). Such a perspective on development has influenced many of the
more recent social developmental theories described above.
For this reason, I have developed an analysis that can account for those
aspects of development that remain unaccommodated even when overall
transformation occurs. Such an analysis, I· believe, best accounts for
Piaget's idea of a "multiple, divided and contradictory self," and is central
to what I am calling the theory of biography and transformation. This
theory emphasizes how particular experiences resist integration into a
higher order system-a process I term overassimilation (incorporating ex-
periences into an earlier structure when a more developed structure ex-
ists). The consequences or products of these overassimilations are encap-
sulations. This term refers to pockets of old meaning systems that are
guided by the cognitive and affective logic (Ciompi, 1982) that governed at
the time the encapsulation occurred. By uncovering the logic used in the
encapsulation, a developmental theory of self can reconstruct approx-
imately when in development the formation of structure occurred.
Encapsulations persist in the forms of themes and attributions about the
self and important relationships. Depending on the time that the encap-
sulations occurred, they can be more "physical" (e.g., based on magical
thinking, focused on the body-self and images of bodily survival during
physical separations), "concrete-action oriented" (e.g., based on a view of
the self as an agent that acts on the world or needs to manipulate the world
deceptively to achieve need gratification), or "psychological" (e.g., a state
where needs are expressed in symbolic form around identifications with
others). The encapsulations, as well as the later developmental positions,
can be understood by the rules and stages of an underlying self logic.
Depending on their severity, encapsulations either will have a pervasive
influence on present day adaptation or will inhabit only a small part of the
self. Regardless of severity or time of occurrence, however, encapsulations
are not "archeological entities." Rather, they are a living part of the self's
internal and interpersonal life and could be compared with earlier com-
munallife coexisting with complex forms of urban life in modern societies.
Wheither the relationship between the two forms will be productive or
destructive is determined by the nature of their interaction.
It is important to introduce the distinction between types of encapsula-
tions. First, I will illustrate the encapsulation that I term vertical. We have
often observed that young children can perceive themselves as responsible
for their parents' divorce. What begins as one area of developmentally

appropriate, egocentric thinking in children can be arrested to become a
strong, primitive, living part in an otherwise evolving self-system. Further-
more, such factors as sense of loyalty to the parents can reinforce an un-
changed belief in the person's "basic fault."
Vertical encapsulations result from these early belief systems and are
also a main source of unintegrated affective states. Both remain organized
by the early developmental principles that governed when the encapsula-
tion occurred. Since these "developmental pockets" are also connected to
strong feelings and vulnerabilities about important primary relationships,
much energy is expended to preserve their separateness from the most
mature self-structure. So, with vertical encapsulations, earlier experiences
are clearly separated from the later developmental structures, creating a
distinct, hierarchical discrepancy in the self.
The second type of encapsulations are horizontal. They are camouflaged
discrepancies and appear under the governance of the most advanced
developmental position. In reality, the earlier conflcts are very much in-
volved, but have led not so much to an arrest, as to a specific sophisticated-
appearing derailment. The early experience continues to shape an aspect
of the person's world view, but appears differently (i.e., more complex) at
each point in development. By adulthood, the person has usually attained
a full formal-operational system that can observe the self and develop an
ideology and theory of relationships. It will appear, therefore, that the
most mature developmental position dominates. In fact, however, the ear-
ly derailment has caused a separate developmental line in the self, which
has transformed an early state along the developmental positions of the
rest of the self. Although the integration of horizontal encapsulations into
the overall self-system might seem simpler because there has been con-
tinued development in that realm, the dissolution of the horizontal split
clinically is very difficult. The person often experiences the split as a matter
solely of the present, which makes the critical return to the past source of
the derailment a matter of particular resistance.
Strength and Development of the Self:

Equilibration and Integration
We have seen that accommodative processes do not guarantee that all
internalized relationships will be transformed. Often, the loss involved in
detachment from the old interpersonal constructions is greater than the
gains anticipated in development. Yet the present interpersonal world pro-
vides the opportunities for transformation. What tends to occur in such a
situation is a partial transformation, one that does not generalize to all
aspects of the self.
The idea of partial transformation is in line with some of the findings of
14 Gil G. Noam
recent Piagetian research, which show considerable discrepancies in de-

velopmentallevels achieved across domains. Most interpretations focus on
a lack of generalization due to insufficient performance conditions, such as
the lack of opportunity to apply newly used skills. From the perspective of
the self, I am suggesting an interpersonal interpretation for discrepancies
in competence and performance. The emotional strength of interpersonal
ties, ties that are often distorted and enlarged inside the self (for example,
the negative belief systems about the self of the depressed or the anorexic
patient, which are clear distortions when placed under careful scrutiny),
provide a protective foil against an application of newly found abilities in
another domain. And so for the generalization of new structural abilities
whose roots can be found in interpersonal relationships, I am suggesting an
intrapersonal interpretation.
When we pursue this argument further, we find that a basic developmen-
tal dimension can now be introduced and receive a new importance. The
strength of the self, a construct largely forgotten in Piagetian personality
psychology and central to the theory of biography and transformation, is
tied to the principles of integration and to the origin of encapsulations. The
strong self has integrated and synthesized earlier experiences each time a
developmental transformation has occurred. While there are other non-
developmental contributors to the strength of the self (e.g., temperament,
attention, will), I have observed that integration is a central component of
such strength. With integration, I mean how encompassing earlier trans-
formations have been, or conversely, how much of the earlier organizations
have been encapsulated. If encapsulations have occurred, the person will
have to expend a great deal of energy on the resultant multiple experiences
that pull the individual in different directions. One of the most extreme
examples of nonintegration is given to us by patients who are diagnosed as
multiple personality. Their very basic function of integration is impaired,
leading to a major weakness of the self. Again, observing such patients can
help us to see more clearly the nature of integration in normal develop-
ment.
Even though Erikson proposed his developmental psychopathology
model long before social cognitive self theories were introduced, his em-
phasis on ego virtues remains the most modem statement on "ego strength."
For Erikson, the strengths and weaknesses of the ego always arise out of
the present crisis (e.g., wisdom versus despair in old age). Each new stage
has a positive, strengthening outcome and a negative, weakening one. In
structural theory, the developmental transformation itself is viewed as a
positive outcome.
Another important contribution to knowledge about self-strength is
found in Kohut's concepts (Kohut, 1977) of self-cohesion and fragmenta-
tion. His complex developmental theory not only describes the process of
cohesion and fragmentation in psychoanalysis, but simultaneously offers
an explanation about the causality of the strong and the weakened (e.g.,
narcissistic) self. As is widely known, he views disturbed self-object rela-

tionships between the child and the parents as the core of the weak (frag-
mented) self. Much like Erikson, Kohut offers an interpersonal rather than
a traditional "drive" explanation by viewing the evolution of self-weakness
as a break in empathy in the early mirroring and grandiose phases of de-
velopment. From my perspective, however, Kohut overemphasizes the
earliest conflicts in their relationship to self strength. Most of Kohut's case
examples depict adults in treatment. It is unlikely that the healing interac-
tions (transmuting internalizations) are solely based on a return to child-
hood and parental images, rather than on experiences that strengthen the
self in the course of its continued evolution in adulthood.
By viewing the strength of the self as emerging from a sense of integra-
tion of early and later constructions, I attempt to build a bridge between
the most mature developmental position and the encapsulation.
Consequently, structural development is no guarantee for better mental
health. At each developmental point, problem areas can be transformed
rather than disappear and create greater damage in the self. Because each
stage transition is a form of internalization, there is the possibility for a
paradox outcome: From the perspective of self-development, the person
can have achieved more complexity and a broader self awareness while
from the perspective of self-strength, the person can be more alienated
from their original problems (i.e., vertical encapsulation) or have a more
complex "false self" (horizontal encapsulation). This phenomenon can
often be observed when we compare an adult at the mutual-inclusive and
the systemic-organizational stages. Because at Stage 3 the self is strongly
tied to identifications with significant others, often the vitality and strength
of the self can be restored through more appropriate interpersonal rela-
tionships. At Stage 4, the problems of self-esteem and fragmentation have
become part of the inner organization of the self. Often people feel more
isolated, less able to reach out to others. The alienation potentially turns
into hatred of the self, with suicide becoming one aspect of the theory of
the self (e.g., the only power people have is to control their self-chosen
time of death). It is thus possible that structural development of the self not
only creates mental health or strengthening of the self, but can lead to
worse outcome on these dimensions.
We will now turn to the bridge between encapsulations and transforma-
tions of the self-the core life themes.
Core Themes: A Bridge Between Equilibration

and Encapsulation
Each person holds core themes that are central reference points through-
out life. These themes organize a multitude of life experiences into key
interpersonal and intrapsychic forms. Some refer to them as a narrative,
16 Gil G. Noam
story line, or script. In the study of the self, we need to be attentive to the
major images and metaphors that not only are elaborated and changed
throughout the course of development, but that also show a great deal of
consistency. Life themes are based on continuous relationship with the
environment, instructions about the relationships in the world, and past
experiences filtered through memory. They are a biography of meanings.
Similar to a symphony with its numerous movements, each developmental
step has its own internal organization, rhythm, and mood. However, the
more general themes that span across the entire piece, continuously under-
going elaborations, are part of each separate movement. What sounds at
first like a repetition often ends up being a reworking with important
changes. As with a symphony, one can focus on the new movements and
their differences of tempo and tone or on the continuity of an overarching
"musical project."
The study of evolving lives is most frequently guided by one of two
approaches. One deals primarily with the continuous elaboration of per-
sonality traits, temperament, or cognitive styles. The other approach
focuses on the discontinuous psychological organizations. Life is viewed as
consisting of major points of reorganization-new narratives and styles get
created as a consequence of these transformations. In a musical piece or a
novel, one would have little difficulty embracing both perspectives at once
or receiving pleasure and excitement from the tension. This dynamic inter-
play is difficult to capture in psychological theory and research. But life
themes and their relationship to mature self processes and encapsulations
put this interplay at center stage. While the stages of self-transformation
focus our view on discontinuous constructions about life, core themes
speak to an unfolding of biographical continuities.
The methodology of extracting core themes does not have to rely solely
on clinical and intuitive material. Preferably, one works from the indi-
vidual's detailed descriptions of relationships, scenes, and locations of past
and present. Clinical research interviews, diaries, or more standarized
stimuli such as the Thematic Apperception Test (TAT) are data sources
that can provide the necessary information to generate specific life themes.
Czikszentmihalyi and Beattie (1979) have outlined systematic research in
the area of "life themes." They have defined "theme" as a hierarchical
affective-cognitive system that is composed of a central existential prob-
lem. The problem is surrounded by interpretations of its supposed causes
and by strategies to be used for solving it.
Given the transformational aspect of the theory of biography and trans-
formation, the core themes are used to reveal the bridge between the most
mature aspect of the self and the encapsulations. They are filtered through
the present point of view and through the less differentiated, earlier
perspectives. First we turn to the transformational side of the self. When
the individual is at the mutual-inclusive stage, we expect to see themes of
self-definition through group membership. We expect biographical content
that reflects concern about others' acceptance of us. In contrast, at the
reciprocal-instrumental stage, we can expect to find themes that reveal a

need for concrete control and mastery. Biographical preoccupations are
with behaving in consonance or discordance with others' rules placed on
the self. There is generally an orientation towards people as either "being
on my side" or "curtailing my freedom." But core themes must not be
understood as a function of the continued evolution of the self alone.
In the example of a person at the mutual-inclusive stage, it is important
to determine whether the concern for acceptance by others is "fueled" by a
more primitive belief system that the self will be abandoned and is not
worthy of love. Similarly, it is necessary to explore whether the person at
the reciprocal-instrumental stage believes that there is no possibility for
intimate relationships because all earlier experiences have been hurtful. In
that case, the need for concrete control is linked to a belief that giving up
control will lead to abandonment. What has been viewed as themes emerg-
ing from the Piagetian stage perspective (e.g., Fowler, 1982; Kegan, 1982;
Loevinger, 1976), now becomes part of a larger configuration, housing
early and later images.
It would be a mistake to view the analytic categories of equilibration,
integration, and core theme analysis as separate functions. Rather, I view
them as being intricately intertwined. Much like photography, which builds
on the relationship between light, speed, and distance, the self entails
numerous and complex functions. It is impossible to discuss the essential
functions without getting lost in a multitude of influencing variables. It is
also unnecessary to reduce the study of the self to only one principle. We
must focus on a few basic functions and their inter-relationships when we
study the self.
Thus, structural activity, so brilliantly outlined by Piaget and the social-
cognitive school, must be placed within the larger theoretical construct of
the developing self. The core themes are continuously elaborated and
reorganized through the process of accommodation, yet each reorganiza-
tion requires a return to those parts of the self that are encapsulated in
order to foster integration. This self-integration is important at all ages of
life, but adulthood provides special opportunities to broaden its scope. As
the person deals with the conflict of intimacy, the care for the next genera-
tion, and the anxieties about mortality, the self's radius is frequently
expanded. Developmental transformations hold a special opportunity for
returning to the encapsulations and overcoming the negative aspects of
recapitulation. It is easier to generalize the process of accommodation to
previously non accommodated parts of the self during times of transition
than to do so during times of stability when assimilation prevails. However,
any student of personality knows that even when rapid developments
occur, the forces associated with encapsulations are very powerful and
resist integration. As mentioned earlier, encapsulations are infused with
strong affects and are connected to internalized primary relationships. The
consequences of ignoring the encapsulations and focusing only on the
present transformation have been discussed earlier.
18 Gil G. Noam
In the process of development, new tasks create possibilities for aware-

ness and reactivation of earlier encapsulations that had previously been
less available to self-observation. With this reactivation comes an inten-
sified period of recapitualation. The new developmental position allows for
a reorganization ofthe self. I term this process "weaving in," meaning that
the encapsulations are overcome through a synthesis by which the new
structure encompasses all earlier ones into a larger whole. Frequently, a
return to the encapsulations not only is required to further integration and
self-strength, but is also necessary for the continued evolution of the self to
occur in the first place.
We will now turn to an interpretation of selections from a famous letter
Franz Kafka wrote to his father. Best known for its literary power, the
letter has also inspired psychologists to interpret both biographic themes
and general dimensions of the father-son relationship. My presentation
of Kafka's letter, in nonsequential excerpts, is to demonstrate the rela-
tionship among encapsulations, mature stages of self-organization, and
core themes.
Prisoner of Adulthood: Kafka's Letter to His Father

In 1919, one year before his death, Franz Kafka wrote a letter to his father,
Hermann (Kafka, 1953). This letter has become a well-known literary
document and has been interpreted from a variety of different psychologi-
cal perspectives. What is less known is that Kafka gave the 45-page letter to
his mother to deliver to the father. According to Max Brod, Kafka's long-
time friend, Kafka hoped the letter might remedy his estrangement from
his father. His mother, who disagreed, did not pass the letter on, but in-
stead returned it to Franz. The intensity of feeling and the brilliance of
observation make this an extremely powerful document.
The adult Franz Kafka writes both about his past and present rela-
tionship with his father, an extremely useful approach for our develop-
mental analysis. Interpretations previous to mine have focused mainly on
narcissistic injuries or oedipal feelings evident in Franz's early years. This
is not surprising since Kafka was influenced by Freud's thinking and was in
agreement over the centrality of childhood experiences for the rest of life.
Sokel (1980) writes: "This letter is a remarkable document, not least be-
cause of its cultural-historical significance as one of the most consistent
applications of Freud's theory of the oedipus complex to one particular
life" (p. 146). But clearly, even with Kafka's interpretive orientation, not
all aspects of the relationship between father and son can be reduced to
early trauma and upbringing. So how are we to understand the psycho-
logical developments of Kafka as an adult and how are they related to the
early experiences in his life? I will address these questions in my interpreta-
tions following each selected excerpt from the letter. I do so with consider-
able hesitation because Kafka described not only himself, but also the his-
toric moment, and because our psychological tools are always insufficient
when we address literary creativity.
Dearest Father,
You asked me recently why I maintain that I am afraid of you. As usual, I was
unable to think of any answer to your question, partly for the very reason that I am
afraid of you, and partly because an explanation of the grounds for this fear would
mean going into far more details than I could even approximately keep in mind
while talking. And if I now try to give you an answer in writing, it will still be very
incomplete, because, even in writing, this fear and its consequences hamper me in
relation to you and because the magnitude of the subject goes far beyond the scope
of my memory and power of reasoning. (p. 7)
I was a timid child. For all that, I am sure I was also obstinate, as children are. I
am sure that Mother spoilt me too, but I cannot believe I was particularly difficult
to manage; I cannot believe that a kindly word, a quiet taking by the hand, a
friendly look, could not have got me to do anything that was wanted of me. Now
you are, after all, at bottom a kindly and softhearted person (what follows will not
be in contradiction to this, I am speaking only of the impression you made on the
child), but not every child has the endurance and fearlessness to go on searching
until it comes to the kindliness that lies beneath the surface. You can only treat a
child in the way you yourself are constituted, with vigor, noise, and hot temper, and
in this case this seemed to you, into the bargain, extremely suitable, because you
wanted to bring me up to be strong brave boy. (p. 15).
The first lines of the letter show Kafka's excellent analytic skills, sensitiv-
ity, and ability to give form to his experience. While we are drawn into his
childhood experiences, we cannot ignore the adult Kafka's recognition and
analysis of, as well as reflection on, those experiences. The descriptions in
the first paragraph help us view some of Kafka's core life themes that are
elaborated throughout the letter. His first sentence refers to a discussion
between father and son regarding anxiety. Kafka acknowledges the anxiety
and states that it is at the core of his inability to answer Hermann's ques-
tion. He then refers to feelings of deficiency, that even under the best of
circumstances, he will fall short of answering the question and he will be
incomplete. Finally, the themes are elaborated and the stage is set for the
lifespan drama of father and son.
Kafka makes a distinction between his father's surface appearance and
underlying kindness, an analytic distinction that requires a relatively com-
plex cognitive system. He states an implicit theory that his father's treat-
ment of others is a reflection of the father's own psychological constitution.
Kafka also shows he is very capable of putting himself into his father's
position when he describes Hermann's motivation for his hurtful behavior:
father wants Franz to grow up to be a strong, brave boy.
The first paragraph of the letter intimates the complex psychological
operations of perspective taking, theory building, and motivational analysis
that mark Kafka's adult relationship with his father. If these adult capaci-
ties are present-and indeed they continue to emerge throughout the
20 Gil G. Noam
letter-our analyses should recognize their integrity. Rather than ego

functions we can take for granted, such operations are aspects of a psy-
chological system of meanings that shape the experience of self and
relationships. The life themes that come to the fore link the most mature
self- organization to earlier experiences. Kafka's memories give us a sense
of the early injuries responsible for his sense of deficiency and self-
recrimination.
There is only one episode in the early years of which I have a direct memory. You
may remember it, too. One night I kept on whimpering for water, not, I am certain,
because I was thirsty, but probably partly to be annoying, partly to amuse myself.
After several vigorous threats had failed to have any effect, you took me out of bed,
carried me out onto the pavlatche, and let me there alone for a while in my night-
shirt, outside the shut door. I am not going to say that this was wrong-perhaps
there was really no other way of getting peace and quiet that night-but I mention
it as typical of your methods of bringing up a child and their effect on me. I dare say
I was quite obedient afterwards at that period, but it did me inner harm. What was
for me a matter of course, that senseless asking for water, and the extraordinary
terror of being carried outside were two things that I, my nature being what it was,
could never properly connect with each other. Even years afterwards I suffered
from the tormenting fancy that the huge man, my father, the ultimate authority,
would come almost for no reason at all and take me out of bed in night and carry
me out onto the pavlatche, and that meant I was a mere nothing for him. (p. 17)
Here we witness one of the origins of the difficult father-son relationship.

This theme of the destructive relationship reemerges throughout the letter
for all the stages of development and gives us insight into Kafka's inter-
pretation of the early antecedents of his sense of deficiency. He not only
did not receive the support he required, but was punished for the wish to
be acknowledged. His desire for a stronger connection to his father was
disappointed and led to punitive separation. His ensuing obedience re-
sulted in what Kafka calls "harm." It is difficult to read this passage with-
out reexperiencing with Kafka the terror and the protest over this senseless
rejection. What is most tragic is that the rejection was inflicted with the
authority of the educational principles of the time and the intensions of
raising a "good" person.
Again, Kafka recognizes that the way he asked for attention was irritat-
ing to Hermann. But he also recognizes that his reflective ability to under-
stand his father now does not undo the scars of the injured child, nor their
effect on his sense of wholeness as an adult. Kafka describes a life-long fear
of his father rooted in senseless strictness and rejection. And sadly, this
fear is transformed into a rejection of the self; the lack of acknowledgment
and support from his father turns into Kafka's self-hatred.
In this section, Kafka teaches us that we cannot study adult experiences
without entering the person's biography. The consternations of childhood
despair are the source of an enduring emotional and cognitive grammar
that become engraved into the person's mind, and that can last a lifetime.
The impossibility of getting on calmly together had one more result, actually a
very natural one: I lost the capacity to talk. I dare say I would not have become a
very eloquent person in any case, but I would, after all, have acquired the usual
fluency of human language. But at a very early stage you forbade me to speak.
(p.33)
I cannot recall your ever having abused me directly and in downright abusive
terms. Nor was that necessary; you had so many other methods, and besides, in talk
at home and particularly at business the words of abuse went flying around me in
such swarms, as they were flung at other people's heads, that as a little boy I was
sometimes almost stunned and had no reason not to apply them to myself too, for
the people you were abusing were certainly no worse than I was and you were
certainly not more displeased with them than with me. And here again was your
enigmatic innocence and inviolability; you cursed and swore without the slightest
scruple; yet you condemned cursing and swearing in other people and would not
have it. (p. 35)
You have a particularly beautiful, very rare way of quietly, contentedly, approv-
ingly smiling, a way of smiling that can make the person for whom it is meant
entirely happy. I can't recall its ever having expressly been my lot in my childhood,
but I dare say it may have happened, for why should you have refused it to me at a
time when I still seemed blameless to you and was your great hope? Yet in the long
run even such friendly impressions brought about nothing but an increased sense of
guilt, making the world still more incomprehensible to me. (p. 43)
Kafka now describes additional symptoms that arose out of the conflict
with his father-his inability to express himself. His father's influence
appears to have been omnipresent. Why else was Kafka, who seems to
have felt loved by his mother, unable to develop the capacity to talk in
other settings? We deal with one of the most eloquent writers of the twen-
tieth century, who experiences himself as unable to deal fluently with the
human language. Although there is no reason to doubt Kafka's childhood
fears of his father's powers, it is curious that the adult Kafka still attributes
to his father the power to curb his participation in human discourse. As a
result of the inconsistent rules (Kafka is forbidden to curse yet the father
does so himself), power is viewed not as rational, but as connected to the
authoritarian rule of the more powerful. Kafka's confusing images and
rules are aided by the powerful smile, which was rarely directed to him.
Even when it was "his lot" to receive that smile, it created even greater
confusion, guilt, and aloneness in a world already full of contradictions.
Not even positive expressions can have a productive impact since they are
overassimilated into a negative belief system about the self. Such belief
systems, which I earlier termed encapsulations, exist even when the self
has been transformed in other domains. That these self-constructions are
not only cognitive but can have a persuasive influence on a person's health
becomes strikingly apparent in the next section.
All that occupied my mind was worry about myself, and this in various ways. There
was, for instance, the worry about my health; it began imperceptibly enough, with
now and then a little anxiety about digestion, hair falling out, a spinal curvature,
22 Gil G. Noam
and so on; intensifying in innumerable gradations, it finally ended with a real ill-
ness. But since there was nothing at all I was certain of, since I needed to be
provided at every instant with a new confirmation of my existence, since nothing
was in my very own, undoubted, sole possession, determined unequivocally only by
me-in sober truth a disinherited son-naturally I became unsure even of the thing
nearest to me, my own body. I shot up, tall and lanky, without knowing what to do
with my lankiness, the burden was to heavy, the back became bent; I scarcely dared
to move, certainly not to exercise, I remained weakly; I was amazed by everything I
could still command as by a miracle, for instance, my good digestion; that sufficed
to lose it, and now the way was open to every sort of hypochondria; until finally
under the strain of the superhuman effort of wanting to marry (of this I shall speak
later), blood came from the lung ... (p. 91)
Kafka illustrates a vivid awareness of his own actions against himself. At

developmentally less advanced positions, found quite frequently in adult-
hood, he would blame his father for "having made him the way he is." But
although Kafka regards his relationship with the father as the ultimate
reason for his suffering, he also highlights his self-destructive worry and
hypochondria. Franz Kafka does not own himself, but is estranged from
himself, and he views this estrangement as a central reason for his suffer-
ing. This alienation from the self, a sense of being "disinherited," is very
typically associated with the systemic-organizational stage of self develop-
ment. As described in Table 1.1, Chandler called a similar phenomenon in
cognitive development "epistemological loneliness of formal operations."
Piaget's full formal-operational thinking underlies the constructions of the
systemic-organizational self. In contrast to earlier stages, the person now
reflects on hypothetical selves and deals with choices of what to make the
"me." We often find a tension between feeling entitled to have control
over one's destiny and a sense of paralysis over the many parts of the self
that defy control. The result is quite frequently a loss of ownership of the
self, with accompanying fears of being trapped and despair over isolation.
The intensity of Kafka's childlike view of his father as all-powerful
makes it understandable why psychoanalysts have focused almost exclu-
sively on the early father-son relationship. It would be a mistake, however,
to view this process as tied solely to early childhood or to complex self-
constructions. As we have seen in Kafka's case, the themes of self-hate
have earlier biographical roots that have remained alive throughout his
life. But their life is now also formed by the adult self who gives the emo-
tions, experiences, and beliefs (all part of what I introduced as themes)
their new form. The rejection continues, so we are told, not much dif-
ferently during adulthood than it did throughout childhood. But there is
now an added dimension, the most mature aspect of the self. This mature
self provides Kafka with reflective abilities capable of rationalizing the im-
pact of the rejection. We find in these capacities the tools for overcoming
the self-destruction. Quite frequently, the kind of complex insights Kafka
uncovered are escorted by feelings of loss and mourning. In the process, a
more humane and flexible self-system can develop that feels less dedicated
to willing fate and to a past that could not blossom. Some can work out
these issues within the bounds of the systemic-organization self. But when
the structure has such rigid boundaries and is so strongly infused by self-
hate, what is required is more dramatic breakdown of the system and
movement beyond the systemic-organizational perspective.
We now jump many years forward and deal with two aspects of the adult
Kafka's relationship with his father-work and marriage.
You struck nearer home with your aversion to my writing and to everything that,
unknown to you, was connected with it. Here I had, in fact, got some distance away
from you by my own efforts, even if it was slightly reminiscent of the worm that,
when a foot treads on its tail end, breaks loose with its front part and drags itself
aside. To a certain extent I was in safety; there was a chance to breathe freely. The
aversion you naturally and immediately took to my writing was, for once, welcome
to me. My vanity, my ambition did suffer under your soon proverbial way of hailing
the arrival of my books: "Put it on my bedside table!" (usually you were playing
cards when a book came), but I was really quite glad of it, not only out of rebellious
malice, not only out of delight at a new confirmation of my view of our relationship,
but quite spontaneously, because to me that formula sounded something like:
"Now you are free!" Of course it was a delusion; I was not, or, to put it most
optimistically, was not yet, free. My writing was all about you: all I did there, after
all, was to bemoan what I could not bemoan upon your breast. It was an inten-
tionally long-drawn-out leave-taking from you, yet, although it was enforced by
you, it did take its course in the direction determined by me. But how little all this
amounted to! (p. 85)
Kafka's need to free himself from his father through adult activities
(e.g., work) is "supported" by his father's rejection. It is an ingenious re-
framing: the father's distance can become the son's freedom. But even
that freedom is fueled by the encapsulated belief of being deficient and
unworthy, expressed through the image of the worm. The subhuman has
good survival skills (can survive a split), but can do little more than drag
itself. Here again, Kafka repreats the basic theme of having been injured.
Many authors have interpreted Kafka's writings as being about his father
and his family (e.g., Heller, 1974; Stern, 1980). And in this passage, Kafka
confirms that view: "my writing was all about you. All I did there, after all,
was to bemoan what I could not bemoan on your breast."
Kafka knows that autonomy built on rejection cannot lead to true free-
dom. But with a tinge of optimism, he recognizes that maybe one day he
will be able to achieve this freedom. Still, in the context of his illness and
other statements in the letter, the optimism is not very credible. I view
Kafka as moving out of the systemic-organizational self position; he takes a
perspective on its many limitations. The move, however, is incomplete in
that he has not delivered an intimate form of self-acceptance or recognition
of the limitations in his relationships to his father. I contend that Kafka's
encapsulated relationship with his father has created a ceiling effect in his
24 Gil G. Noam
continued self-evolution. The earlier conflicts and problems cannot be re-

solved since he is not able to move far enough beyond the systemic-organi-
zational self to relativize the system in which he feels trapped. Thus, he
cannot make peace with the earlier constructions. For this reason, I view
Kafka, while developmentally far advanced and cognitively complex, as
suffering from a strong sense of self-fragility. Kafka continues to shape his
experience by the earlier images of his father, but at the time applies his
more mature developmental perspectives.
Since he continues to employ earlier belief systems of thinking and feel-
ing while carrying knowledge about his condition, Kafka is trapped twice
without being able to get untrapped. This situation frequently leads people
to "choose suicide." This feeling of being trapped becomes even stronger
when Kafka presents his images of marriage.
Now, regarding my attempts at marriage there is much you can say in reply, and
you have indeed done so: you could not have much respect for my decision since I
had twice broken the engagement with F. and had twice renewed it; since I had
needlessly dragged you and Mother to Berlin to celebrate the engagement, and the
like. All this is true-but how did it come about? (p. 109)
Why then did I not marry? There were certainly obstacles, as there always are,
but then, life consists in taking such obstacles. The essential obstacle, however,
which is unfortunately, independent of the individual case, is that obviously I am
mentally incapable of marrying. This manifests itself in the fact that from the mo-
ment I make up my mind to marry I can no longer sleep, my head burns day and
night, life can no longer be called life, I stagger about in despair. (p. 111)
I will try to explain it in more detail. Here, in the attempt to marry, two seeming-
ly antagonistic elements in my relations with you unite more intensely than
anywhere else. Marriage certainly is the pledge of the most acute form of self-
liberation and independence. I would have a family, in my opinion the highest one
can achieve, and so too the highest you have achieved; I would be your equal; all
old and ever new shame and tyranny would be mere history. It would be like a fairy
tale, but precisely there does the questionable element lie. It is too much; so much
cannot be achieved. It is as if a person were a prisoner, and he had not only the
intention to escape, which would perhaps be attainable, but also, and indeed simul-
taneously, the intention to rebuild the prison as a pleasure dome for himself. But if
he escapes, he cannot rebuild, and if he rebuilds, he cannot escape. If I, in the
particular unhappy relationship in which I stand to you, want to become indepen-
dent, I must do something that will have, if possible, no connection with you at all;
though marrying is the greatest thing of all and provides the most honorable inde-
pendence, it is also at the same time in the closest relation to you. To try to get out
of all this has therefore a touch of madness about it, and every attempt is almost
punished with it. (p. 113)
Again, Kafka is quite capable of describing the complex and ambiguous

situation of deciding to marry. He is even capable of giving a truly inspiring
definition of marriage, one usually connected to an advanced form of self-
development. Kafka's view that a union with another person is simul-
taneously the most intimate act and the most mature form of self-liberation
and independence attests to Kafka's complex self-theory. But Kafka is so

far away from the realization of this ultimate bond that he refers to it as a
fairy tale. And more importantly, as we think about the core themes that
house earlier encapsulated issues and the most mature aspect of the self,
the metaphor of prison can provide us a most important entry point into
Kafka's inner life. Kafka wants to escape his inner prison. But rather than
just wanting freedom, he thinks of rebuilding the prison into a pleasure
dome. I assume this means that Kafka does not view separation as the
solution to his problems (breaking out of prison), but rather a recon-
structed and more pleasurable connection. Still, the image of a dome built
out of a prison also possibly shows Kafka's continued indebtedness to a
confined systemic-organizational world in contrast to viewing pleasure-
attainment as a process.
It is precisely this close relation that partly lures me toward marrying. I picture
the equality which would then arise between us-and which you would be able to
understand better than any other form of equality-as so beautiful because then I
could be a free, grateful, guiltless, upright son, and you could be an untroubled,
untyrannical, sympathetic, contented father. But to this end everything that ever
happened would have to be undone, that is, we ourselves should have to be can-
celled out.
But we being what we are, marrying is barred to me because it is your very own
domain. Sometimes I imagine the map of the world spread out and you stretched
diagonally across it. And I feel as if I could consider living in only those regions that
either are not covered by you or are not within your reach. And in keeping with the
conception I have of your magnitude, these are not many and not very comforting
regions-and marriage is not among them. (p. 113)
In this section of the letter, Kafka's pessimism about the possibility of
change becomes apparent one more time. We have encountered the bleak
hopelessness in many of his books. Why is it that everything would have to
be undone and father and son would have to be cancelled out? Is there not
human metamorphoses (instead of becoming an insect) that make hu-
mans more human and forgiving, even when injury has occurred in the
past? Kafka cannot fathom a world in which he can transform his rela-
tionships and himself. He cannot enter that process, even though he has
some glimpses of a psychological world in which such freedom can exist.
All the creativity and expressive genius has not provided him with an
image of how he could move forward without forgetting nor being impris-
oned by the memories. The picture of the father who inhabits most of this
globe tells us how impossible Kafka finds it to develop a realistic and adult
construction of him. Clearly, this image is not in line with the complex
systemic-organizational self-structure, but it is a powerful internal force of
constraint. It is one of the reasons why Kafka cannot develop further. The
relationship between encapsulated beliefs and the most mature self system
needs also to be reversed: Would Kafka begin to exit his adult prison, he
could also begin to reconstruct his past view of his important childhood
26 Gil G. Noam
relationships. The tragedy we feel is that he is so close and yet so far from
being able to take the liberating steps. And he knows it.
Conclusion
In beginning this project, it was my goal to enter an exciting stage of theory
development of ego, self, and identity by proposing a broad view on the
development of the self. Consequently, the chapter sets out to take the
reader on a difficult journey. Many concepts needed to be reviewed, terms
introduced, and implications discussed. And despite the detailed descrip-
tions, what is presented here is but a beginning of uncovering new develop-
mental principles relevant to normality and pathology.
First, I questioned the viability of a unified stage model, arguing that
discrepancies in the self-system and personality development are known
phenomena and yet are not adequately addressed within existing structual
models. I argued that psychopathology serves as a magnifying glass for
those aspects of the self that do not become synthesized in the process of
development and require scrutiny for our understanding normal develop-
ment. Second, it was shown that a new model of development is necessary,
one that encompasses processes of biography and transformation. Building
blocks of this perspective were introduced and related to the key concepts
of the development and the integration of the self. Third, this framework
was then applied to issues of adult development. Kafka's letter to his father
exemplified aspects of the theory.
Even though psychological interpretations of literary documents usually
do violence to the original text, I was willing to take the risk. Rarely has
anyone written as honestly about himself, his childhood, and his adult rela-
tionship with his father as Franz Kafka has. It is easy to distance oneself
from Kafka's experiences by viewing them as relevant only to another
epoch. Or, one can interpret him as suffering from psychopathology. In-
deed, it is more difficult to acknowledge his writing as part of the human
condition when his special sensitivities and talents converge in statements
more blunt and extreme than most children describe to their parents. If we
take Kafka seriously, we also have to rework some of our notions about
adult development. The clean, stepwise, and hierarchical movements
through stages of ego or self can create an illusion of order. As ideologies
of psychological upward mobility, they send up screens against another
adult reality: the struggle against disintegration, repetition, conflicting
affective states, and early belief systems. It does not require Franz Kafka to
know about the many adults who, for example, view people they are close
to as inhabiting "regions of the globe."
These observations have led some to abandon developmental theory
altogether and to divide up adult experience into smaller and only loosely
connected units (e.g., ego functions, traits, etc.). The theory of biography
and transformation builds on the strength of a view of consistency and

organization of experience, while allowing for discrepancies and disor-
ganization in the self-system. Although clinical observations and literary
cases have provided excellent insights, more systematic research is present-
ly underway. At present, interviews are being performed in a longitudinal
study of adolescents. Future research will include studies on adults at dif-
ferent points in their lives. But for now, Kafka's letter provides us with an
example of the explanatory model provided by the theory of biography and
transformation.
Acknowledgments. I would like to thank Barbara Anderson, Sophie

Borst, Nadine Bozek, and Tamara Claman for excellent comments and
editorial assistance in the preparation of this chapter.
REFERENCES
Baldwin, 1.M. (1902). Social and ethical interpretations in mental development. New
York: Grune & Stratton.
Blasi, A. (1983). The self and cognition. In B. Lee & G. Noam (Eds.), Develop-
mental approaches to the self. New York: Plenum.
Biates, P.B., Reese, H.W., & Nesselroade, 1.R. (1977). Life-span developmental
psychology: Introduction to research methods. Monterey, CA: Brooks/Cole Pub-
lishing.
Block, 1. (1971). Lives through time. Berkeley: Bancroft.
Broughton, 1. (1978). The development of concepts of self, mind, reality, and
knowledge. New Directions for Child Development, 1, 75-100.
Buhler, C. (1933). Der menschliche Lebenslauf als psychologisches Problem. Leip-
zig: Hirzel.
Chandler, M. (1973). Egocentrism and antisocial behavior: The assessment and
training of social perspective-taking skills. Developmental Psychology, 9, 326-
332.
Chess, S., Thomas, A., & Birch, H.G. (1959). Characteristics of the individual
child's behavioral response to the environment. American Journal of Ortho-
psychiatry, 29, 791-802.
Ciompi, S. (1982). Affektologik. Stuttgart: Klett-Corta.
Czikszentmihalyi, M., & Beattie, O.V. (1979). Life themes: A theoretical and
empirical exploration of their origins and effects. Journal of Humanistic Psychol-
ogy, 19, 45-63.
Damon, W., & Hart, D. (1982). The development of self-understanding from
infancy through adolescence. Child Development, 53(4),841-864.
Edelstein, W., & Noam, G. (1982). Regulatory structures of the self and post
formal operations in adulthood. Human development, 25,407- 422.
Erikson, E. (1950). Childhood and society. New York: Norton.
Erikson, E. (1968b). Insight and responsibility. New York: Norton.
Flavell, 1.H. (1977). Cognitive development. Englewood Cliffs, Nl: Prentice Hall.
Fowler, 1.W. (1982). Stages offaith. New York: Harper & Row.
28 Gil G. Noam
Gilliagan, e. (1982). In a different voice. Cambridge, MA: Harvard University

Press.
Heller, E. (1974). Franz Kafka. Princeton, NJ: Princeton University Press.
James, W. (1980). Principles of psychology. New York: Holt, Rinehart & Winston.
Jung, C.G. (1923/1971). Psychological Types. Bollingen Series, Vol. XI. Princeton,
NJ: Princeton University Press.
Kafka, F. (1953). Letter to his father. Cambridge, MA: Harvard University Press.
Kagan, J., & Moss, H.A. (1962). Birth to maturity: A study in psychological
development. New York: Wiley.
Kegan, R. (1982). The evolving self. Cambridge, MA: Harvard University Press.
Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach
to socialization. In D. Goslin (Ed.), Handbook of socialization, theory and re-
search (pp. 347-480). New York: Rand McNally.
Kohlberg, L. (1984). The psychology of moral development. San Francisco: Harper
& Row.
Kohut, H. (1977). The restoration of the self. New York: International Universities
Press.
Lahey, L. Souvaine, E., Kegan, R., Goodman, R., & Felix, S. (1987). The subject-
object interview: A guide to its administration and coding. Unpublished manu-
script. Cambridge, MA: Harvard Graduate School of Education.
Laing, R.D. (1962). The divided self. New York: Partheon Books.
Lapsley, D., & Rice, K. (1988). The "new look" at the imaginary audience and
personal fable: Toward a general model of adolescent and ego development. In
D. Lapsley & F.e. Power (Eds.), Self, ego, and identity: Integrative approaches
(pp. 109-129). New York: Springer-Verlag.
Lerner, R.M., & Busch-Rossnagel, N. (Eds.), (1981). Individuals as producers of
their development: A life-span perspective. New York: Academic.
Levinson, D.J. (1978). The seasons of a man's life. New York: Knopf.
Loevinger, J. (1976). Ego development. San Francisco: Jossey-Bass.
Mead, G.H. (1934). Mind, self and society. Chicago: University of Chicago Press.
Noam, G., & Kegan, R. (1983) Social cognition and psychodynamics: Towards a
clinical-development psychology. In W. Edelstein & M. Keller (Eds.): Perspekti-
vitat und Interpretation (pp. 218-238). Frankfurt: Suhrkamp.
Noam, G. (1984). Self, morality and biography: Studies in clinical-developmental
psychology. Unpublished doctoral dissertation. Harvard University.
Noam, G. (1985). Stage, phase, and style: The developmental dynamics of the self.
In M. Berkowitz, & F. Oser (Eds.), Moral education: Theory and application
(pp. 322-346). Hillsdale, NJ: Lawrence Erlbaum Associates.
Noam, G. (1986a). Borderline personality disorders and the theory of biography
and transformation (part I). McLean Hospital Journal, XI (1), 19-43.
Noam, G. (1986b). The theory of biography and transformation and the borderline
personality disorders: A developmental typology (Part II). McLean Hospital
Journal, XI, (2),79-105.
Noam, G. (1987). Marking time in the hardest movement: Borderline psycho-
pathology in lifespan perspective. In K. Field, G. Wool, & B. Cohler (Eds.),
Psychoanalytic perspective on learning and education. New York: International
University Press.
Noam, G. (in press). The structural theory of biography and transformation:
Foundation for clinical-developmental therapy. In S. Shirk (Ed.), Cognitive de-

velopment and child psychotherapy. New York: Plenum Press.
Noam, G., Kohlberg, L., & Snarey, J. (1983). Steps toward a model of the self. In
B. Lee & G. Noam (Eds.), Developmental approaches to the self (pp. 59-134).
New York: Plenum Press.
Noam, G., Powers, S., Kilkenny, R., & Beedy, J. (in press). The interpersonal self
in lifespan perspective. In R. Lerner, P. Baltes, & D. Featherman (Eds.), Life-
span development and behavior (Vol. 10). Hillsdale, NJ: Lawrence Erlbaum &
Associates.
Oser, F., Power, c., Gmeunder, P., Fritzsche, U., & Widmer, K. (1980). Stages
of religious judgment. In J. Fowler (Ed.), Inward moral and religious maturity
(pp. 277-315). Morristown, NJ: Silver Burdett.
Piaget, J. (1973). Affective unconscious and cognitive unconscious. The child and
reality. New York: Grossman.
Selman, R.L. (1980). The growth of interpersonal understanding: Developmental
and clinical analyses. New York: Academic Press.
Sokel, W.H. (1980). Freud and the magic of Kafka's writing. In J.P. Stern (Ed.),
The world of Franz Kafka (pp. 145-158). New York: Holt, Rinehart & Winston.
Stern, J.P. (1980). The world of Franz Kafka. New York: Holt, Rinehart & Win-
ston.
Sullivan, H.S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Thomae, H. (1979). The concept of development and life-span developmental
psychology. In P.B. Blates, & O.G. Grim, Jr. (Eds.), Life-span development and
behavior, 2, 281-312.
Vaillant, G. (1977). Adaptation to life. Boston: Little Brown.
Vygotsky, L. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Vygotsky, L. (1979). Consciousness as a problem of psychology of behavior. Soviet
Psychology, 170, 29-30.
Vygotsky, L. (1986). Thought and language. Cambridge, MA: MIT Press.
Winnicott, D.W. (1958). Through pediatrics to psychoanalysis. New York: Basic
Books.
2
A Social-Cognitive Account of the
Self's Development
ANTHONY G. GREENWALD
Reviewing recent social psychological literature on the self, Greenwald

and Pratkanis (1984) concluded that the (adult) self is "a complex, person-
specific, central, attitudinal schema." In this description, the self is char-
acterized as complex because it incorporates a great variety of knowledge;
as person-specific because it is an idiosyncratic knowledge structure; as cen-
tral because it is a major (perhaps the major) structure of personality; as
attitudinal because it is invested with the affect that is associated with one's
sense of self-worth; and, most importantly, the self is identified as a schema
because it is an organized structure of knowledge.
Part of the importance of this recently developed organization-of-
knowledge conception of the self (Greenwald & Pratkanis, 1984; Kihl-
strom & Cantor, 1984) is that it firmly links the self, as a psychological
construct, to cognitive and social psychological research methods. As ex-
amples of this link, the self has recently been investigated in studies using
procedures that were introduced in other contexts to examine (a) depth
of (cognitive) processing (e.g., Keenan & Baillet, 1980; Rogers, Kuiper,
& Kirker, 1977), (b) cognitive schemata or prototypes (Markus, 1977), (c)
memory (Greenwald, 1981; Rubin, 1982), (d) judgment biases (Green-
wald, 1980), and (e) unconscious processes (Bargh, 1982). (See Pratkanis
and Greenwald [1985] for an overview of research procedures used in
recent investigations of the self.)
Comparative and developmental psychologists (e.g., Gallup & Suarez,
1986; Lewis, 1986) have documented what is presumably the earliest
appearance of the self in development, the achievement of mirror self-
recognition (occurring between 18 and 24 months of age in humans). It is a
long developmental path from the mirror-recognized self to the adult self
that has been the focus of recent research by social, personality, and cog-
nitive psychologists. This chapter describes an interpretation of the self's
development that includes a characterization of the intermediate de-
velopmental stages. However, the chapter's main focus is on variations in
the foundations of the adult's sense of self-worth, including a social-
developmental analysis of the basis for those variations.
2. A Social-Cognitive Account of the Self's Development 31
The Self as an Organization of Knowledge

A central insight of the emerging discipline of cognitive science is that in-
telligent performances depend on highly organized knowledge structures
that are often domain specific. Terms used to describe these knowledge
structures include prototype, script, frame, semantic network, mental mod-
el, and schema. As noted in recent reviews by Greenwald and Pratkanis
(1984) and Kihlstrom and Cantor (1984), such terms have also been em-
ployed extensively in recent treatments of the self. The broad adoption of
these usages signals the evolution of a new metaphor for the self-the self
as an organization of knowledge. Pratkanis and Greenwald (1985) describe
the manner in which this metaphor for the self has replaced older
metaphors that did not serve so well to guide research.
LEVELS OF REPRESENTATION
The present analysis interprets the self as a knowledge structure that

emerges in the course of human development. (In contrast, most pre-20th-
century treatments equated the self with an ever-present soul; see, e.g., the
critical review by James [1890].) The cognitive developmental assumptions
of the present analysis are based on an account of mental representation in
terms of levels of representation (Greenwald, 1987). In particular, the
analysis assumes that human mental representations can be characterized
by five levels, which, in order of increasing complexity, employ as their
representational units features, objects, categories, propositions, and
schemata. These levels of representation (LOR) are interrelated in a
hierarchical structure, such that the units of each level above the lowest
are composed of multiples of units of the immediately lower level. This
five-level (feature-object-category-proposition-schema) representational
structure is designated LOR hs ("h" to indicate that it is a system for human
representations and "5" for its number of levels).
Each of LORhs's five systems of representations models qualitatively
distinct properties of the environment. Features represent basic sensory
properties, such as red color, loud sound, warm temperature, bitter taste,
and textural roughness. Features combine by gestalt grouping relations
into objects. Objects are grouped by the class inclusion relation into cate-
gories. Propositions combine categories, usually including a function (e.g.,
action) category together with a suitable set of argument (e.g., agent,
object, and instrument) categories. Schemata are ruleful combinations of
propositions, such as scripts (Schank & Abelson, 1977), frames (Minsky,
1975), and mental models (Johnson-Laird, 1983).
In LORhs (or other theories using the hierarchical LOR form), each
ascending level models properties of the environment that are not captured
by lower levels (these are emergent properties of the representational
structure). The emergent property of the object level of representations is
32 Anthony G. Greenwald
physical identity, which is a conservation of object-ness through physical

transformations of features such as rotation or occlusion. (This is an emer-
gent property in the sense that features, considered separately, do not
possess object-ness.) For category representations, the major emergent
property is abstract identity, a relationship between the objects that share
category membership. Emergent properties of the system of propositional
representations include agency (the relation between actor and action in a
proposition) and instrumentality (the relation between an object and an
action that employs the object). Consistency is a general designation for
the emergent properties of schema representations; it is the characteristic
of propositions that are eligible to coparticipate in a schema. Among the
forms of consistency, for different types of schemata, are narrative coher-
ence, analogy, logical proof, cognitive consonance (Festinger, 1957),
cognitive balance (Heider, 1958), self-consistency (Lecky, 1945), legality
(consistency with a body of laws or rules), morality (consistency with prin-
ciples of correct conduct), and empirical validity (consistency with a body
of data).
DEVELOPMENT OF THE SELF THROUGH

LEVELS OF REPRESENTATION
The proposition and schema levels of LORhS have a conceptual corres-
pondence to the two highest stages of Piaget's conception of cognitive de-
velopment, concrete operations and formal operations (e.g., Inhelder &
Piaget, 1956; Piaget, 1954). Use of Bruner's (1966) subdivision of Piaget's
sensorimotor and preoperational stages into enactive, iconic, and symbolic
stages permits a loose mapping of all five LORhS levels onto the resulting
five developmental stages, as follows: (a) enactive (feature representa-
tions), (b) iconic (objects), (c) symbolic (categories), (d) concrete opera-
tions (propositions), and (e) formal operations (schemata). This mapping
implies that LORh5 's five levels conform to a developmental ordering,
which can be applied to the development of self by identifying stages in the
selfs development that correspond to LORhS's levels.
At the lowest level of LORhS , stimulus features associated with in-
teroceptive and proprioceptive (versus exteroceptive) stimulation provide
a primitive basis for self-other discrimination. At the object level, the self
is perceived as an object distinct from its background. At the category
level, the self is identified as a member of the category persons. As the self
becomes a frequent occupant of the role of agent (or actor) in propositions,
the cognized self acquires the property of agency.
The self's greatest cognitive elaboration occurs at the schema level; the
self participates in several types of schemata. The self-concept is a descrip-
tive schema, a collection of related, complementary attributes. (Descrip-
tive schemata are sometimes referred to as frames [Minsky, 1975] or men-
tal models [Johnson-Laird, 1983].) One's autobiographical, or episodic,
memory constitutes a narrative schema-a temporal ordering of events.

The ability to cognize causal relationships involving the self presumably
underlies one's sense of personal control or self-efficacy (or the lack there-
of). Logical or quasi-logical relationships among propositions define a class
of inferential schemata, which can range from the loose inferences of com-
mon sense to the highly constrained ones of formal logic. Theoretical
accounts of quasi-logical cognitive processes (e.g., Festinger, 1957; Heid-
er, 1958; Lecky, 1945) assert that these processes play an important role in
self-cognition. The type of schema that is of greatest concern in this chap-
ter is the evaluative self-schema-a knowledge structure that underlies
one's sense of self-worth, or self-esteem, and that plausibly involves de-
scriptive, narrative, and inferential subschemata. This chapter's focus on
evaluative self-schemata can be contrasted with the metacognitive focus of
Lapsley and Quintana's (1985) recent social-cognitive account of the self's
development.
Chronology of the Self's Cognitive Development

Although an alignment of LORhs's five levels with Piagetian stages of
cognitive development has been suggested, it is premature to associate the
corresponding levels of the self's cognitive development with the age
ranges that have traditionally been associated with Piaget's stages. In par-
ticular, it seems likely that cognition of the self's causal efficacy substantial-
ly precedes the traditional age 7 onset of concrete operations, and similarly
that schematic representations such as autobiographical memory and self-
concept substantially precede the age 13 onset of formal operations.
(These traditional age boundaries of Piaget's stages are in any event con-
troversial, see, e.g., Gelman & Baillargeon [1983]. The LOR hs theory is
more concerned with the order in which representational abilities appear
than with their absolute age levels, which remain a matter for empirical
investigation. )
Varieties of Evaluative Self-Schemata

and Their Development 1
The concept of ego tasks was developed in a review (Greenwald, 1982) that
attempted to interpret the often-confusing body of research on ego-
involvement that was published between the mid-1930s and the early
1960s. That review established that, even though there was no general rec-
ognition of the fact by researchers using the term, by approximately 1960
"ego-involvement" had acquired three conceptually distinct meanings.
1 The material in this section is based on development of the concepts of ego task
analysis by Greenwald (1982), Greenwald and Pratkanis (1984), Greenwald and
Breckler (1985), and Breckler and Greenwald (1986).
Ego-involvement}: Concern about public impression, or evaluation by other per-

sons; similar to evaluation apprehension or need for approval.
Ego-involvement2 : Concern about self-evaluation, private self-image, or evaluation
by oneself; similar to self-esteem maintenance or need for achievement.
Ego-involvement3: Concern about importance in relation to central values.
Greenwald and Breckler (1985) linked these three senses of ego-

involvement to an analysis of self-presentation processes, in which each
sense of ego-involvement was taken to reflect concern about the evaluation
of one's worth from the perspective of a particular audience. These were,
respectively, the audience of important or powerful others (ego-involve-
ment l ), an inner audience (ego-involvement2), and the audience of one's
reference groups (ego-involvement3). An "ego task" is the task of
earning the favorable evaluation of one of these audiences. The ego tasks
corresponding to the three types of ego-involvement are social accredita-
tion (favorable evaluation by important others), individual achievement
(favorable evaluation by one's self), and collective achievement (achieve-
ment of reference group goals). Breckler and Greenwald (1986) inter-
preted these three ego tasks as corresponding, respectively, to public,
private, and collective facets of the self. Table 2.1 summarizes this analysis.
The public, private, and collective facets of the self are accompanied there
by a fourth facet, identified as the diffuse self. The diffuse self is, in some
senses, a preself or a nonself. Rather than being oriented toward achieving
favorable self-evaluation, the diffuse self is conceived as having the hedo-
nic goals of seeking pleasure and avoiding pain. In terms of the representa-
tional concepts of LORh5 , each of the ego tasks in Table 2.1 is linked to a
schema-level cognitive structure that, in effect., provides an algorithm for
computing one's self-worth in terms of the appropriate audience's evalua-
tions.
The diffuse self is assumed to precede the other facets, being the self that
exists prior to the formation of schema-level cognitive structures. When
the child learns to value the approval of parents, authorities, and peers, the
diffuse self is supplemented by a public self. Those others' criteria for
approval will eventually become familiar, and can be anticipated and inter-
nalized, leading to development of the private self, which is initially an
internalized subset of others' standards. The collective self similarly de-
pends on internalization, but also on the establishment of an identity rela-
tionship with a reference group, leading to adoption of the group's values
and goals. The development from diffuse to public to private to collective
self is conceived as a sequence in which (a) new facets of the self are added
to those already developed, (b) each facet continues to develop after its
initial appearance, and (c) the relative strengths of facets will vary from
person to person. The diffuse, public, private, and collective facets of the
self can be seen as having a rough equivalence to the first four of Kohl-
TABLE 2.1Interrelation offacets ofthe self, ego tasks, personality measures,

experimental procedures, and performance strategies.
Facets of the
self Diffuse self Public self Private self Collective self
Ego task Hedonic Social Individual Collective

designation .satisfaction accreditation; achievement achievement
self-definition
Basis for self- Attainment of Approval of Internal Internalized
evaluation positive others (outer standards goals of
affect audience) (inner reference
audience) group
Individual Public self- Private self-
difference consciousness; consciousness;
measures of need for need for
task approval; high achievement;
orientation self- low self-
monitoring monitoring
Situation Anonymity in Minority status Privacy; Reference group
inducers of group; drug in group; solo exposure to salience;
task intoxication before performance cohesive
orientation audience; replay; mirror; group;
camera; public private failure superordinate
failure goals
Strategies in Norm Conformity; Independence;
service of violation obedience; defiance;
task opinion opinion
moderation; resistance
basking in
reflected glory
berg's (e.g., 1976) six stages of moral judgment. However, the present
analysis's focus on evaluating self-worth differs substantially from Kohl-
berg's focus on evaluating what is right.
SCHEMATIC ASPECTS OF EGO TASKS

The Diffuse Self
The diffuse self has its roots in the body's innate pleasure and pain
responses. This hedonistic facet of the self may appear not to require
development of schematic representations. Nevertheless, as the human
cognitive apparatus develops, its capabilities can be applied to the goals of
the diffuse self as effectively as to any other tasks. Thus, extensive reper-
tories of schematic knowledge can develop in association with the con-
sumption of food, the pursuit of sexual pleasure, the cultivation of bodily
comforts, the use of intoxicants, and so forth.
The Public Self

Development of the public self occurs when others' approval is valued to
the point at which it competes with innate sources of satisfaction. In
minimal form, the public self does not require complex (schematic) cogni-
tive structures. However, the simple strategies that children use to get
approval will not be tolerated indefinitely by parents and others. The child
must begin to understand the ways in which appropriate behavior varies as
a function of situation (Le., situational norms). A sophisticated public self
will use evaluative schemata to anticipate what will be valued by important
others in novel situations, as well as to avoid action that will be ineffective
because powerful others will perceive the actions as ingratiating.
The Private Self

As the public self develops, it begins to encounter conflicts among others'
standards-for example, between the (likely) pro authority criteria of
parents and teachers and the anti authority criteria of peers. The emerg-
ence of the private self may be prompted partly by the difficulty of dealing
with such conflicts. As a schematic knowledge structure, the private self
evaluates behavioral options by assessing their consistency with a set of
cross-situationally applicable values.
The Collective Self

The collective self has its likely origin in identification with the family
group. Religious, racial, ethnic, national, political, corporate, and profes-
sional groups subsequently provide contexts in which the collective facet of
the self can develop further. The collective self emerges when the person
begins to resolve conflicts between group and private goals by allowing the
values and goals of the reference group to take precedence over more per-
sonal goals. The collective facet of the self nevertheless faces potential
conflicts between goals of different reference groups. These conflicts can be
avoided by aligning one's identity firmly with a single group, or by com-
partmentalizing one's life in order to minimize the opportunity for group
goals to come into conflict.
INDIVIDUAL DIFFERENCES IN EGO TASK ORIENTATIONS

The public-private-collective developmental ordering is not an obligatory
one. For example, a collective facet of the self may never develop, or a
collective self might develop without ever having been preceded by a well-
developed private self. (The latter course may be especially likely for a
person whose history of social experience is marked by repeated failures in
winning others' approval [cf. Hoffer, 1951; Maslach, 1974].) The hypothe-
sized facets of the self are not logically or socially incompatible, and can
coexist; development of a new facet does not oblige abandonment of

existing ones. For these reasons, there can be substantial variation across
persons in the relative prominences of the several facets. These variations
are referred to as differences in ego task orientations (Greenwald, 1982).
Measures of ego task orientation for use with adult subjects are presently
under development (Breckler, Greenwald, & Rubinsky, 1986). Breckler
et al. 's measures of ego task orientations are related to measures that
have achieved widespread use in the past decade, in particular, the Self-
Monitoring Scale (Snyder, 1974) and the Self-Consciousness Scale (Fenig-
stein, Scheier, & Buss, 1975). Snyder's Self-Monitoring Scale was original-
ly conceived as a measure of the disposition to attend to the expressive
aspects of one's behavior in social situations. More recently, however
(see Gangestad & Snyder, 1985; Snyder & Campbell, 1982), high self-
monitoring scores have been characterized as indicating a "pragmatic self"
(which corresponds closely to the present public facet of the self), and low
scores as indicating a "principled self" (which corresponds to the present
private facet). The Self-Consciousness Scale produces measures of public
and private self-consciousness, which were interpreted by Fenigstein et al.
(1975) as indicating dispositions to attend, respectively, to publicly observ-
able and privately observable aspects of the self. More recently Scheier and
Carver (1983) have interpreted public and private self-consciousness as
representing "two sides of the self," and Breckler et al. (1986) have estab-
lished that scores on the public and private self-consciousness measures
correlate in the expected fashion with their new measures of public and
private ego task orientations.
Importance of Self-Worth
The most distinctive characteristics of the present social developmental
account of the self are its assertions (a) that the search for self-worth is one
of the strongest motivating forces in adolescent and adult human behavior,
and (b) that differences between persons in their manner of, and effective-
ness in, establishing self-worth are fundamental to personality. Although
these conclusions are ones that have been sometimes resisted in the recent
research literatures of personality and social psychology, they are never-
theless conclusions for which empirical support has become increasingly
strong.
INDICATIONS THAT ESTABLISHING SELF-WORTH IS IMPORTANT

Several recent, prominent lines of research in social psychology have set
out to investigate phenomena that had no apparent initial connection to
self-esteem, but evolved into demonstrations of self-esteem's centrality to
a domain of social behavior. A major case in point is provided by the
history of research on cognitive dissonance theory. Originally interpreted

by Festinger (1957) in terms of motivation to maintain quasi-logical con-
sistency among items of knowledge, "dissonance reduction" phenomena
have subsequently been reinterpreted as involving cognitive changes in the
service of maintaining self-esteem (see, e.g., Aronson, 1968; Greenwald &
Ronis, 1978).
Other socially important phenomena for which research has evolved to-
ward the discovery that self-esteem maintenance was the operative motive
in a domain that was not originally conceived in its terms include: (a) in-
fluence effects due to the social character of the psychological experiment
(reviewed by Weber & Cook, 1972), (b) effects of others' presence on task
performance (Cottrell, 1972; Zajonc, 1965), (c) effects of working in
groups on reduction of work effort ("socialloafing"-Latane, Williams, &
Harkins, 1979; Williams & Jackson, 1986), and (d) social interaction deter-
minants of interpersonal attraction (e.g., Aronson, 1984, Chapter 7).
Two of the most widely known social psychological experiments, Asch's
(1951) conformity experiment and Milgram's (1963) obedience experi-
ment, have a compelling quality (both to subjects who participate in them
and to students of social psychology) that appears to be due to the unusual
way in which they engage the subject's motivation to maintain self-worth.
Asch's conformity experiment puts the subject into conflict between giving
a perceptual judgment that the group (apparently) regards as correct and a
judgment that appears correct according to one's private perception (of the
experimental line-length judging stimuli). Similarly, Milgram's obedience
experiment puts into conflict the subject's desire to do what the experimen-
ter declares to be correct (delivering shocks to the "learner") and what
seems correct according to internalized standards of behavior (i.e., not
causing pain to another). The subject's dilemma, in either experiment, is
that establishing one's self-worth by a public-self strategy (doing what
appears right to others: conforming or obeying) can only be managed at
the expense of losing self-worth in terms of a private-self strategy (doing
what seems wrong to oneself: lying or causing harm), and vice versa.
Still further evidence for the importance of self-esteem motivation
comes from investigations of normal cognitive processes that create and
maintain an inflated self-image. The widespread existence of inflated self-
images was effectively documented by Myers and Ridl (1979). And the
pervasiveness of cognitive biases that build and maintain an inflated sense
of self-worth was demonstrated in a review by Greenwald (1980). In sum-
mary, the evidence from studies of social behavior and social cognition
establish that building and maintaining self-worth are primary social mo-
tives. A corollary conclusion is that individual differences in preferred
strategies for establishing self-worth must playa major role in producing
individual differences in social behavior. The demonstrated success of
measures of self-monitoring, public self-consciousness, and private self-
consciousness in predicting individual differences in social behavior is
consistent with this suggestion (see previous section on "Individual

Differences in Ego Task Orientations").
WHY Is SELF-WORTH So IMPORTANT?

Animal species typically have extensive repertories of genetically provided
self-protective behaviors. Humans' elaborate and flexible neural apparatus
permits the addition of cognitive processes to the self-protective repertory.
At present, there appear to be three mutually supporting interpretations
that indicate why the establishment of self-worth belongs in the category of
cognitive processes that contribute to human adaptive success.
First, high self-esteem is associated with perseverance in problem solv-
ing, which is presumably a product of the high-self-esteem person's
expectation of success (see review by Breckler & Greenwald, 1986). The
high-self-esteem person's perseverance in problem solving is likely to be a
strong adaptive asset in a world in which many problems are solvable as
much through persistent effort as through skill. Second, because a positive
self-image is associated with the conception that one is engaged in effective
courses of action, an affectively positive self-concept is more likely to be
cognitively stable than is a neutral or negative self-concept. Greenwald
(1980), in discussing the adaptive significance of self-inflating cognitive
biases, offered an interpretation that credited their success to their role in
constructing an intrapsychically stable cognitive system. Third, self-esteem
is an attitude that, like other attitudes, can function to guide actions that
maintain and protect the social objects that are critical to adaptive success
(see the review of attitude functions in Pratkanis, Breckler, & Greenwald,
in press). After infancy (during which the social objects most critical to
adaptive success are one's parents), the self is the object that is most essen-
tial to one's adaptive success. Consequently, a positive attitude toward self
has the adaptive value of functioning to protect and foster this critical ob-
ject. (Findings just beginning to become available suggest that positive
self-attitudes are also associated with superior functioning of the body's
immune system.)
Conclusions
This chapter presents a cognitive and social conception of the self's de-
velopment. It focuses on the development of the sense of self-worth, which
is motivationally central to the adult self-concept. It identifies four major
variations in the basis for perceiving self-worth. These four motivational
facets of the self-the diffuse, public, private, and collective selves-
correspond to cognitive/schematic differences in conceiving the relation of
person/self to society. Each of these ego task orientations is ordinarily pre-
sent to some degree in adult behavior. Nevertheless, differences in their
strengths appear to be stable dispositional differences that guide social be-

havior in a great variety of situations. The forces that, in the course of
normal development, lead to variations in the relative strengths of the four
facets remain to be established.
REFERENCES
Aronson, E. (1968). Dissonance theory: Progress and problems. In R.P. Abelson
et al. (Eds.), Theories of cognitive consistency: A sourcebook (pp. 5-27). Chica-
go: Rand McNally.
Aronson, E. (1984). The social animal (4th ed.). New York: Freeman.
Asch, S.E. (1951). Effects of group pressure on the modification and distortion of
judgments. In H. Guetzkow (Ed.), Groups, leadership, and men (pp. 177-190).
Pittsburgh, PA: Carnegie Press.
Bargh, 1.A. (1982). Attention and automaticity in the processing of self-relevant
information. Journal of Personality and Social Psychology, 43, 425-436.
Breckler, S.l., & Greenwald, A.G. (1986). Motivational facets ofthe self. In E.T.
Higgins and R. Sorrentino (Eds.), Handbook of motivation and cognition
(pp. 145-164). New York: Guilford Press.
Breckler, S.l., Greenwald, A.G., & Rubinsky, E.C. (1986, April). Public, private,
and collective self-evaluation: Measurement of individual differences. Paper
presented at conference, "Self and Social Involvement," Princeton, Nl.
Bruner, 1.S. (1966). On cognitive growth. In 1.S. Bruner et aI., Studies in cognitive
growth (pp. 1-67). New York: Wiley.
Cottrell, N.B. (1972). Social facilitation. In e.G. McClintock (Ed.), Experimental
social psychology (pp. 185-236). New York: Holt, Rinehart & Winston.
Fenigstein, A., Scheier, M.F., & Buss, A.H. (1975). Public and private self-
consciousness: Assessment and theory. Journal of Consulting and Clinical
Psychology, 43,522-527.
Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford Uni-
versity Press.
Gallup, G.G., & Suarez, S.D. (1986). Self-awareness and the emergence of mind
in humans and other primates. In 1. Suls & A.G. Greenwald (Eds.), Psycho-
logical perspectives on the self (Vol. 3, pp. 3-26). Hillsdale, Nl: Erlbaum.
Gangestad, S., & Snyder, M. (1985). "To carve nature at its joints": On the exis-
tence of discrete classes in personality. Psychological Review, 92, 317-349.
Gelman, R., & Baillargeon, R. (1983). A review of some Piagetian concepts. In
1.H. Flavell & E.M. Markman (Eds.), Handbook of child psychology (4th ed.,
Vol. 3, pp. 167-230). New York: Wiley.
Greenwald, A.G. (1980). The totalitarian ego: Fabrication and revision of personal
history. American Psychologist, 35, 603-618.
Greenwald, A.G. (1981). Self and memory. In G.H. Bower (Ed.), The psychology
of learning and motivation (Vol. 15, pp. 535-553). New York: Academic Press.
Greenwald, A.G. (1982). Ego task analysis: A synthesis of research on ego-
involvement and self-awareness. In A.H. Hastorf & A.M. Isen (Eds.), Cognitive
social psychology (pp. 109-147). New York: ElsevierlNorth-Holland.
Greenwald, A.G. (1987). Levels of representation. Manuscript submitted for pub-
lication.
Greenwald, A.G., & Breckler, S.J. (1985). To whom is the self presented? In B.R.
Schlenker (Ed.), The self and social life (pp. 126-145). New York: McGraw-Hill.
Greenwald, A.G., & Pratkanis, A.R. (1984). The self. In R.S. Wyer and T.K.
Srull (Eds.), Handbook of social cognition (Vol. 3, pp. 129-178). Hillsdale, NJ:
Erlbaum.
Greenwald, A.G., & Ronis, D.L. (1978). Twenty years of cognitive dissonance:
Case study of the evolution of a theory. Psychological Review, 85,53-57.
Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.
Hoffer, E. (1951). The true believer. New York: Harper.
Inhelder, B., & Piaget, J. (1956). The growth of logical thinking from childhood to
adolescence. London: Routledge and Kegan Paul.
James, W. (1890). Principles of psychology (2 vols.). New York: Holt.
Johnson-Laird, P.N. (1983). Mental models. Cambridge, MA: Harvard University
Press.
Keenan, J.M., & Baillet, S.D. (1980). Memory for personally and socially signif-
icant events. In R.S. Nickerson (Ed.), Attention and performance VIII (pp. 651-
669). Hillsdale, NJ: Erlbaum.
Kihlstrom, J.F., & Cantor, N. (1984). Mental representations of the self. In L.
Berkowitz (Ed.), Advances in experimental social psychology (Vol. 15, pp. 1-
47). New York: Academic Press.
Kohlberg, L. (1976). Moral state and moralization. In T. Lickona (Ed.), Moral
development and behavior (pp. 31-53). New York: Holt, Rinehart & Winston.
Lapsley, D.K., & Quintana, S.M. (1985). Integrative themes in social and develop-
mental theories of self. In J. Pryor and J. Day (Eds.), The development of social
cognition (pp. 153-176). New York: Springer.
Latane, B., Williams, K., & Harkins, S. (1979). Many hands make light the work:
The causes and consequences of social loafing. Journal of Personality and Social
Lecky, P. (1945). Self-consistency: A theory ofpersonality. New York: Island Press.
Lewis, M. (1986). Origins of self-knowledge and individual differences in early
self-recognition. In J. Suls & A.G. Greenwald (Eds.), Psychological perspectives
on the self (Vol. 3, pp. 55-78). Hillsdale, NJ: Erlbaum.
Markus, H. (1977). Self-schemata and processing information about the self. Jour-
nal of Personality and Social Psychology, 35, 63-78.
Maslach, C. (1974). Social and personal bases of individuation. Journal of Per-
sonality and Social Psychology, 29, 411-425.
Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and So-
cial Psychology, 67,371-378.
Minsky, M. (1975). A framework for representing knowledge. In P.H. Winston
(Ed.), The psychology of computer vision (pp. 211-277). New York: McGraw-
Hill.
Myers, D.G., & Ridl, J. (1979, August). Can we all be better than average?
Psychology Today, pp. 85, 95, 96, 98.
Piaget, J. (1954). The construction of reality in the child (Translated by M. Cook).
New York: Basic Books.
Pratkanis, A.R., Breckler, S.J., & Greenwald, A.G. (Eds.) (in press). Attitude
structure and function. Hillsdale, NJ: Erlbaum.
Pratkanis, A.R., & Greenwald, A.G. (1985). How shall the self be conceived?
Journalfor the Theory of Social Behavior, 15,311-330.
Rogers, T.B., Kuiper, N.A., & Kirker, W.S. (1977). Self-reference and the encod-
ing of personal information. Journal of Personality and Social Psychology, 35,
677-688.
Rubin, D.C. (1982). On the retention function for autobiographical memory. Jour-
nal of Verbal Learning and Verbal Behavior, 21,21-38.
Schank, R.C., & Abelson, R.P. (1977). Scripts, plans, goals, and understanding.
Hillsdale, NJ: Erlbaum.
Scheier, M.F., & Carver, C.S. (1983). Two sides of the self: One for you and one
for me. In J. Suls & A.G. Greenwald (Eds.), Psychological perspectives on the
self (Vol. 2, pp. 123-157). Hillsdale, NJ: Erlbaum.
Snyder, M. (1974). Self-monitoring of expressive behavior. Journal of Personality
and Social Psychology, 30, 526-537.
Snyder, M., & Campbell, B. (1982). Self-monitoring: The self in action. In J. Suls
(Eci.), Psychological perspectives on the self (Vol. 1, pp. 185-207). Hillsdale, NJ:
Erlbaum.
Weber, S.J., & Cook, T.D. (1972). Subject effects in laboratory research: An
examination of subject roles, demand characteristics, and valid inference. Psy-
chological Bulletin, 77,273-295.
Williams, K.D., & Jackson, J.D. (1986, May). Social loafing: A review and theo-
retical integration of the literature. Presented at the Midwestern Psychological
Association, Chicago, IL.
Zajonc, R.B. (1965). Social facilitation. Science, 149,269-274.
3
The Construction and
Conservation of the Self:
James and Cooley Revisited
SUSAN HARTER
The 1980s have witnessed a resurgence of interest in the self. Not only have
new models proliferated, but a number of investigators have sought
theoretical guidance from historical scholars of the past. In our own work,
we have drawn upon the insights of William James (1892) and Charles
Horton Cooley (1902). In this chapter we build upon the legacy of James
and Cooley, particularly with regard to an understanding of the determi-
nants of self-regard or self-worth. We adopt a developmental perspective
in that we focus on how the child's sense of overall worth as a person is
constructed. A major goal of our research has been to operationalize the
formulations of both James and Cooley, in order to provide a test of their
appropriateness in accounting for individual differences in self-worth. We
have also sought to examine the issue of whether self-worth is merely an
epiphenomenal construct or whether it plays a role in mediating one's
affect and one's motivational level. Since a detailed description of this
work has appeared elsewhere (see Harter, 1986b, 1987), this chapter mere-
ly summarizes the model and supporting evidence.
Following this description, we examine the extent to which children and
adolescents are able to conserve the self over time as well as across the
various roles that they must adopt. We also discuss the mechanisms
through which children and adolescents attempt to protect and enhance the
self. Finally, this more Western approach to the self is contrasted to a
more Eastern, Buddhist perspective, and the implications of each are
explored.
I -Self Versus Me-Self

The majority of scholars who have devoted thoughtful attention to the self
have come to the conclusion that two conceptually distinct, but experien-
tally intertwined, aspects of the self can be meaningfully identified, the
I-self and the Me-self. From a historical perspective, James (1892) was
perhaps the most articulate on this issue, although Mead (1934) also pur-
44 Susan Harter
sued this theme in his own theorizing. For James, the I-self was the self-
as-subject or knower, in contrast to the Me-self, which represented "an
empirical aggregate of things objectively known" (1892, p. 197). Thus, the
I-self is the processor of information, whereas the Me-self is the product,
namely what is known about one's self.
More recent scholars of the self have echoed this distinction in their own
theorizing (see Dickstein, 1977; Lewis & Brooks-Gunn, 1979). Wylie
(1974, 1979) has summarized the essence of this distinction, contrasting the
self as active agent or the processor with the self as the object of one's
knowledge and evaluation. The I-self, therefore, is the active observer,
whereas the Me-self is the observed, the product of the observing process
when attention is focused on the self.
This chapter, focuses on the role of the I-self as the constructor of the
Me-self, that which is constructed. We also address those mechanisms
through which the I-self attempts to enhance and protect the Me-self.
Finally, we speculate on whether these activities constitute the best use of
the I-self's time and energy.
An examination of the literature reveals that considerable theoretical
and empirical attention has been devoted toward the study of the Me-self,
whereas the I-self has been virtually ignored (see Damon & Hart, 1982;
Harter, 1983). Thus, hundreds of studies have been conducted on the self-
concept, on the self as an object of one's knowledge and evaluation (see
Wylie, 1961, 1974, 1979). Much less is known about the processes through
which the I-self comes to construct the Me-self.
Within this self-concept literature, two approaches can be contrasted,
one in which the self is viewed as a unitary construct, best represented by a
single score, and one in which the self is viewed as multidimensional, best
represented by discrete evaluations across a variety of specific domains
(see Harter, 1985b, 1986b, for a review of these positions). In our own
work we have found both of these approaches wanting. Rather, it appears
more fruitful to consider the Me-self from the perspective of a global judg-
ment about one's worth as a person as well as from a domain-specific pers-
pective in which one examines an individual's profile of self-judgments
across the most relevant arenas of one's life.
Legacy of James and Cooley

The need to consider domain-specific, as well as global, judgments about
the self represents a salient theme for James. In his formal theorizing,
James differentiated a number of specific Me-selves (e.g., the Material Me,
the Social Me, the Spiritual Me). However, he also acknowledged that there
were as many potential Me-selves as there were arenas in one's life that
were critical to one's self-definition. In fact, these domain-specific self-
evaluations constituted the building blocks that formed the edifice of one's
overall sense of self-esteem or regard for oneself as a person.
3. The Construction and Conservation of the Self 45
James specifically postulated that one's global self-esteem represented

the ratio of one's successes to one's pretensions. Thus, it was not sufficient
to aggregate one's self-evaluative judgments over each and every domain
of one's life. Rather, one's global self-esteem derived from one's estima-
tion of one's success in those domains where one aspired to success. Thus,
if one judged that one was successful in these critical domains, a high level
of self-esteem would ensue. Conversely, if one's pretensions vastly ex-
ceeded one's perceived level of success, the outcome would be low self-
esteem. Thus, James's model required that the individual compare his or
her hierarchy of perceived successes to a hierarchy of one's pretensions in
the corresponding domains, in order to arrive at a judgment about one's
overall level of self-regard. The I-self, therefore, must have the ability to
make these comparisons in order to effectively construct the Me-self.
In contrast to James's more cognitive-evaluative model, Cooley (1902)
postulated that the self was a social construction consisting of our percep-
tions of the regard that significant others held toward the self. Mead (1934)
elaborated on this theme in his concept of the "generalized other," which
represented the pooled or collective judgments of the significant others in
one's life. Implicit in these formulations is a modeling process wherein the
individual adopts the attitudes that others hold toward the self. These
reflected appraisals come to define what Cooley metaphorically labeled the
"looking-glass self."
In the formulations of both James and Cooley, one finds philosophical
speculation on the nature of global self-esteem or self-regard in adults. The
goal of our own work has been to examine the applicability of these pro-
cesses in children and adolescents. Toward this end we have put forth a
model of the determinants, as well as mediational role, of self-worth (see
Harter, 1986; Harter & Hogan, 1985). The two determinants, success in
domains deemed important and the reflected appraisals of others, were
derived from James's and Cooley's formulations, respectively. Thus, our
first task was to operationalize these constructs, in order to determine the
relative contribution of each of these potential determinants of self-regard
or what we have termed self-worth.
An Empirical Test of James' Formulation

Self-worth, as we have come to define it, constitutes self-evaluative judg-
ments about one's overall worth as a person. Thus, questionnaire items
asking how much one likes oneself as a person, how satisfied one is with
oneself, and how much one likes the way one is leading one's life, are
administered to tap this construct (see Harter, 1982, 1985b). This defini-
tion is very similar to Rosenberg's (1979) conceptualization of self-esteem.
Thus, self-worth represents the global judgment that we are attempting to
predict in our model, drawing upon both James's and Cooley's formula-
tions.
46 Susan Harter
Our attraction to James's formulation was not merely theoretical in na-

ture. Our own empirical approach to assessing the self-concept in children
yielded profiles that were initially rather puzzling. As mentioned at the
outset, we adopted the perspective that children (above age 8) not only
make domain-specific evaluations of the self, but possess a more global
concept of their self-worth. The specific domains that we have identified to
date include scholastic competence, athletic competence, social accept-
ance, physical appearance, and behavioral conduct. Thus, our self-
Perception Profile for Children (Harter, 1985b) yields a total of six scores,
five domain-specific scores and one global self-worth score, allowing us to
examine a child's profile of self-preceptions.
However, in initially examining large numbers of individual profiles, we
were puzzled by the relationship that the domain-specific scores bore to the
global self-worth score. Consider the profiles of Child A and Child B pre-
sented at the top of Figure 3.1. The domain-specific scoress for these two
children are quite different, particularly with regard to their perceptions of
their scholastic and athletic competencies where the opposite patterns are
observed. Child A feels very competent scholastically but inadequate in
athletics, whereas Child B feels inadequate scholastically but quite compe-
tent athletically. Despite what appear to be two very different domain-
specific profiles, the global self-worth scores for these two children are vir-
tually identical. Thus, we needed some explanation for how two children
could look so different in their domain-specific evaluations, but look so
similar in terms of their overall sense of worth as a person.
We were also perplexed by another type of comparison, represented by
the profiles of Child C and Child D at the bottom of Figure 3.1. Each of
these children displays extremely similar profiles across the five specific
domains. Yet their global self-worth scores are extremely different. Child C
feels very good about his or her overall worth as a person, whereas Child C
has extremely low self-worth. If these children are so similar in their
domain-specific self-perceptions, why is their sense of worth so different?
It was at this point that James's formulation came to our interpretive
rescue. We reasoned that perhaps it was essential to consider the impor-
tance that these domains held for children (James's pretensions), in order
to comprehend the relationship that the specific domain profiles bore to
global self-worth. We inferred, therefore, that Child A must feel that scho-
lastic competence and behavioral conduct were domains in which success
was quite important, and that since this child felt quite adequate in these
domains, his or her sense of self-worth was very positive. By the same logic
we inferred that Child B valued the domains of athletic competence, social
acceptance, and physical appearance, and that since he or she perceived
the self to be quite adequate in these domains, his or her global self-worth
was also very positive. According to this analysis, the importance hierar-
chies of these two children should be very different.
Inferences about the importance hierarchies of Child C and Child D also
4 4
Child A Child B
High
3 3
Medium
2 2
Low
., J:
., ., J:
. - :;::&"
a> a> a> I- a> a> I-
og
.. -
0 a: 0 a:
~~ 0 0 0 0
""
iii iii
a>
"co 00 " 0 :;:;Q)
"
co 00 me" 0
-6 ~;
0- 0-
.. a>
-0.
:!~ .~ ~ .. a>
-0. :;:;[
:!2- -6
.~ .~ m ~
£5
~E .. a> ~E
~5
(/)0
:50
«0
00
00
(/)«
~"
,,0
mo
,.,0.
~c.
Q.«
...J
UJ
(/) (/)0
:50
«0
00
00
(/)«
~"
.,0
mo
,.,0.
~c.
Q.«
...J
UJ
(/)
4 4
High
3 3
Medium
2 2
Low
.,0 J: ., ., .,0 J:
~~" "
a>
. .
I- I-
.,a;"
... -., :;:;[.,"
og
. .,
a> a:
0
iii 0 a: 0
0 iii
0- " 00 "
mf!!
0 :;:;Q)
lil-
c:
00 cat!!
c: 0
-0.
~E :!~ .~ -6 .~[ ~ -
~E
0. 0.
E :!i ";: ::J
IOU .~ m ~
~5
(/)0
:50
«0
00
00
(/)«
~c:
.,0
mo ~c.
Q.«
...J
UJ
(/)
00
(/)0
0
0
00
00
(/)«
~c:
,,0
mo
,.,0.
~c.
Q.«
...J
UJ
(/)
FIGURE 3.1. Profiles of four children showing the relationship between domain-
specific self-evaluations and global self-worth.
provided us an interpretation of why their self-worth scores were so dif-

ferent. We assumed that Child C valued social acceptance and physical
appearance very highly, while discounting the importance of scholastic and
academic success. In this manner, his or her level of self-worth could re-
main quite high. Conversely, we inferred that Child D must place great
importance on scholastic and athletic competence. However, since this
48 Susan Harter
child feels extremely inadequate in these domains, it takes its toll on the
child's overall sense of worth. Thus, a Jamesian analysis provided a very
tempting post hoc explanation for these initially puzzling profiles.
Our next step, therefore, was to examine directly the importance that
children attach to success in these five domains. Thus, we operationalized
the importance of success in each domain by giving children a separate
rating scale in which they were asked to think about how important it was
to do well in each domain in order for them to feel good about themselves
as a p'erson (see Harter, 1985b). Thus, we obtained five competence/
adequacy judgments (from the original Self-Perception Profile), and five
ratings of the importance of success across the corresponding domains on
our new measure, as well as the child's perception of their global self-
worth. All rating scales utilized a 4-point scale.
In order to capture the essence of James's ratio between successes (com-
petence) and pretensions (importance of success), we obtained a discre-
pancy score (competence minus importance) averaged across just those
domains that the child considered to be important to his or her sense of
worth as a person. (Note that for James, one's adequacy in domains
deemed unimportant should not impact one's global self-esteem.) The big-
ger this discrepancy in a negative direction, that is, the more one's impor-
tance scores exceed one's competence or adequacy judgments, the more
negative one's self-worth should be. In contrast, high self-worth should be
associated with discrepancy scores that are close to zero, indicating that
one's hierarchy of perceived competencies and one's hierarchy of impor-
tance ratings are very congruent.
Our initial findings across several studies of children between the ages of
8 and 15 strongly supported James's formulation. There is a systematic
linear relationship between individuals' discrepancy scores and their self-
worth scores. Those at the lowest levels of self-worth have discrepancies in
which their average rated importance of domains is 1.75 scale points higher
than their average competence or adequacy across the five domains (on a
4-point scale). These discrepancies decrease as self-worth increases, such
that among those with the very highest self-worth, the discrepancy between
importance ratings and competence/adequacy judgments is virtually negligi-
ble. Thus, individuals with very low levels of self-worth have the highest
discrepancy scores, reflecting the fact that their importance ratings far ex-
ceed their perceived competence judgments. As self-worth increases, the
discrepancy score systematically decreases, and, as predicted, approaches
zero for those children with the highest level of self-worth. For these high
self-worth children, their importance ratings and competence judgments
are relatively congruent.
Across several studies with children and adolescents, the correlation be-
tween self-worth and the competence/importance discrepancy score has
ranged from .55 to .76. More recently, we have examined this relationship
among college students and adults, based on an administration of the Self-
Perception Profile for College Students (Harter & Neeman, 1986) and the
Self-Perception Profile for Adults (Messer & Harter, 1986). Here we found
virtually the same relationship between self-worth and the discrepancy
score, a finding that we had anticipated, given that James was attempting
to elucidate the determinants of adult self-esteem.
From a more psychological perspective, this relationship reflects the fact
that the low self-worth individual is one who cannot discount the impor-
tance of areas in which he or she is unsuccessful. Conversely, the individual
with high self-worth seems able to discount the importance of those do-
mains in which he or she does not feel adequate. In pursuing this inter-
pretation, we have studied the discounting process even more directly.
We have presented children with vignettes, one for each domain, in
which a story character (the same gender as the subject) feels that success
in that domain is very important, but later comes to learn that he or she is
not very adequate in that domain. An illustrative vignette from the athletic
domain is given below:
Julie and the Soccer Team. When the school year started, Julie decided that
she really wanted to be on the girls' soccer team. It was something that was
IMPORTANT for her to do. She joined the team, but she soon learned that she
was not very good at soccer. She knew she didn't play well and the other kids
and the coach knew it too. No one thought she would get much better, either.
Finally Julie had to make a decision. Which decision to you think she made:
--She decided the soccer was not that important after all and thought about
quitting the team.
--She decided that soccer was still important and that she would stay on the
team, even though she was not that good.
We were particularly interested in the choices of children whom we had
identified as high and low in self-worth. In order to pick the most credible
vignette for each individual subject, we selected the story that corre-
sponded to that subject's least competent domain. The findings (see Harter,
1986b) clearly revealed that the vast majority of high self-worth children
picked the first option in which the importance of the domain is dis-
counted. In contrast, low self-worth children were much more likely to
pick the second option in which one continues to maintain that the domain
is important. Thus, these findings suggested to us that the discounting pro-
cess is one mechanism through which children of high self-worth are able to
maintain positive feelings about the self in general. Unfortunately, the low
self-worth child does not appear to be able to invoke this protective
strategy.
In point of fact, low self-worth children are also among the least com-
petent according to their own perceptions. Given that the domains we have
selected are rather difficult for children in our culture to discount, the large
discrepancy between the competence judgments and the importance rat-
ings of these low self-worth children is perhaps not surprising. Children can
50 Susan Harter
neither discount those domains touted by our society nor escape from
situations in which their lack of competence is evident. As a result, inade-
quacy in domains that they deem as important takes its toll in terms of their
overall feelings of worth. Our more specific analyses at the domain level
reveal that two domains are particularly difficult to discount, namely
behavioral conduct and scholastic competence, domains with very high
normative importance ratings. These are also the two domains that chil-
dren feel their parents think are the most important. Thus, it is exceedingly
difficult for children to discount these domains, even in the face of low
competence. Therefore, although James's formulation contains a logical
analysis of how one might enhance self-esteem by reducing the discrepan-
cy, eiter by lowering one's importance ratings, or by increasing one's com-
petence, in reality, both may be relatively difficult to accomplish for many
low self-worth children.
A Test of Cooley's Formulation

Having demonstrated that James's theory of the determinants of self-worth
was alive and well in the minds of children, we next turned to an empirical
examination of Cooley's looking-glass-self formulation. Since our goal was
to predict global self-worth, we needed to assess the worth, esteem, or
regard that significant others attached to the self. That is, if others feel that
one is a worthwhile person, then one should adopt this attitude toward the
self, and experience high self-worth. Conversely, if others have little re-
gard for the self, then one's self-regard or self-worth should be quite low.
From this perspective, therefore, the I-self must infer the attitudes of
others and then adopt these perceptions in constructing the Me-self.
In operationalizing the regard that others hold for the self, we designed
an instrument that tapped the perceived regard of four potential significant
others: parents, classmates, close friends, and teachers. Items tapped the
extent to which these significant others treated them like a person who
mattered, liked them the way they are, cared about their feelings, and so
forth. (Harter, 1985c).
The findings clearly revealed a strong correlation between the combined-
regard-from-others score and self-worth. In several studies of children ages
8 to 15, the correlations range from .45 to .60. These relationships are quite
robust across this entire age range. In addition, we have been interested in
which specific sources of support are most critical or predictive in answer-
ing the question, "Mirror, mirror on the wall, whose opinion is the most
important of all?" Interestingly, the parental regard score is most highly
predictive of self-worth across this entire age range. It is followed closely
by classmate support. Close friend and teacher regard are less predictive of
self-worth, although these correlations (rs between .27 and .38) are sig-
nificant.
Thus, it would appear that Cooley's model of self-worth is also appli-

cable to children, in that the higher the level of regard from significant
others, the higher the level of overall regard for the self. Moreover, as
Rosenberg (1979) has cogently argued, not everyone's opinion is equally
critical. Although there are individual differences in the patterns governing
just which significant other has the most impact, the group findings are
interesting in highlighting the importance of parental regard to self-worth.
Here we concur with Damon (1983), who have noted that parental in-
fluence does not wane during adolescence to the extent that others have
claimed. While we are not denying the growing importance of the peer
culture as one moves into later childhood and adolescence, our findings
suggest that parents continue to remain important as a source of self-worth.
(See Harter, 1987 for a further discussion of this issue.)
The Larger Theoretical Model

In our previous work, we have been interested not only in the determinants
of self-worth, but in the extent to which self-worth influences other con-
structs within the self-system (i.e., affect and motivation). Since these
theoretical and empirical efforts have been presented elsewhere (see
Harter, 1986b), they are only briefly described here. The larger model
postulates two determinants of self-worth: the competence/importance
discrepancy construct (derived from James) and the positive regard of sig-
nificant others (derived from Cooley). Evidence for these two deter-
minants has been extended in the present chapter.
The model further postulates that self-worth, in turn, mediates one's
general affect along a dimension of cheerful to depressed. Affect, in turn,
serves as the primary mediating link between self-worth and motivation,
defined as the energy one has available to engage in age-appropriate activi-
ties. The choice of these constructs, affect and motivation, was partially
guided by recent sequential models demonstrating that self-judgments
elicit an affective reaction that, in turn, motivates the individual to engage
in particular behaviors (e.g., Bandura, 1978; Harter & Connell, 1984; Kan-
fer, 1980; Wicklund, 1975). Our interest in affect and motivation was also
sparked by our efforts to identify theoretically derived components that
have been implicated in the study of depression, and to devise a self-report
instrument to assess these components. Both the affect and motivation
constructs, therefore, were assessed by administering subscales from our
Dimensions of Childhood Depression Profile (Harter & Nowakowski,
1987). The affect scale provides a general index of perceived sadness or
depressed mood versus happiness or cheerfulness. The motivation subscale
taps the energy, interest, and desire that children have to engage in age-
appropriate activities. Judgments of affect and motivation, therefore, were
not domain specific, but required subjects to provide a general assessment of
52 Susan Harter
both constructs, just as they were requested, on the self-worth subscale, to

make a global judgment about their worth as a person.
We have postulated, therefore, that self-worth is not an epiphenomenon
with no particular function, but that it has a major mediational role impact-
ing affect, such that the child with high self-worth will be relatively cheer-
ful, whereas the low self-worth child will be relatively depressed. Affect, in
turn, will have an impact on motivation, such that the cheerful child will be
much more likely to have energy to engage in age-appropriate activities,
whereas the depressed child will have little energy to devote to such activi-
ties.
Employing path-analytic techniques (Harter & Hogan, 1985), we have
obtained strong support for this model, particularly in comparison to
alternative models involving these same constructs. For example, we have
compared this model to one in which the mediational role of self-worth has
been removed (Le., a model in which self-worth is considered to be epiphe-
nomenal). The latter model provided a very poor fit to the data. Thus, it
would appear that self-worth does have an important function in mediating
both the affective and motivational systems, a finding consistent with a
variety of sequential models in which self-judgments elicit an affective
reaction that in turn motivates the individual to engage in a particular
behavior (see Bandura, 1978; Harter & Connell, 1984; Kanfer, 1980;
Wicklund, 1975).
Interestingly, in keeping with the homage due both James and Cooley,
these historical scholars also spoke to the function of self-esteem. James
(1892) was perhaps the most explicit in suggesting that self-esteem pro-
vokes an emotional reaction, which in turn has implications for how we
behave. Cooley devoted more attention to the affective component, iden-
tifying the variety of self-feelings that could be provoked by the regard in
which we hold the self (e.g., pride, self-aggrandizement, shame, humilia-
tion, mortification). As Cooley observed (1902), self-feelings represent a
unique subset of emotions in that they are experienced by, as well as
directed toward, the self. In our I-self, Me-self parlance, it is the I-self
registering a feeling about the Me-self. These self-affects can be contrasted
to emotions that are directed externally, toward events or the actions of
others (e.g., frustration over being grounded or anxiety over a bad grade).
A fruitful avenue for future study will be the motivational properties of
these two types of affects, an issue that we have recently begun to address
in our own work (Harter, in press a).
Conservation of the Self

We have just described certain processes through which the self is con-
structed, namely how the I-self compares competence and importance
hierarchies, as well as assesses the opinions of significant others, in arriving
at a global judgment of the Me-self. We now turn to a consideration of how

and whether the I-self conserves the Me-self.
The conservation metaphor has been selected quite purposely. Piaget
invoked this term primarily to describe what he observed to be physical
conservation. That is, when one performs a physical transformation, such
as pouring water from a short, wide beaker to a tall, thin beaker, does the
child appreciate the fact that the amount of water remains the same, that
is, is conserved?
Here, we are referring to psychological conservation. That is, we all
experience certain transformations (certain perturbations to use Piaget's
terminology) that seemingly catapult us from the short, wide beakers of
life to the tall, thin beakers. In the face of such transformations, to what
extent is the self conserved in that our sense of self remains constant, our
identity unshaken? What happens, for example, when our manuscript is
rejected, when our grant isn't funded, or when we don't get the job? We
prefer the term conservation of self to the commonly used construct of sta-
bility, since it highlights the individua1's response to events or potential
transformations that may, or may not, impact the self. The allusion to the
conservation metaphor is all the more powerful to the extent that the
effects of these events may not be reversible.
Interestingly, both James and Cooley addressed this very issue in their
comments on the desirability of possessing a balanced sense of self that
is not affected by the vicissitudes of daily life. In the words of James,
" ... there is a certain average tone of self-feeling which each one of us
carries about with him, and which is independent of the objective reasons
we may have for satisfaction or discontent" (1892, p. 171). Cooley voiced a
similar sentiment in observing that the person with balanced self-respect
has ". . . stable ways of thinking about the image of self that cannot be
upset by passing phases of praise or blame" (1902, p. 201).
It should be noted, however, that both James and Cooley were making
reference to the adult self. From a developmental perspective, however,
one may question whether a balanced sense of self is potentially possible
for children or adolescents whose life is characterized by change. Para-
doxically, perhaps, the very determinants of one's global self that were
identified by James and Cooley, namely the relationship between one's
successes and pretensions, as well as the opinions of significant others, can
readily be cast into mechanisms of change. For example, events or trans-
formations that alter either one's perceived level of success or the impor-
tance that an individual attaches to a given domain have the potential for
altering one's overall sense of self-esteem, from a Jamesian perspective.
Analogously, events that alter the level of regard that significant others
manifest toward the self should also have a direct impact on one's global
sense of self, from Cooley's looking-glass-self perspective. In the face
of either type of change, one would anticipate that the child or adolescent
would not be able to conserve the self. We would anticipate that this
54 Susan Harter
process is not readily reversible by the individual unless the determinants

themselves are altered accordingly.
In our own work, we have addressed this issue by examining the potential
effect of one particular educational transformation, namely the transition
from elementary school to junior high school. We have followed students
longitudinally, assessing their self-worth in the spring of their sixth-grade
elementary school year and then again in the fall of their seventh-grade
year in junior high school. The correlation between self-worth at the two
time periods spanning this transition is approximately .5, revealing that a
substantial number of students alter their perceptions of their worth.
In pursuing this question, therefore, we have identified three groups of
children: those who manifest decreases in self-worth, those whose self-worth
increases, and those for whom there is no significant change. To give some
indication of the magnitude of change, those decreasing in self-worth had
scores of 2.75 in the spring compared to 2.20 in the fall, a sizable drop.
Those increasing in self-worth went from scores of 2.80 in the spring to 3.52
in the fall, an even greater change. Those manifesting no change had scores
of 2.90 in the spring and 3.00 in the fall, a nonsignificant difference.
From a Jamesian perspective, one would predict that changes in self-
worth should be directly related to changes in the competence/importance
discrepancy score. More specifically, if the transition to junior high pro-
vokes changes in either one's competence hierarchy or one's importance
hierarchy, self-worth should change correspondingly.
The findings (initially reported in Harter, 1986b) clearly support this
prediction. Those children reporting decreases in self-worth showed a cor-
responding increase in their discrepancy score, revealing a greater gap be-
tween their competencies and the importance of success. Those children
reporting increases in their self-worth showed a corresponding decrease in
their discrepancy score, reflecting the fact that their competence hierarchy
and their importance hierarchy were becoming more congruent. Children
reporting no changes in self-worth manifested no change in their discrepan-
cy score.
Thus, it would appear that the transition to junior high school does, for
certain children, cause them to reevaluate their competencies as well as the
value that they attach to success in the domains that we have tapped. The
resulting discrepancy or congruence between these hierarchies, in turn,
would appear to impact their self-worth. Thus, the determinants of self-
worth that were derived from James's theory can clearly be cast into
mechanisms of change, namely factors that make it difficult for some chil-
dren to conserve their sense of self-worth. Since we have only looked at the
relatively immediate impact of educational transitions, we do not yet know
the extent to which such processes may be reversible. That is, it may well
be that those who experience lowered self-worth may, ultimately, be able
to discount the importance of domains in which they feel incompetent in
the new setting or, conversely, attempt to increase their competence in
order to reduce the importance/competence discrepancy. Longitudinal

studies would be valuable here in order to better understand the dynamic
nature of these processes and the potential for further change.
In our study of children's transition to seventh grade, we had not yet
considered the impact of positive regard on self-worth and, therefore,
could not test the implication of Cooley's formulation with this sample.
However, in our subsequent studies of children within a middle school
(sixth, seventh, and eighth grades), we have identified children whose self-
worth has either increased, decreased, or remained the same over a year's
period in which these children did not change schools. Very interestingly,
these self-worth changes were systematically related to changes in the posi-
tiveregard score, particularly the perceived positive regard from parents.
We do not know, of course, whether changes in one's perceptions of the
regard that one receives from others involves actual changes in the level of
support from these individuals, or whether the children merely perceive
there to be a difference. Nevertheless, the findings suggest that the
looking-glass-self model also provides an analysis of the mechanisms that
can lead to alterations in one's level of self-worth.
SITUATIONAL CONSERVATION OF THE SELF
The preceding discussion dealt with the issue of how self-worth is, or is not,
conserved over time. We now turn to the question of how and whether the
self is conserved across roles as well as situations. In previous treatments of
the self, this issue has been discussed from the standpoint of the consistency
of the self. Numerous scholars of the self have underscored the need for
the individual to possess an integrated or unified self (Allport, 1955, 1961;
Horney, 1950; Jung, 1928; Lecky, 1945; Maslow, 1961; Rogers, 1950). A
number of theorists (e.g., Brim, 1976; Epstein, 1973; Kelley, 1955) have
emphasized that the individual organizes his or her self-attributes or post-
ulates about the self into a theory. Much like a scientific theory, this orga-
nization is threatened by evidence that is inconsistent with the portrait that
one has constructed of the self. Epstein (1981) has recently formalized
many of these observations under the rubric of the unity principle, emph-
asizing that one of the most basic needs of the individual is to maintain the
unity and coherence of the conceptual system that defines the self.
Other theorists, however, have focused on the multiple selves that reside
within the person, emphasizing that the most fruitful approach to the self is
to consider the multiple roles that the individual must play in different
contexts (Baldwin, 1897; Gergen, 1968; Jourard, 1971). Gergen contends
that the notion of the unified, consistent self is probably ill-conceived, and
that people invariably adjust their behavior in accord with the specific
nature of the relationship or situational context in which they find them-
selves.
This issue has also been recast into the controversy over whether be-
56 Susan Harter
havior is best conceived of as trait-like in nature or situation specific. The

most recent literature has emphasized the situational differences in
people's behavior and self-perceptions (see Gergen, 1971; Mischel, 1969,
1973; Vallacher, 1980). From this point of view, the association of different
selves with different roles or contexts represents differentiation rather than
inconsistency.
Within the developmental literature, two age-related shifts in one's
conception of the self are noteworthy in this regard. With increasing de-
velopment, self-descriptions shift from concrete, observable or behavioral
attributes (e.g., I am short, have brown hair, live on Blake street, have
two dogs, and am a very fast runner) to trait labels or more conceptually
abstract descriptions of the self (see Harter, 1983; Rosenberg, 1986, for a
summary of these findings). Thus, in middle childhood and beyond, the
postulates of the self-theory might include such descriptions as smart,
friendly, trustworthy, cheerful, popular, and athletic.
The second trend is for the self to become less global or more differenti-
ated as well as complex (see also Harter, 1983; Rosenberg, 1986). Thus,
the number of postulates in the self-system increases with development, as
does the penchant for organizing the varying attributes that one possesses
or displays across the various contexts in one's life.
On the surface, these two trends may appear incompatible, since the
possession of traits is typically viewed as antithetical to a more differenti-
ated, situation-specific view of the self. To clarify this apparent paradox,
it is critical to specify the meaning of the term trait. From a cognitive-
developmental perspective, the shift to trait labels represents a conceptual
advance in that the older child becomes capable of organizing his or her
observable, behavioral attributes into a higher order generalization about
the self. Thus, one's prowess at soccer, tennis, and gymnastics can be com-
bined into the generalized trait of athletic. One's mastery of a number of
school subjects (e.g., science, social studies, language arts) may lead one to
conclude that one is smart. One's ability to listen, to offer assistance, and to
share one's possessions with one's peers may lead one to see the self as
friendly.
The ability to conceptualize the self in terms of trait labels, however,
does not necessarily imply that these attributes are consistent or stable
across situations or time. That is, we need to make the distinction between
a trait label as a higher order generalization about the self and the use of
the term trait to connote stability or consistency. The need for this distinc-
tion becomes apparent when we consider the fact that older children and
adolescents, as well as adults, can perceive the self to be friendly or smart
or athletic in some contexts but not in others. That is, the conceptual
ability to construct trait labels does not necessarily connote self-attributes,
it does not imply that the self is conserved across the various contexts of
one's life.
We have addressed this issue empirically, from a number of perspec-

tives. In one study, we focused on children's self-perceptions of the trait
labels smart and dumb (Harter & Haltiwanger, 1986). Sixth- and seventh-
grade children were presented with an array of seven "pie-like" circles
portraying different combinations of smart and dumb. At the extremes, the
circles were labeled all smart and all dumb. Various combinations of smart
and dumb were depicted, including a circle in the center that represented
half-smart and half-dumb. Thus, children could describe themselves along
a continuum ranging from all-smart at one extreme to all-dumb at the other
extreme, with five combinations of smart and dumb in between.
We then asked the children to think about how smart and dumb they
really are and to check how many combinations they have experienced
within just the past several weeks. We emphasized that they could check as
few or as many as they wished, ranging from one circle to all seven. The
findings of this initial choice revealed that on the average, children selected
approximately three combinations of smart and dumb, revealing that their
level of smartness or dumbness was not conserved over time or situation.
Later, in a questionnaire, we asked these same children to respond to a
forced-choice question asking them whether they felt that they were exact-
ly the same combination of smart and dumb all the time or whether they
felt that they changed in that sometimes they were one combination of
smart and dumb, whereas at other times they were a different combination.
The responses revealed that 77% of the children selected the second op-
tion, indicating that they felt that at different times they perceived them-
selves to possess different combinations of smart and dumb. Here again,
the findings indicated that these self-attributes were not conserved.
From a psychological perspective, perhaps the more critical question is
whether children are concerned or bothered by these fluctuations within
the self. Thus, in a separate question, we asked just those 77% of the
children who acknowledged such fluctuations to indicate whether or not
they were bothered by such changes. The results indicated that 33% of
these sixth and seventh graders were bothered, while the remaining 67%
were not. Interestingly, the mean number of combinations experienced by
those who were bothered was 3.9 compared to a mean of 2.7 for those who
were not bothered. Thus, it would appear that greater fluctuation is associ-
ated with one's concern over such changes.
The children were also asked to generate reasons for why they felt they
changed. That is, they were asked, "What happens to make how smart and
dumb you are change"? A content analysis of these reasons revealed two
general categories, those that could be characterized as more internally
controlled by the individual and those that involved more external control,
typically by others. Examples of the more internal reasons were: "I know I
have to do something to change," "It depends on whether I pay attention
or not," "You can straighten up (and get smarter) and study or you can
58 Susan Harter
just relax and mess around (and get dumber)," "You can get confidence
and work hard to change from dumb to smart or you can do the opposite to
change from smart to dumb."
Examples of the external reasons included: "Your friends and your
parents make you change because they can make you feel domb some of
the time and smart some of the time," "Sometimes my friends and the way
they act change me," "The teachers make me change," "What makes me
change is being around people, when I'm around other people my per-
sonality changes."
When we examined the reasons for those who were bothered by their
changing smartness or dumbness and those who were not bothered, we
discovered an interesting difference: Those who were bothered by these
changes were more likely to give external reasons for change, seeing these
changes as due to the effects of other people, whereas those who were not
bothered were more likely to feel that they were responsible for these
changes, that is, they were under more internal control. Thus, it would
appear that not only is one's level of smartness not conserved over time
and situation for the majority of children at this age level, but that one's
concern over such changes may be directly related to who is primarily re-
sponsible for these shifts, the self or others.
Further research will be necessary to determine just how apt the con-
servation may be to these processes. For example, if changes in smartness
are perceived to be under one's control, such that one can reverse these
perceptions through one's actions, and one feels that one's "true" level of
smartness has not been altered, then such changes may not represent lack
of conservation, as Piaget conceived it. Thus, we need to probe further into
this issue and its implications for the possibility that more enduring charac-
teristics represent the backdrop against which situational alterations are
brought about.
COGNITIVE-DEVELOPMENTAL FACTORS DURING ADOLESCENCE
We have seen how the inability to conserve the self can be due to environ-
mental factors such as educational transitions, to the effects of other
people, and to one's own role in producing change. In addition, there are
certain cognitive-developmental factors that also conspire with socializa-
tion pressures to make it difficult to conserve the self, particularly during
the period of adolescence. With regard to the demands of socialization, the
period of adolescence is characterized by the need to forge an identity that
is necessarily based on the need to differentiate the self vis-a-vis the roles
one must play with the different significant others in one's life (Erikson,
1968). Thus, one comes to develop somewhat different selves in relation to
one's parents, one's friends, one's teachers, one's romantic interests, one's
occupational ambitions, and so forth.
This need to construct multiple selves was acknowledged by James
(1892) who cogently observed the potential contradictions that this might
pose, a phenomenon that James labeled the "conflict of the different
Me's." As an example he describes how "Many a youth who is demure
enough before his parents and teachers swaggers like a pirate among his
tough young friends." The developmental task of this period, however,
is to consolidate somehow these multiple selves in order to construct an
integrated identity.
From a cognitive-developmental perspective, there are conceptual ad-
vances that should potentially aid in this process. However, these advances
also make salient the contradictory elements within one's self-portrait,
underscoring the reality that the self is not conserved across roles. With the
advent of formal operations, the cognitive apparatus is now capable of
constructing a more comprehensive, more formal, theory of oneself. This
theory, as Epstein (1973) has cogently argued, must meet the criteria for
any formal theory, (e.g., parsimony, usefulness, testability, and internal
consistency) .
Thus, the very ability to create such a self-theory during adolescence
represents both an asset as well as a potential liability , since the conceptual
tools that allow one to consider and compare one's role-related attributes
simultaneously also allow one to detect inconsistencies across roles. To the
extent that the cognitive apparatus is capable and primed to construct a
theory that is internally consistent, this lack of conservation across should
become psychologically troublesome.
In a series of studies from our own laboratory, we have documented that
this is clearly the case (Carson, 1985; Harter, 1986a; Harter & Bresnick,
1987; Monsour, 1985). Our procedure involves asking subjects to first gen-
erate lists of self-attributes by roles, after which they transfer these self-
descriptors to a circular diagram, organizing them according to whether
they are important and central to the self (in which case they are placed
in a center ring), less important (next ring), or relatively unimportant
(peripheral ring). When they have organized their self-portrait in this man-
ner, we ask them to indicate any pairs of self-attributes that appear to be
opposites or contradictory. After identifying these opposing attributes, we
next ask subjects to indicate whether any of these opposites appear to be
clashing, are at war with one another, are struggling within their personal-
ity, or are in conflict.
Our adolescent subjects not only identify a number of contradictory
attributes but many are extremely bothered by the fact that they appear to
manifest one attribute in one role (e.g., cheerful with friends) and then
display the opposite characteristic in a different role (e.g., depressed with
family). As one IS-year old put it,
I really think of myself as a cheerful person and I usually am with my friends, but
then it really bugs me, how can I be so depressed with my parents when I am really
a cheerful person, it doesn't make sense. I contemplate on this a lot but I can't
really resolve it.
60 Susan Harter
Our findings reveal that this type of concern and preoccupation increases
dramatically from early to middle adolescence. In interpreting this shift,
we have relied on Fischer's (1980) cognitive-developmental theory, which
provides a more differentiated analysis of the substages of formal opera-
tions than was provided by Piaget. Although young adolescents are cap-
able of constructing "single abstractions" about the self (e.g., describing
the self as considerate, liberal, self-conscious, introspective, obnoxious,
depressed, understanding), they cannot yet relate each of these abstrac-
tions about the self to one another; they are not yet capable of what Fischer
labels "abstract mappings," in which one self-attribute is compared to, or
mapped onto, another. As a result, they tend not to detect, or be con-
cerned over, opposing self-attributes.
The cognitive skills necessary to compare such abstractions about the
self appear in middle adolescence, about the age of 15. With the advent of
the ability to relate one's attributes to each other, one can now evaluate the
postulates of one's self-theory from the standpoint of whether they are
internally consistent. Thus, the adolescent can now detect the fact that
there are opposites within one's self-portrait, opposites that are particular-
ly evident across the various roles that one has come to adopt, according to
our findings. These opposites are not only detected, but are quite con-
flictual beginning in middle adolescence (age 15 to 16), in contrast to early
adolescence (ages 12 to 13) where conflict is negligible.
From a cognitive-developmental perspective (Fischer, 1980), conflict
should diminish as individuals move into later adolescence, a period in
which they become capable of a new cognitive level of abstract systems.
That is, they develop the ability to coordinate or integrate these abstrac-
tions into meaningful, noncontradictory, higher order abstractions about
the self. Thus, one can integrate one's cheerful and depressive attributes
into the higher order abstraction of "moody."
Through such cognitive advances, therefore, previous contradictions be-
tween opposing attributes of the self can be resolved, particularly those
occurring across roles. For example, one could assert that one is flexible,
adaptive, sensitive to the needs of others, open, and appropriate, thereby
rendering numerous potentially opposing characteristics compatible, since
they are now subsumed under higher order abstractions about the self.
There is a related strategy that a number of our older adolescents em-
ploy that also seems to reduce the phenomenological conflict experienced
by those in middle adolescence. Many individuals adopt a more philo-
sophical stance toward the desirability and normalcy of behaving different-
ly in different roles. Thus, they will indicate that "it wouldn't be normal to
be the same way with your girlfriend as with your mother" (a conclusion
that appears reasonable to all but the most ardent psychoanalysts!). Other
examples include: "You act one way with your friends, and a different way
when you are in class, that's the way you should be," and "Its good to be
able to be different with different people in your life, you'd be pretty
strange, and also pretty boring, if you weren't."
The foregoing analysis places considerable emphasis on how cognitive-

developmental factors, in the face of the demands of socialization, during
the period of adolescence may impact the degree to which one can con-
serve the self. In this arena, as we have suggested in the application of the
conservation-of-self metaphor, we need to investigate further the extent
to which one feels one has truly changed, and therefore cannot readily
reverse these processes, or whether we are dealing with more manifest
change in one's behavior. The answer to this question may well be perplex-
ing to the adolescent, since we have suggested that development during
adolescence requires the construction of multiple selves in relation to the
role differentiation that is required (at least in our society), and that cogni-
tive advances that emerge during middle adolescence provoke one to de-
tect role inconsistencies that cause phenomenological contlict and distress.
Indeed, as other writers of studies on adolescence have cogently pointed
out, adolescence ushers in the search for the true self, an identity that is yet
to be consolidated. In the absence of such an identity, seemingly role-
related, ephemeral changes in the self, may well take on greater signif-
icance in the absence of a core, stable, sense of self that is conserved
across these role-related transformations.
This type of cognitive-developmental analysis can be meaningfully
supplemented by a thoughtful consideration of entire social context in
which the adolescent is developing. Rosenberg (1986) has provided such
a framework in his discussion of what he terms the "barometric self-
concept," namely the day-to-day and sometimes moment-to-moment
tluctuations in self-attitudes that can be observed, particularly during the
period of adolescence. Rosenberg reports on a number of studies revealing
that the self-concept is particularly barometric or volatile during adoles-
cence. He cites a number of reasons that go beyond our own analysis,
providing a more comprehensive picture of the nature of the self-concept
during this period. Four specific reasons for the volatility of the adolescent
self-concept are suggested.
First, he observes that during adolescence, there is tremendous concern
with the impression that one is making on others and relatedly with what
others think of oneself. Given that adolescents depend so heavily on what
others think, the difficulty of divining their impressions leads to ambiguity
about others' attitudes toward the self, which in tum accentuates one's
uncertainty about what one is really like.
In addition, self-concept volatility is also fostered by the fact that differ-
ent people, in different roles or contexts, will have different impressions of
the self. Given that adolescents are so heavily dependent on the retlected
appraisals of others, these differing perspectives that others have of the self
create contradictory feedback that must necessarily result in uncertainty
about the self.
As a third, related consideration, Rosenberg points out that the adoles-
cent is preoccupied with self-presentation or impression management. As
part of this process, one may tentatively adopt, and as readily abandon, a
62 Susan Harter
variety of roles. To the extent that one observes oneself enacting these
varying, and often contradictory, roles, one comes to experience the self as
highly mutable.
Finally, Rosenberg draws upon the ambiguous status of adolescence
within our American society. Since there are no clear age markers as to
when this period begins or ends, one may be treated more like a child by
some, more like an adult by others, and like someone of uncertain status
by yet other people with whom one interacts. Given these fluctuations, in
the face of the adolescent's dependence upon the opinions of others toward
the self, one's own appraisal of the self will undoubtedly vary.
The concern over contradictions within the self during the period of
adolescence, leading to what we have termed lack of conservation of self,
would not merely appear to be a problem within contemporary American
society. As we noted earlier, James introduced this theme at the turn of the
century, as did one of his colleagues, Mary Calkins. Calkins (1930, p. 216)
contended that the everyday self was experienced as a hierarchy of many
partial selves (e.g., as one who is capable of both reason and impulsive-
ness; as one who is both conscientious and reckless). Calkins wrestled with
the issue of whether these seemingly conflicting selves were in actuality
combined into an articulated whole, whether the human "Me" could come
to represent "One Self" that somehow integrated the many selves that one
experienced. She did this issue poetic justice by drawing upon a historical
verse that poignantly punctuated the underlying dilemma:
Within my earthly temple there's a crowd,
There's one of us that's humble, one that's proud;
There's one in eager search for earthly pelf,
And one who loves his neighbour as himself.
There's one who's broken-hearted for his sins,
And one who, unreprentant, sits and grins.
From much corroding care I should be free
If once I could determine which is me.
Protection and Enhancement of the Self

We have shown how the I-self constructs the Me-self and how although the
I-self may also attempt to conserve the Me-self, there may be environmen-
tal, social, personal, and cognitive-developmental factors that mitigate
against such conservation over time or across situations. We now turn to
another preoccupation of the I-self, namely its attempts to protect and
enhance the Me-self. There are four interrelated processes that our own
research has revealed in children and adolescents.
The first process, discounting, is discussed earlier. Our findings have in-
dicated that in order to maintain one's overall sense of worth, one attempts
to discount the importance of domains in which one is not competent. As
James himself noted, there are two strategies for equalizing the balance
between one's successes and one's pretensions: one can either increase
one's level of success or alter one's pretensions, namely lower one's aspira-
tions. (This latter strategy would also appear to be part of contemporary
college lore as revealed in a poster that I recently observed on a hall bulle-
tin board at the University of Notre Dame: "Mom always says, if all else
fails, lower your standards!")
A related strategy can be observed in the tendency for children and
adolescents to take more credit for their successes than their failures, a
process that Greenwald (1980) has labeled "beneffectance." Greenwald
amassed a considerable array of findings demonstrating how this type of
enhancement strategy operates in adults. In our own work with children
(see Harter, 1985a), we have operationalized this strategy by comparing
the credit that one takes for one's academic successes to the responsibility
that one takes for one's scholastic failures. The evidence reveals that the
large majority of children take more responsibility for their successes than
their failures, and that this tendency increases over the age range that we
have studied to date, Grades 3 through 9. Thus, it would appear that with
development, the I-self becomes more adept at protecting and enhancing
the Me-self through this type of strategy.
A similar self-enhancement mechanism has been revealed in our work
on the adolescent self-portrait. In the work described in the preceding sec-
tion, we have utilized a procedure in which adolescents organize their self-
attributes in such a way that they designate which characteristics represent
the most important, core attributes of the self, in contrast to those that are
less important, and those that are the least important. An examination of
these patterns of organization reveals a very dramatic effect in that the
vast majority of positive attributes (e.g., caring, understanding, friendly,
nice, smart, curious, outgoing) are placed at the core of the self, as the
most important attributes. Conversely, one's negative attributes (e.g.,
frustrated, inconsiderate, lazy, nervous, shy, embarrassed, confused, de-
pressed) are relegated to the periphery of the self reflecting the fact that
they are judged to be the least important aspects of one's personality.
Fully 94% of one's core attributes were positive, and 75% of one's least
important attributes were negative. Thus, one's positive and negative
attributes would appear to be filtered through a protective lens that ac-
centuates the positive while deemphasizing the negative.
There is a further elaboration of this strategy, a fourth mechanism, that
our most recent research has revealed (Harter & Bresnick, 1987). This
strategy emerged as a result of our efforts to determine whether the self-
attributes that adolescents generated were viewed by them as traits,
defined as enduring characteristics of the person, or whether these self-
descriptors were considered to be behaviors that one occasionally displays.
Consistent with our previous discussion of the need to be precise in our use
of the term trait label, we did not want to assume that these higher order
64 Susan Harter
generalizations about the self necessarily represented relatively enduring

characteristics of the person over time or situation.
To address this question empirically, we employed a very simple proce-
dure. After each subject had generated his or her role-related list of attri-
butes, we asked that he or she placed that attribute in one of two positions
in the following sentence: "Would you say that you were a __ person
or that you act __ now and then? The pattern of findings was quite clear.
Seventy-five percent of all positive self-descriptors were seen as attributes
of the person, rather than mere behaviors. In contrast, among the negative
self-descriptors, 79% were identified as behaviors, ways that one acts occa-
sionally, rather than attributes of oneself as a person.
Thus, not only do individuals attempt to protect and enhance the self by
viewing their positive attributes as central to the self, but in addition they
see them as characteristics of the self as a person, whereas negative charac-
teristics are relegated to the periphery and are viewed merely as behaviors
in which one occasionally engages. These processes have much in common
with those identified in our earlier research where we found that children
and adolescents attempt to discount those domains in which they are not
successful in order to maintain their sense of self-worth. This pattern also
converges with our evidence indicating that children take more respon-
sibility for their successes than their failures. While there are individual
differences in the extent to which children and adolescents can effectively
utilize these mechanisms, they do seem to represent a package of strategies
through which the I -self attempts to protect and enhance the Me-self
(Harter & Robinson, 1987).
Is It Really Worth the Energy?

We have seen that there are several potentially powerful mechanisms
through which the I-self attempts to protect and enhance the Me-self, pro-
cesses that on the surface would appear to be extremely adaptive and
therefore quite desirable. However, let us for a moment relax our
ethnocentric grip on the mechanisms touted by Western psychologists, and
ask whether all of this effort is really necessary? Is it worth the energy? Is
this really how the I-self should be spending its time, devoted to the pro-
tection and enhancement of the Me-self? Should this be its primary job
description? Aren't there better things for the I-self to be doing? Appre-
ciating art? Enjoying nature? Practicing the inner game of tennis? Openly
experiencing other people?
The perspective that I have developed thus far represents a very Western
view of the self. Westerners appear to be preoccupied with the self. Wit-
ness the number of professional publications on the topic, not to mention
the vast array of books and articles in the popular press that are devoted to
enhancing one's self-concept. A judicious selection from this menu is
viewed as the contemporary route to psychological health.
However, this is by no means a universal window on self-processes. In

fact, it departs dramatically from those perspectives that have grown out of
the Eastern, Buddhist tradition. Consider the following observations from
a Buddhist scholar (Tulku, 1978) who would appear to begin with a very
Cooley-like perspective, although the implications are decidedly different.
Tulku observes:
Each of us has a self-image that is based on who we think we are and how we think
others see us. When we look in a mirror, we know that what we see there is only a
reflection; even though our self-image has the same illusory quality, we often be-
lieve it to be real. Our belief in this image draws us away from the true qualities of
our nature .... Because the self-image is based on how we wish we were, on what
we fear we are, or how we would like the world to see us, it prevents us from seeing
ourselves clearly. (pp. 102-103)
Another Buddhist scholar (Trungpa, 1976) describes how we attempt

to create the illusion of our egos as continuous, solid, and consistent. "So
we build up an idea, a preconception that self and other are solid and
continuous. And once we have this idea, we manipulate our thoughts to
confirm it, and are afraid of any contrary evidence" (p. 13). Interestingly,
from this particular Buddhist perspective, the preoccupation with the self,
watching oneself as an external object, is actually a form of ignorance,
self-observing ignorance. As Trungpa (1973) notes, self-observation can be
quite dangerous, it can involve watching yourself like a hungry cat watch-
ing mice. For Trungpa, the watcher is actually a very complicated bureau-
cracy that we set up to seemingly protect and enhance the self. (In our own
parlance, the watcher represents a rather frenetic, if not desperate, I-self,
preoccupied with managing the impression that the Me-self is making upon
the world.) However, from the Buddhist perspective, one must go beyond
this form of self-observation, one must remove the watcher and the compli-
cated bureaucracy that it creates to preserve the permanence of the self.
As Trungpa observes, once we take away the watcher, there is a
tremendous amount of space because the watcher and the bureaucracy
take up so much room. Thus, if one eliminates the watcher, the space
becomes sharp, precise, and intelligent. In fact, one does not really need
the watcher or observer of the self at all.
One can see in this comparative analysis rather divergent views of how
the I-self should be occupying its time. From the perspective of Western
psychology, the I-self should be gainfully employed in observing and pro-
tecting the Me-self, packaging it as a valued commodity in the psycho-
logical marketplace. From an Eastern, Buddhist perspective, far more
fringe benefits will accrue if the I-self averts this myopic gaze since it repre-
sents a distorted lens that obscures one's true nature. Rather, the I-self
should direct its energies outward, exercising its capabilities and enjoy-
ing life experiences rather than turning inward in its preoccupation with
the creation of a Me-self that it considers to be acceptable to the society at
large.
66 Susan Harter
On-I-People and Me-People

Having contrasted these two orientations, it becomes apparent that the
I-self does have a variety of potential job definitions, and that peoples'
lives differ with regard to which particular path they have opted, or been
compelled, to pursue. Metaphorically speaking, we can recognize I-people
and Me-people in the world, a task that is presumably easier than recogniz-
ing these qualities in ourselves. Even casual observation will permit us to
recognize those I-individuals who appear devoted to experiencing life, who
are attuned to what it feels like to appreciate a rainbow or a sunset or the
fragrance of the garden after a spring rain, who seem to have a zest for life,
and who appear curious to explore the outside world rather than the world
of their innermost thoughts, fears, and self-perceptions.
Me-people, in contrast, would appear to interpret their experiences from
the standpoint of what these might reflect about them as a person, what
their actions connote to the observing world as well as the observing ego.
The Me-stance is far more judgmental. What does this say about Me? As
a light-hearted excursion into exploring such a typology, consider the
following contrasts:
George Plimpton versus Howard Cosell
Icarus versus Narcissus
Mozart versus Salieri
Gaugin versus Van Gogh
Alice in Wonderland versus Miss Piggy
Paul Newman versus Woody Allen
Superman versus Clark Kent
E.T. versus Charlie Brown
Are we really serious about such a distinction, and is it likely that this
represents an individual difference of any interest or merit? We are just
beginning to pursue this issue in our own work (Harter & Rose, 1987).
Although we are merely in the beginning phases of this investigation, our
early efforts appear promising. In an attempt to determine whether people
could be contrasted in this manner, we have designed forced-choice items
in which individuals respond to a range of potential trait labels, indicating
whether they think about this attribute in terms of what it feels like to them
(namely the experience) or whether they are more likely to think about
what this says about them and the kind of person they are. We have found
that attributes such as smart, stupid, competent, assertive, patient, kind,
tolerant, nice, mean, caring, and aggressive function best to discriminate
between these two styles.
Our preliminary findings also suggest the fruitfulness of exploring certain
correlates of these two styles, namely the extent to which individuals intro-
spect about themselves or think a lot about the kind of person they are, as
well as the extent to which they are concerned about what other people
think about them. One can see in the description of these correlates certain
parallels to the dimension of self-monitoring (Snyder, 1979). For example,
our Me-individuals would appear to have many dynamics in common with
high self-monitors. The parallels between our I-people and low self-
monitors is less obvious, however, since we are clearly defining I-people
as those who are focused on experiencing the actions associated with attri-
butes. This feature does not necessarily define the low self-monitor as
operationally defined by the measures designed to tap the self-monitoring
dimension. Nevertheless, there are certain convergences that we shall
continue to explore.
Concluding Remarks
Our excursion into the processes by which the I-self constructs, conserves,
protects, and enhances the self began as an exploration in how these
mechanisms come to operate in the developing child and adolescent.
However, as we have seen, these processes can also be viewed from the
perspective of individual differences in how people interpret their experi-
ences. Finally, these constructs invite us to step out of our typical frame of
reference regarding the value of promoting self-protective strategies, to
consider alternative world views on the nature of the self. We have sug-
gested one such alternative, namely an Eastern, Buddhist perspective in
which the preoccupation with self-observation and self-enhancement may
well obscure our true nature and prevent one from being the "real me."
The underlying question provoked by such a shift in perspective can be
asked in one final form: How many times in the last month have you had an
experience or a very comfortable interaction with someone that has caused
you to exclaim, with some combination of surprise, joy, and relief: "I real-
ly felt like myself!"? Once, twice, three times? For most people it is re-
latively rare. But then if one so rarely has this experience, if one so rarely
feels like one's true self, the "real me" (with a small "m"), then who is one
the vast majority of the time? I rest my case.
REFERENCES
Allport, G.W. (1955). Becoming: Basic considerations for a psychology of perso-
nality. New Haven: Yale University Press.
Allport, G.W. (1961). Pattern and growth in personality. New York: Holt, Rine-
hart & Winston.
Bandura, A. (1978). The self system in reciprocal determinism. American Psycho-
logist, 33, 344-358.
Baldwin, 1.M. (1897). Social and ethical interpretations in mental development. New
York: MacMillan.
Brim, O.G. (1976). Life span development of the theory of oneself: Implications
68 Susan Harter
for child development. In H.W. Reese (Ed.), Advances in child development and
behavior. New York: Academic Press.
Carson, J. (1985). Adolescent development: Conflicts within the self. Honors
thesis, University of Denver.
Calkins, M. W. (1930). The philosophical credo of an absolutistic personalist. In
G.P. Adams and W.P. Montague (Eds.), Contemporary american philosophy,
vol. 1. New York: Macmillian, pp. 197-218.
Cooley, C.H. (1902). Human nature and the social order. New York: Charles
Scribner's Sons.
Damon, W. (1983). Social and personality development. New York: W.W. Norton.
infancy through adolescence. Child Development, 53,841-864.
Dickstein, E. (1977). Self and self-esteem: Theoretical foundation and their
implications for research. Human Development, 20,129-140.
Epstein, S. (1973). The self-concept revisited or a theory of a theory. American
Psychologist, 28,405-416.
Epstein, S. (1981). The unity principle versus the reality and pleasure principles, or
the tale of the scorpion and the frog. In M.D. Lynch. A.A. Norem-Hebeisen, &
K. Gergen (Eds.), Self concept: Advances in theory and research. Cambridge,
MA: Ballinger.
Erikson, E. (1968). Identity, youth and crisis. New York: Norton.
Fischer, K.W. (1980). A theory of cognitive development: The control and con-
struction of hierarchies of skills. Psychological Review, 87, 477-53l.
Gergen, K.J. (1968). Personal consistency and the presentation of self. In C.
Gordon & K.J. Gergen (Eds.), The self in social interaction. New York: Wiley.
Gergen, K.J. (1971). The concept of self New York: Holt.
history. American Psychologist, 7,603-618.
Harter, S. (1982). Developmental differences in children's understanding of self-
affect labels. Unpublished manuscript, University of Denver.
Harter, S. (1983). Children's understanding of multiple emotions: A cognitive-
developmental approach. In W.F. Overton, (Ed.), The relationship between
social and cognitive development. Hillsdale, NJ.: Erlbaum.
Harter, S. (1985a). Competence as a dimension of self-evaluation: Toward a com-
prehensive model of self-worth. In R. Leahy (Ed.), The development of the self
New York: Academic Press, pp. 55-122.
Harter, S. (1985b). The social support scale for children: manual. University of
Denver, Unpublished manuscript.
Harter, S. (1986a). Cognitive-developmental processes in the integration of con-
cepts about emotions and the self. Social Cognition.
Harter, S. (1986b). Processes underlying the constrution, maintenance, and en-
hancement of the self-concept in children. In S. Suls & A. Greenwald (Eds.),
Psychological perspectives on the self, (Vol. 3), Hillsdale, NJ.: Erlbaum, pp. 136-
182.
Harter, S. (1987). Self-concept formation in children. In N. Eisenberg (Ed.), Con-
temporary topics in developmental psychology.
Harter, S. (in press a). The relationships between children's self-perceptions,
affect, and motivation within the academic domain. In A. Boggiano and T.
3. The Construction and Conservation ofthe Self 69
Pittman (Eds.), Motivation and achievement from a social-developmental

perspective.
Harter, S., & Bresnick, S. (1987). Attributes as traits versus behaviors: Processes to
protect the self Unpublished manuscript, University of Denver.
Harter, S., & Connell, J.P. (1984). A comparison of alternative models of the
relationships between academic achievement and children's perceptions of com-
petence, control, and motivational orientation. In J. Nicholls (Ed.), The develop-
ment of achievement-related cognitions and behaviors. Greenwich, CT: J.A.I.
Press, pp. 219-250.
Harter, S. & Haltiwanger, J. (1986). Conservation of self: Stability and change in
children's perceptions of their smartness and dumbness.
Harter, S., & Hogan, A. (1985). A causal model of the determinants of self-worth
and the affective and motivational systems that it mediates. S.R.e.D. presenta-
tion, Toronto.
Harter, S., & Neeman, J. (1986). Dimensions and dynamics of the college students'
self-concept. Unpublished manuscript, University of Denver.
Harter, S. & Nowakowski, M. (1987). The dimensions of depression profile for
children and adolescents. University of Denver, Unpublished Manuscript.
Harter, S., & Robinson, N. (1987). Self-enhancing styles among middle school chil-
dren. Unpublished manuscript, University of Denver.
Harter, S., & Rose, S. (1987). On I-people and Me-people and the legacy of William
James. Unpublished manuscript, University of Denver.
Horney, L. (1950). Neurosis and human growth. New York: Norton.
James, W. (1892). Psychology: The briefer course. New York: Holt, Rinehart &
Winston.
Jourard, S.M. (1971). Self-disclosure: An experimental analysis of the transparent
self New York: Wiley.
Jung, e.G. (1928). Two essays on analytical psychology. New York: Dodd, Mead.
Kanfer, F.H. (1980). Self-management methods. In F.H. Kanfer & A.P. Goldstein
(Eds.), Helping people change: A textbook of methods, (2nd ed.). New York:
Pergamon Press, pp. 334-389.
Kelly, G.A. (1955). The psychology of personal constructs. New York: Norton.
Lecky, P. (1945). Self-Consistency: A theory of personality. New York: Island
Press.
Lewis, M., & Brooks-Gunn, J. (1979). Toward a theory of social cognition: The
development of the self. In I. Uzgiris (Ed.), New directions in child development:
Social interaction and communication during infancy. San Francisco: Jossey-Bass.
Maslow, A.H. (1961). Peak-experiences as acute identity-experiences. American
Journal of Psychoanalysis, 21,254-260.
Mead, G.H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Messer, B. Harter, S. (1986). The adult-self perception profile. Unpublished manu-
script, University of Denver.
Mischel, W. (1969). Continuity and change in personality. American Psychologist,
24, 1012-1018.
Mischel, W. (1973). Toward a cognitive social learning reconceptualization of per-
sonality. Psychological Review, 80, 252-283.
Monsour, A. (1985). The structure and dynamics of the adolescent self-concept.
Doctoral Dissertation, University of Denver.
70 Susan Harter
Rogers, C.R. (1950). The significance of the self-regarding attitudes and percep-
tion. In M.L. Reymert (Ed.), Feeling and emotions: The Moosehears sympo-
sium. New York: McGraw-Hill.
Rosenberg M. (1979). Conceiving the self. New York: Basic Books.
Rosenberg, M. (1986). Self-concept from middle childhood through adolescence.
In S. Suls & A. Greenwald (Eds.), Psychological perspectives on the self, (vol. 3).
Hillsdale, NJ: Erlbaum, pp. 107-135.
Snyder, M. (1979). Self-monitoring processes. In Berkowitz, L. (Ed.), Advances in
experimental social psychology (vol. 12). New York: Academic Press, pp. 86-
13l.
Trungpa, C. (1973). Spiritual materialism. Berkeley: Shambhalla Books.
Trungpa, C. (1976). The myth offreedom. Berkeley: Shambhalla Books.
Tulku, T. (1978). Skillful means. Berkeley, CA.: Dharma Publishing.
Vallacher, R.R. (1980). An introduction to self theory. In D.M. Wegner & R.R.
Vallacher (Eds.), The self in social psychology. New York: Oxford University
Press, pp. 3-30.
Wicklund, R.A. (1975). Objective self-awareness. In L. Berkowitz (Ed.). Ad-
vances in experimental social psychology (Vol. 8). New York: Academic Press.
Wylie, R. (1961). The self concept: A critical survey of pertinent research literature.
Lincoln, NB: University of Nebraska Press.
Wylie, R. (1974). The self concept: Revised Edition (Vol. 1). A review of metho-
dological considerations and measuring instruments. Lincoln, NB: University of
Nebraska Press.
Wylie, R. (1979). The self concept: vol. 2. Theory and research on selected topics.
Lincoln, NB: University of Nebraska Press.
4
The Adolescent Self-Concept in
Social Context
DANIEL HART
Adolescents come to know and define themselves largely through social

interaction. In part, the adolescent's self-concept may be a product of the
incorporation of others' communications. If family relatives, for instance,
all proclaim the daughter to be a spitting image of the mother, the daughter
may come to think of herself in those terms. Even self-attributions that are
not first suggested by others derive their meaning within a social medium.
The adolescent who thinks of the self as "shy" is perhaps implicitly com-
paring the self to friends or peers who are more outgoing socially. Not only
is "shy" in this sense social-relational, its selection as a component of the
self-conception may reflect a feature that differentiates the self from others
(McGuire & Padawer-Singer, 1976).
Although the influence of the social context on the adolescent's self-
concept is apparent, self-concept research has proceeded with little con-
cern for it. My purpose in this chapter is to demonstrate the importance
of considering the adolescent's self-concept, or self-understanding, in its
social context. In the first section of the chapter, the relationships between
this approach and more common themes of developmental self-concept
investigation are examined. In a later section, some new research is pre-
sented on the adolescent's self-concept in its social context.
Themes of Self-Concept Research

In The Principles of Psychology (1890/1950), William James described for
the first time what has become a most influential topography of the self-
concept and self-experience. His map to the landscape of the self-concept
has led directly to three lines of self-concept research, two of which are
The research described in this chapter was supported in part by a National Science
Foundation Research Opportunity Award (BNS 83-01027, Seymour Rosenberg,
Principal Investigator) and by an award from the Rutgers Research Council.
72 Daniel Hart
frequently followed in developmental investigation. The first is the study

of self-evaluation. James proposed that one type of self-evaluation, self-
esteem, is a function of both one's goals and one's achievements, as repre-
sented in the equation, self-esteem = success/pretensions. According to
James's formula, someone who had not obtained any of their goals would
be at the zero end of the self-esteem dimension, while another who had
met success for all of his or her pretensions would be at the opposite end.
Investigators pursuing the self-evaluation type of self-concept research
usually attempt to elicit the individual's own assessment of the self in re-
spect to a polar dimension like self-esteem. Because individuals evaluate
themselves in various ways, there are subtypes of self-evaluation studies,
concerning, for instance, self-esteem (e.g., Rosenberg, 1979), locus of con-
trol (e.g., Nowicki & Strickland, 1973), and perceived competence (e.g.,
Harter, 1982). Typically, each subject in one of these studies receives a
numerical score representing his or her position along the evaluative
dimension.
A second, commonly traveled line of research focuses on self-
understanding, which consists of the individual's beliefs, attitudes, and
cognitions about the self. These various thoughts are elicited by asking
each subject to respond to questions such as "Who am I?" (Kuhn &
McPartland, 1954). In self-understanding too, there are subtypes corre-
sponding to the different sorts of thoughts and beliefs that an individual has
about the self. Again, returning to the theory of James, the individual's
various thoughts of self can be parsed into the "self-as-object" (subsuming
the physical, active, social, and psychological qualities), and the "self-as-
subject" (beliefs about the nature of the self's continuity over time, dis-
tinctness from others, and volition). The self-understanding researcher
attempts to describe in qualitative terms the individual's general under-
standing of self (e.g., Montemayor & Eisen, 1977; Secord & Peevers,
1974). While there are no studies (to my knowledge) that investigate the
interaction between self-evaluation and self-understanding, such an in-
teraction must surely exist. An evaluation of an object is based upon one's
conception of it; self-understanding is the conception of self upon which
self-evaluation is based.
Recently, Damon and Hart (Damon & Hart, 1982, 1986; Hart &
Damon, 1986; Hart, Lucca-Irizarry, & Damon, 1986; Hart, Maloney, &
Damon, 1987) have proposed and validated a developmental model of
general self-understanding development that utilizes the distinctions pro-
posed by James. This model is depicted in Figure 4.1. James's self-as-object
corresponds to the definitional self represented by the front face of the
model, which is divided into the four types of self-characteristic identified
by James. The side of the model, labeled the subjective self, corresponds
to the self-as-subject aspect of self-experience in James's theory. Again,
the subdivisions proposed by James are also included in the model (sense
of self-continuity, sense of distinctness from others, sense of volition).
o~~
",6«'
",'3-\ '3-. \\"
?e~so. :(\s \(\
\.'0-'3-\\0
e"'lJ. se\\
e(\Ge
. ",'3-",6
'0--0\G'3-\\O~.
Cp«'«' ~\",\e
.("\~oG'lJ. e(\ce
~ev\Y \(\\\"1'>
'3-v\\o(\
Developmental ~
Level se\1
.r ~\s'(\es, e(\ce
Physical Attributes Active Attributes Moral or Personal Belief Systems, \O~\S, \(\\\\)
;2
('1)
4. Late Reflecting Voli- That Reflect Choices Concerning Personal Philo- I>\ ~e"'\S
Adolescence '3-",6\'3- :»
tional Choices, or Choices, Personal Social Relations or sophy, Self's Own ;e\-.Je':l e'<s
I se\1
0.
Personal & Moral or Moral Standards Social-Personal ity Thought Processes 0\'(\ o
Standards Characteristics 6 0 '" \'(\e \. '0-"0Co(\'- (p
'0'3-5e \i\(\~ ese\\ 'J>
(')
3, Early Physical Attributes Active Attributes Social-Personal ity Social Sensitivity, oCiJ'" 01 <,,:1-\e~(\'3- \'3-v\O~s ('1)
Adolescence I That Influence So- That Influence So- Characteristics Communicative Com- o\\'3-'OW se\1 I:l
pete nee, & Other '0'3-('\\ \..\'«'e x>\e ....
cial Appeal & So- cial Appeal & So- :-Je'{ '0\'0. \~\e~~\"0e
0: .\\0(\ r/)
cial Interactions cial Interactions Psychologically Re- o ",\«'«' \ -.,\0\\
lated Social Skills 6 0 . c'/).\ ~
'O'3.Se o\O~\ Se(\se 0 .....
(J
2. Middle & Late Activity-Related Capabilities Activities That Are
Considered With Ref-
Knowledge, Learned
I~
Qsi~~eS o
Childhood I Physical Attributes Relative to Others Skills, Motivation or 0..'0'3.\\ I:l
(')
I I erence to Reactions Activity-Related
('1)
(Approval or Disap- Emotional States
proval) of Others "S-
Bodily Properties Typical BehaVior Fact of Membership Momentary Moods,
S·
1. Early
r/)
Childhood or Material in Particular Social Feelings, Prefer-
Possessions Relations or Groups ences & Aversions
o
D.
e:..
PhYSical Self Active Self Social Self Psychological Self (J
o
The Definitional Self
....I:l
('1)
~
FIGURE 4.1. Developmental model of self-understanding.
-.J
V>
74 Daniel Hart
The four horizontal rows along the model's face represent the four se-
quential developmental levels of organization. These levels are the major
modes of construing the self from childhood through adolescence, and ap-
ply across both the definitional and subjective selves. At Levell, the self is
viewed as a collection of physical qualities, typical activities, group mem-
berships, and simple thoughts and feelings, all of the which are determined
by external factors. Level 2 is characterized by the examination of the selfs
characteristics in light of social standards and the characteristics of others.
Self-continuity and distinctness from others rest on the degree to which the
self relative to others possesses a particular psychological characteristic.
The self determines its own future through wishes and effort. At Level 3,
self-characteristics are important in so far as they influence the selfs in-
teractions with, or the selfs attractiveness to, others. Interactions with per-
sons in the immediate social context help to shape the self, and one's loca-
tion in the social context underlies self-continuity and distinctness from
others. At Level 4, the importance of the immediate social context wanes
in influence, and is replaced by the self's subjective experiences and belief
systems as the core of the self-concept. As indicated by the unique content
of each box, the developmental levels apply somewhat differently to each
scheme. Thus there is a fundamental structural unity across different
aspects of the self-concept, but also some disunity, due to the distinct
nature of each scheme X level interaction.
What does the model and its associated research reveal concerning the
nature of the adolescent's self-concept? It is during early and mid-
adolescence, according to the model, that the adolescent's self-concept is
most embedded within the immediate social context of family, friends,
schoolmates, and so on. The adolescent claims that the various self-
characteristics are important to the degree to which they foster integration
into the social context. One's position in this social network provides the
adolescent with the sense of self-continuity, as well as the experience of
uniqueness or distinctness from others. Finally, the adolescent believes
that the nature of the self is determined through communication and
interaction with others. Because the adolescent's sense of self is so em-
bedded within a social context, it seems particularly important to consider
the self-concept within a network of relationships, a third line of research
suggested by William James.
Social Selves and Social Contexts

The adolescent's social interactions are not all the same, of course, and
some facets of the self-concept are more apparent to self-reflection in par-
ticular relationships. For example, knowledge of the selfs popularity
among peers is more accurately drawn from conversations with and the
reactions of friends and schoolmates than from the statements of parents.
4. The Adolescent Self-Concept in Social Context 75
The different social contexts of the adolescent not only reveal self-
knowledge, but shape it as well. Each social context exerts unique role
demands: for instance, with parents, there may be expectations to be obe-
dient, helpful, and polite; with friends, loyal, talkative, and active; and so
on. These role expectations may not determine the adolescent's behavior,
but they surely influence it. This is a point well-noted by James, who wrote,
"Many a youth who is demure enough before his parents and teachers,
swears and swaggers like a pi1ate among his 'tough' young friends" (18901
1950, p. 169). One consequence of this role-adaptive behavior is that the
self upon which the adolescent reflects may differ according to the social
context in which it is considered.
Because what the adolescent thinks about the self varies according to the
social context, it is possible to describe the adolescent as having a number
of different social selves. This has important theoretical and methodologi-
cal ramifications. For theory, the differentiation of the sense of self into a
number of different selves means that there may be not be a monolithic
self-concept. Instead of a single self to be evaluated and thought about,
there could be multiple selves, each of which is evaluated and thought
about in different ways. A more appropriate conceptualization would then
be to characterize the sense of self as a confederacy of self-concepts, each
one appropriate for a particular range of social contexts.
Within this perspective, the nature of the confederacy becomes a pivotal
concern. Are the various selves anarchists, with little organization and few
similarities among them? Or are the selves soldiers drawn together within a
tight organization and bound by numerous commonalities? As James
(1890/1950, p. 294) noted, the degree of organization and similarity
between the different self-concepts can vary from person to person: there
"may be a discordant splitting" or it may "be a perfectly harmonious divi-
sion."
The degree of organization in the self-concept confederacy is likely to
vary according to the extent of similarity among the social selves. At one
extreme, an individual may not recognize any differences among the social
selves, and there would be no need for organization among them. This is
unlikely for most people however. If there are no differences in the sense
of self for different social settings, the individual is either failing to adjust
appropriately to the situation, or is completely without an awareness of
what changes in the self take place. At the other extreme is the individual
who perceives few connections in the sense of self across social roles. No
satisfactory degree of organization can be imposed on selves that share few
common characteristics. Distress and anxiety may be a consequence of a
lack of unity among the different selves, due to the violation of the indi-
vidual's desire for personal consistency (Lecky, 1945) or unity (Epstein,
1981). Dissimilar, unorganized social selves might also lead to the sense
that there is no "real" self, like the central character in Woody Allen's
movie, Zelig, who changes chameleon-like to fit the social context.
76 Daniel Hart
The sense of self as composed of different social self-concepts was first

elaborated within a developmental context by Baldwin (1902) and Mead
(1934). Both theorists argued that, prior to the age of 4 years, the child's
sense of self consists of disparate social role-dependent self-concepts. As a
consequence, the child's behavior varies widely from one social context to
another. Baldwin and Mead suggested that between the ages of 4 and 6,
the different social selves of the child become organized. The emergent
organization of selves results in some behaviors related to a particular
social self being evident across different contexts. This is because the selves
are arrayed in a hierarchical organization, with some selves more impor-
tant than other social selves. For instance, the self with parents may not
steal, and this self may predominate over the peer identity when the child's
friends are stealing.
During adolescence, several factors may contribute to both decreasing
similarity among social selves and the weakening of their organization.
One of the most apparent factors is the discrepancy between what parents
and peers expect from the adolescent. Although there is substantial debate
about the magnitude of the discrepancy between parental and peer ex-
pectations (Hartup, 1983), Berndt (1979) has found that it is greatest dur-
ing early adolescence, particularly for antisocial acts. The growing schism
between the self-with-parents role and the self-with-peers role may make it
especially difficult for the adolescent to discern important similarities
among the different social selves.
A second factor that exacerbates the difficulties involved in organizing
the various social selves is the deterioration during adolescence of the
social self provided by the parents (Adelson & Doehrman, 1980). To the
child, the hierarchy of different social selves is dictated by the demands,
values, and goals of the parents. For instance, religious parents are likely
to have children for whom the self-concept of church-goer is important.
However, as the child becomes an adolescent, the values and goals of the
parents become less important in organizing the various selves. When this
process begins, the adolescent may not have formulated substitute goals
and values to construct an alternative hierarchy, and hence there may not
be a framework within which to order the various selves.
To this point, I have outlined some of the reasons that the self-concept
should be considered, at least in part, as composed of a number of social
self-concepts, and that the organization of these selves may be challenged
by the tasks of adolescence. These ideas are not new, of course, descending
as they do from the well-known work of William James, and elaborated in
the theories of Baldwin and Mead; more recently, theorists such as Kihl-
strom and Cantor (1984) and Martindale (1980), among others, have resur-
rected and reconstructed these same ideas. What is striking is how little
influence these views have had on empirical work, apart from the few
studies to be described in more detail in later sections. In most research on
the self-concept in childhood and adolescence, subjects respond to social-
contextless self-evaluation scales or to self-understanding questions such as

"Who am I?" (the self-understanding research has been reviewed by
Damon and Hart, 1982). Underlying such a research strategy, as Smollar
and Youniss (1985) note, is the implicit assumption that children and
adolescents have one general concept of self that can be described apart
from a social context. However, if the child's or the adolescent's sense of
self varies greatly from one social role to the next, with little organization,
such a methodology would seem to be seriously flawed.
PRIOR RESEARCH ON THE SELF-CONCEPT IN SOCIAL CONTEXT

Does the adolescent's sense of self vary greatly from one social role to
another? There is little research evidence with which to judge. Only three
studies have considered the adolescent's self-concept in social context. Ge-
case (1972) asked 16- to 17-year-old adolescents to imagine themselves in
the roles of student, family member, being with somebody of the opposite
sex, peer group member, and being with adults. For the self in each of
these roles, the adolescent completed a self-esteem inventory. Gecas found
that the average correlation between self-esteem scores in the different
roles was .54, indicating that there is both constancy and variability across
social contexts. The nature of this study does not permit any inferences
about why the social context affects self-esteem in these ways, because only
a gobal measurement was utilized. Also, this study only addresses self-
evaluation, and not whether adolescents think about themselves in similar
ways across social roles.
In a study of self-understanding in social context, Volpe (1981) asked
subjects of ages 10 to 23 years to provide three characteristics for each of
three sentence stems: "When I am with my close friend of the same sex I
am __ "; "When I am with my mother I am __"; and "When I am with
my father I am __ ." Characteristics offered in response to these stems
were assigned to one of eight categories, such as mature (serious, intellec-
tual, quiet), sociable/cooperative (friendly, obedient), openness (honest,
trusting, free), or anxious (confused, frustrated). Volpe found that sub-
jects tended to describe the self-with-father in terms of maturity more
often than the self-with-mother or self-with-friend, and that the self-with-
friend was distinctively described in terms of openness. Although the re-
sults suggest that self-understanding varies from one social context to
another, the procedure used to elicit self-understanding prohibits any
strong conclusions. The subjects were never asked whether a characteristic
offered for one social self described each of the others; so for example, the
qualities of openness that a subject offered to describe the self-with-friend
might also have been found to be characteristic of the self-with-parents if
the subject had been asked. Volpe's implicit assumption appears to have
been that if openness was an important or salient descriptor of the self-
with-parent, it would have been one of the subject's first three responses to
78 Daniel Hart
the appropriate sentence stem. But this is not necessarily true. Research by
McPhail and Tucker (1972) has demonstrated that the order in which char-
acteristics of self are listed in response to questions like "Who am I?" has
little connection to their subjective importance to the subject.
In a study using Volpe's methodology, Smoller and Youniss (1985) in-
ferred similarity between two social selves if the two selves were described
by a characteristic from the same category. For instance, if the self-with-
best-friend was characterized as "friendly" and self-with-mother was de-
scribed as "obedient," the two selves were judged to be similar because
both were described with adjectives from the sociable/cooperative cate-
gory. This procedure is somewhat problematic, however, because the cate-
gories are probably too broad to allow these inferences to be convincing.
For instance, it is not clear why a self described as "friendly" and another
characterized as "obedient" should be considered as similar. Because of
this methodological problem, their finding that there is less similarity
among the different selves for older adolescents than there is for younger
adolescents must be accepted cautiously. Together, these three studies in-
dicate that the social context in which the adolescent envision~ the self
affects the self-concept, but leave mostly unanswered questions about simi-
larity and organization among social selves.
A RESEARCH EXAMPLE
In this section, some research is described that aims to elucidate the simi-
larity and organization among four social selves important in adolescence
(self-with-mother, self-with-father, self-with-best-friend, and self-with-
peers) that are not known well. This study differs from the studies con-
sidered above in two ways. First, similarity among the different selves is
directly assessed by asking each adolescent if a characteristic offered to
describe one social self also defines another (for instance, "You said that
when you are with your mother, you are talkative; are you talkative when
you are with your father? ... with your best friend? ... with kids you
don't known well?"). This procedure also allows the identification of those
features of self that differentiate one social self-concept from another. In
the above example, "talkative" may apply only to the self-with-mother, in
which case it serves to differentiate the self-with-mother from the other
social self-concepts. Because communication and intimacy become in-
creasingly important in relationships as adolescents become older (Youn-
iss, 1980), it is predicted that characteristics of self that reflect these qual-
ities will often differentiate among social selves.
The following study also presents an initial attempt to describe the orga-
nization of the four social selves, using a set-theoretical model (Rosenberg
& Gara, 1985) that is consonant with the developmental theories of Mead
and Baldwin previously discussed. Within this model, one of four different
types of relationships may exist between any two selves: 1) two selves, A
FIGURE 4.2. 1, A and B are identical; 2, A is superordinate to B; 3, B is superordi-

nate to A; 4, A and B are distinct.
and B, may be identical and belong to the same set (the two selves are
described by exactly the same descriptors); 2) Self A may contain all of the
descriptors of Self B, plus some additional descriptors not contained in B
(A is superordinate, B is subordinate); 3) Self B may contain all of the
descriptors of Self A, plus some additional descriptors not contained in A
(B is a superordinate set, A is a subordinate set); and 4) the two selves may
be distinct, constituting either partially overlapping (having some descrip-
tors in common) or disjunctive sets (no shared descriptors). These possible
relationships are depicted in Figure 4.2. Each superordinate self is
hypothesized to be psychologically more important than the subordinate
selves that it encompasses (Rosenberg & Gara, 1985). This is because the
more extensive description of the superordinate self indicates that the indi-
vidual either spends a great deal of time in that social role or is especially
committed to it, each resulting in the exploration and elaboration of the
features of that social self. This hypothesis has received preliminary con-
firmation in two studies, one of professional women (Joseph, 1985), and
80 Daniel Hart
the other of college students (Grubb, 1986). Of the four social selves con-
sidered in this chapter, it is expected that the self-with-mother or the self-
with-father will be superordinate during late childhood and early adoles-
cence. However, for older adolescents who are beginning to separate from
their parents and for whom peer roles are becoming increasingly impor-
tant, it is expected that the self-with-best-friend role will become the most
common superordinate self.
Subjects and Procedures

Seventy-three students from Grades 5, 7, 9, and 11 were individually inter-
viewed at their schools. In the first part of the interview, each student
responded to five questions. These questions were:
1. What kind of person do you think you are? (general self)
2. What kind of person are you when you are with your mother? (self-
with-mother)
3. What kind of person are you when you are with your father? (self-with-
father)
4. What kind of person are you when you are with your best friend? (self-
with-best-friend)
5. What kind of person are you when you are with kids you don't know
well? (self-with-unknown-kids)
Students were encouraged to offer as many descriptors as possible for each

of the different questions.
In the second part of the interview, each student was asked to judge
whether a descriptor offered for the general-self question or for one of
the self-role questions applied to each of the other self-role questions: not
at all (0), not usually (1), most of the time (2). (For example: "You said
that you think of yourself as a shy person when you are with your mother;
are you a shy person when you are with your father? . . . with your best
friend? ... with kids you don't know well?) This procedure results in a
matrix of descriptors by social selves, with 0, 1, and 2 as the entries in the
matrix indicating whether each descriptor is unique to a social self or
whether a descriptor also applies to one or more of the other social selves.
An example of a matrix is found in Table 4.1
Finally, subjects were asked to judge "How real I am" when with the
mother, father, best friend, and unknown kids on a scale of 1 to 10, with 1
representing not being "real" at all and 10 representing that they were
totally authentic, or 100% "real."
The data contained in the matrices were analyzed in two different ways.
First, the descriptors in each matrix were analyzed with a content-coding
scheme. This coding scheme permits each descriptor to be assigned to a
particular category. The purpose of assigning descriptors to these catego-
ries was to allow an analysis of the types of descriptors that describe the
TABLE 4.1. An example of a descriptors x social selves matrix.
Social selves
With With With With

Descriptors mother father friend kids
Easygoing 2 2 2 2
Understanding 2 2 2 0
Independent 1 1 2 1
Help with problems 2 0 2 1
Speak my mind 1 1 2 1
Kind 2 2 2 2
Do active stuff 1 2 2 1
Have a good time 2 2 2 2
Ride horses 0 2 2 2
Competitive 1 1 2 2
Talk about boys 2 1 2 0
Nonhesitant 2 2 2 2
TABLE 4.2. Categories for coding descriptors for the four social selves.
A. Subjective self
1. Individuality ("different," "like anybody else")
2. Effort ("try real hard," "never give 100% ")
3. Independence ("don't lean on anybody," "dependent")
4. Continuity ("always the same," "changing all the time")
B. Objective self
1. Physical
a. Size ("big," "fat")
b. Attractiveness ("pretty face," "nice hair")
c. Other ("brown eyes")
2. Active
a. Capabilities ("can ride a bike," "not able to play football")
b. Typical activities ("play kickball at recess")
c. Activity level ("don't do much," "very busy")
d. Intrepidness ("wild")
3. Social
a. Relationships ("have a brother")
b. (+) Personality traits ("friendly")
c. (-) Personality traits ("mean")
d. Intellectual social comparisons ("know more than other kids," "dumber than
everybody else")
e. Conformity to standards ("clean my room," "don't make my bed")
f. Enjoyable person to be with ("fun to be with," "not fun to be with")
g. Helpful ("help my friends")
3. Psychological
a. Preferences ("like football," "don't like math")
b. Emotional ("get mad easily," "happy")
c. Diffuse ("curious," "reflective")
d. Beliefs ("Christian," "important to work for world peace")
82 Daniel Hart
sense of the general self, as well as to make possible the identification of

the type of descriptors that distinguish each social self from the others. The
content-coding system is based upon a scoring manual developed in other
work (Damon & Hart, in preparation), and reflects the general distinctions
made by James and discussed earlier in reference to the developmental
model of self-understanding. For the purposes of this study, a number of
subcategories were proposed, in order to allow finer distinction. The dif-
ferent categories are presented in Table 4.2. In a previous study (Hart,
1986), two raters assigned a descriptor to the same category for 88% of the
descriptors.
Five types of descriptors from each matrix were coded with the context-
coding scheme: 1) all self-descriptive descriptors (descriptors that were
judged to describe the general self or any of the social selves); 2) descrip-
tors unique to the self-with-mother (all descriptors for which a 2 was reg-
istered in the self-with-mother's column and a 1 or a 0 in the columns for
the other social selves); 3) descriptors unique to the self-with-father (all
descriptors for which a 2 was registered in the self-with-father's column and
a 1 or a 0 in the columns for the other social selves); 4) descriptors unique
to the self-with-best-friend (all descriptors for which a 2 was registered in
the self-with-best-friend's column and a 1 or a 0 in the columns for the
other social selves); and 5) descriptors unique to the self-with-unknown-
kids (all descriptors for which a 2 was registered in the self-with-kids col-
umn and a 1 or a 0 in the columns for other social selves). For each subject,
percentages of descriptors in each category of the coding system were
calculated, in order to control for differences among subjects in the num-
ber of descriptors offered. 1 For each subject and for each of the four
aspects of the matrix, then, the number of descriptors in a particular cate-
gory was divided by the total number of descriptors in that aspect.
Second, the set relationships among the four social selves were deter-
mined by hierarchical classification with a three-rank solution specified on
each matrix using the computer program HICLAS (DeBoeck, 1985). The
underlying model and rationale for this analysis is described by DeBoeck
and Rosenberg (1985). HICLAS only accepts dichotomous data, so it was
necessary to condense the data range of 0-2 found in the matrices. In order
to highlight the differences among the four social selves, the Is were re-
corded as Os, leaving 2s and Os as the only values in the matrices. The
program provides the set relationships among the four selves. These set
relationships were diagrammed for each subject. An example of a diagram
is found in Figure 4.3. In this example, the self-with-best-friend is super-
1 If a subject offered no unique characteristics for a self, zeros were registered for
the percentage of descriptors in each of the categories. Since zeros represent the
absence of any unique characteristics, and not a failure to complete the task, they
are included in the analyses.
Self with
Best Friend
FIGURE 4.3. Self-with-best-friend is superordinate to self-with-mother and self-with-

father; self-with-unknown-kids is distinct from all of the three other selves.
ordinate to the self-with-mother and self-with-father, and is defined by the

descriptors characterizing each of these two social selves. The self-with-
kids is distinct from these three selves, although it is defined by some ofthe
same descriptors.
Differences and Similarities Among Social Selves

Do adolescents differentiate among social selves? The results suggest that
they do, increasingly so with age for the self-with-best-friend. As outlined
above, five aspects of the matrix were coded for content: the self-
descriptive descriptors and the descriptors unique to each one of the four
social selves. The mean number of descriptors for self-description, unique
to self-with-mother, unique to self-with-father, unique to self-with-best-
friend, and unique to self-with-unknown-kids were, respectively: 16, 0.8,
0.7, 1.5, and 0.9. Approximately 25% of the descriptors offered to char-
acterize the self, then, are specific to a particular social context. To deter-
mine if any of the social selves becomes increasingly differentiated from the
others with age, four ratios were constructed: number of descriptors
unique to each social role to the number of self-description descriptors.
Ratios were used in order to control for the general age-related increase in
the number of descriptors. Only one ratio, that involving descriptors uni-
que to the self-with-best-friend was correlated with age, r = .28, P < .05.
This means that the older subjects are attributing relatively more unique
descriptors to the self-with-best-friend than are the younger subjects.
The context analysis augments these findings by identifying the types of
characteristics of self that are unique to a specific social self. Table 4.3
presents the averages for the percentage of descriptors in each of the cate-
gories unique to each social self. For each of the social selves, the general
social category contains the largest percentage of descriptors. This means
that adolescents' descriptions of themselves with parents, friends, and un-
84 Daniel Hart
TABLE 4.3. Percentage of descriptors in each category for self-description and the
four unique roles.
Percentage of descriptors
Unique with
Self-
descriptive Mother Father Friend Peers
B. 2. Active 18 2 16 24 4·
a. Capabilities 2 0 2 1 0
b. Activity level 2 0 0 2 3
c. Typical activities 12 2 14 15 1·
d. Intrepidness 2 0 0 7 0
3. Social 58 39 21 34 46·
a. Relationships 2 3 0 3 0
b. (+) Personality traits 23 3 1 11 4·
c. (-) Personality traits 7 7 5 3 27·
d. Intellectual social comparisons 3 2 3 0 1
e. Conformity to adult standards 4 3 0 1 1
f. Talkative 14 17 9 11 12
g. Listening 1 0 1 1 0
h. Enjoyable person to be with 1 2 0 2 0
i. Helpful 3 1 1 1 0
4. Psychological 17 9 8 10 11
a. Preferences 5 2 0 2 0
b. Emotional 6 4 3 4 3
c. Diffuse psychological 5 2 5 3 8
d. Beliefs 1 1 0 0 0
• p < .05 (for the comparison ofthe four unique social selves).
known kids differ most frequently in terms of the social characteristics

attributed to the self in these varying social contexts. Within the general
social category, the talkative subcategory accounts for the largest percent-
age of descriptors unique to each social self, confirming the hypothesis put
forward that communication characteristics distinguish between social
selves. The averages for each category presented in Table 4.3 appear to
differ in magnitude from one social self to another, however. Repeated-
measure MANOVAs (O'Brien & Kaiser, 1985), with grade and sex as
between-groups factors, were conducted to determine if there were sig-
nificant differences among the four unique perspectives in the percentage
of responses in the active, social, and psychological categories. Signif-
icant effects were found between the four social selves for the general
active (F [3, 63] = 10.2, P < .001) and for the general social categories (F
[3, 63] = 3.9, p < .05), with no significant interactions with grade or sex.
Subsequent MANOVAs of the subcategories of the general active and so-
cial categories revealed significant differences for the percentage of de-
scriptors unique to the four social selves in the following subcategories:
typical activities (F= 6.51, P < .001), positive personality traits (F= 5.4,
P < .01), and negative personality traits (F= 9.9, P < .001).
An inspection of the average percentage of descriptors in Table 4.3 for
each of these significant comparisons gives them psychological meaning.
Considering first the general active category, adolescents appear to make
more unique attributions of active qualities to the self-with-father (16% of
the descriptors unique to this social self) and the self-with-best-friend (24%
of the descriptors unique to the self-with-best-friend), but only rarely is the
self-with-mother or the self-with-unknown-kids characterized with active
category descriptors unique to one of these selves. The trends contained in
the active subcategories suggest: (a) typical activities are important in dif-
ferentiating the self-with-father and self-with-best-friend from other social
selves, and (b) intrepidness (e.g., "I'm wild") is only ever unique to the
self-with-best-friend.
In comparison to the descriptors unique to the other selves, the self-
with-father appears to possess relatively few unique social descriptors
(21 % of the total, as opposed to 39%, 34%, and 45% of the total descrip-
tors for, respectively, the self-with-mother, the self-with-friend, and the
self-with-unknown-kids). Perhaps not surprisingly, adolescents think they
exhibit negative personality traits (e.g., "shy," "unfriendly") with un-
known kids (27% ofthe descriptors unique to self-with-unknown-kids) that
are not characteristic of the self-with-parents or self-with-friends (7%, 5%,
and 3% for the self-with-mother, self-with-father, and self-with-best-
friend, respectively). Finally, positive personality traits appear to be most
commonly unique to the self-with-best-friend descriptions, with 11 % of the
total descriptors unique to the self-with-best-friend of this type.
It is also useful to identify the types of self-characteristics that do not
vary from one social self to another. For each category in Table 4.3, a
comparison of the self-descriptive percentage with that for each of the four
unique selves provides an estimate of constancy across social context.
Several categories of descriptors are frequent in self-description, but are
rarely unique to one social self. For instance, positive personality traits
constitute 23% of all self-descriptive descriptors, but are relatively infre-
quent in the attributions specific to one social context. Psychological de-
scriptors and, in particular, preferences also are found with more frequen-
cy in self-description than among the descriptors unique to one social self.
If personality traits and psychological characteristics are considered central
characteristics of self (Livesly & Bromley, 1973), then adolescents believe
these core characteristics are relatively constant over social contexts.
In summary, the content analyses of the descriptors unique to the dif-
ferent social selves suggest the following: the self-with-father is frequently
differentiated from the other selves by typical activities; the self-with-best-
friend is often distinguished from the other selves by activities, intrepid-
ness, and positive personality traits; and it is only the self-with-unknown-
kids that is regularly separated from the other social selves by negative
86 Daniel Hart
personality traits. Remaining constant across social contexts are the self's
personality traits and psychological characteristics.
Organization Among Social Selves

To this point, the analyses have focused on identifying the degree and
nature of similarities between social selves. In an earlier section of this
chapter, however, it was suggested that an additional issue of some impor-
tance concerns the organization of the social selves. The organization of
the four social selves in this study were assessed within the set model de-
scribed earlier. The analysis yielded several findings. First, for 38 of the 73
subjects, there was no superordinate self. For almost all of these subjects
this means that the self-with-mother, self-with-father and self-with-best-
friend were distinct from each other, with the self-with-unknown-kids be-
longing to the same set as one of these social selves. Within the model
postulated by Rosenberg and Gara (1985), this suggests that these different
selves are equally prominent and important for these subjects.
The second interpretation is that the failure to identify a hierarchical
organization among the four social selves considered in this study accurate-
ly captures the psychology of the adolescents for whom this finding held.
Perhaps all four social selves are equally important to some adolescents.
There is a reasonable argument for this interpretation. All of the adoles-
cents in this study were living at home, and by necessity interact on a regu-
lar basis with their parents, so the self-with-mother and the self-with-father
probably remain important to many of them. The best-friend relationship
is also of special significance to the adolescent, because it provides a forum
in which independence from the parents can be exercised, and problems
and concerns can be shared (Bios, 1968). For these reasons, the self-with-
friend may not be subordinated to the self-with-parents. Finally, the self-
conscious adolescent (Elkind & Bowen, 1979) may believe that the self-
with-unfamiliar-fellow-students is important as well, because these peers
are thought to be critically evaluating the adolescent, and assessment that
may determine the selfs social standing and popularity. Unfortunately, no
data were collected that can be used to test which of these two explanations
for the lack of a superordinate self in these subjects is correct.
Age trends were apparent for which social self is superordinate when a
superordinate self emerged from the hierarchical classification. The self-
with-mother or the self-with-father was superordinate for seven students in
Grade 5, two in Grade 7, four in Grade 9, and none in Grade 11, X 2 = 9.3,
P < .05. This may mean that, for subjects for whom there is a superordi-
nate self, the self-with-parent is dominant during childhood and early
adolescence, but loses its eminence after this age. The self-with-best-friend
was the sole superordinate self for two subjects in Grade 5, three in Grade
7, one in Grade 9, and seven in Grade 11, X2 = 8.0, p < .05. This suggests
that the self-with-best-friend is psychologically the most important social role
for many of the oldest subjects with a superordinate self, but for few of the
younger subjects.
The average estimates for "How real I am" with mother, father, best
friend, and unknown kids, were respectively, 7.7, 7.0, 8.7, and 5.1, indicat-
ing that adolescents believe they are most real with their best friends and
least real with kids whom they do not know well. A repeated-measure
MANOVA indicated that the differences among these averages were signif-
icant F (3, 59) = 31.4, P < .001. Although grade was not a significant factor
(nor was sex) in the MANOVA, there was a significant correlation be-
tween age and the estimate for how real the subject is with the best friend,
r = .31, P < .01. The estimates for "How real I am" for the other three
roles were not significantly correlated with age.
To summarize, it is evident from the results of this study that the adoles-
cent's self-concept varies according to the social context in which it is con-
sidered. Among the social selves included in this study, it is the self-
with-best-friend that is most distinct from the general self, and it becomes
increasingly so with age. Further, the adolescent claims that the self is most
authentic with the best friend, and, finally, among older adolescents the
self-with-best-friend is likely to be superordinate.
Discussion
By considering the self-concept in its social context, our understanding
of the adolescent's sense of self is enriched in several ways. First, this
approach has implications for the other two lines of research (self-
evaluation and self-understanding) frequently followed in the study of
adolescence. The research by Gecas (1972) discussed above can serve as a
case example. Gecas demonstrated that there are commonalities and dif-
ferences in self-esteem across social contexts. The study described in this
chapter can be helpful in making sense of these findings. The similarities in
self-esteem scores across social roles are likely due to the same evaluation
of characteristics that are common to these roles, particularly positive per-
sonality traits and psychological characteristics. If the adolescent is proud
of being "friendly," and if "friendly" is attributed to several social selves,
then "friendly" is one source of the similarity of self-esteem scores in differ-
ent roles. Average differences between self-esteem scores for the different
roles can also be interpreted in light of the findings presented in this study.
For instance, in Gecas's study, the self-with-friend was higher in self-
esteem than was the self-in-the-classroom. The self-with-friend was de-
scribed by adolescents in the present study as possessing a relatively large
number of positive personality traits that are unique to the self in this role,
while the self-with-unknown-kids (the social context in this study closest to
the context of "in the classroom" used by Gecas) was described in terms of
negative personality traits only characteristic of the self in this social con-
88 Daniel Hart
text. Again, if positive personality traits are evaluated by the adolescent

positively, and negative personality traits are evaluated negatively, then
the higher self-esteem of the self-with-friend may be attributable to its
relatively high proportion of positive personality traits and the lower
self-esteem of the self-in-the-classroom could be due to the large number
of negative personality traits unique to that role.
This study also has implications for self-understanding research. The re-
sults confirm the existence of differences in self-understanding across social
contexts. The findings concerning the differences between social selves
found in the study described in this chapter are generally similar to the
findings reported by Volpe (1981) and Smollar and Youniss (1985). Unique
to the present study, however, are two findings concerning self-under-
standing. The first is that, using a new procedure that directly assesses
similarity between social selves, it was possible to determine that about
25% of the characteristics ascribed to the self are specific to the self in one
social relationship. This means that there appears to be a core self-concept
that pervades the various social selves. Methodologically, this finding
implies that attempts to elicit self-understanding without reference to a
specific social context are probably tapping this core, but may also elicit
characteristics that are likely to be specific to only one social context, such
as activities and communicative qualities. The interpretation of studies that
use social contextless questions for self-understanding should take these
results into account.
A second finding concerns the organization of social selves. An attempt
was made in this study to discern the organization among the social selves
of the adolescent, using a new analytical technique. The results of this
technique are interpretable within developmental theory, and indicate that
the self-with-mother and self-with-father social selves provide less orga-
nization to the confederacy of self-concepts in older adolescents, for whom
the self-with-friend is likely to serve the organizing role.
These findings can supplement the description of adolescent self-
understanding drawn from the model described in an earlier section of this
chapter. According to the model, at developmental Level 3, characterizing
early and mid-adolescence, self-understanding is embedded in a network
of social relationships. Adolescents consider the various characteristics of
self in light of their importance for social interaction with others, and their
influence on the selfs attractiveness to others. As indicated by the findings
from this study, however, the various relationships constituting the social
context influence the self-concept in different ways. For the younger
adolescents, it is the self-with-parents that is most likely to provide orga-
nization to the various social selves. Among older adolescents in Level 3, it
is the self-with-best-friend that serves this organizational role. Even within
the broad developmental levels identified by the model, then, there are
important changes in the organization of self-understanding.
The renewal of interest in the self-concept, for which this volume is one
testimonial, will lead to new theoretical and research approaches. The

purpose of this chapter is to demonstrate that these new theories and
methodologies might benefit by considering the self-concept in its social
context, an argument made almost 100 years ago by William James.
REFERENCES
Adelson, J., & Doehrman, M. (1980). The psychodynamic approach to adoles-
cence. In J. Adelson (Ed.). Handbook of adolescent psychology (pp. 99-116).
New York: Wiley.
Baldwin, J.M. (1902). Social and ethical interpretations in mental life. New York:
Macmillan.
Berndt, T. (1979). Developmental changes in conformity to peers and parents.
Developmental Psychology, 15, 608-616.
BIos, P. (1968). The second individuation process of adolescence. In R.S. Eissler et
al. (Eds.), Psychoanalytic study of the child (Vol. 23). New York: International
Universities Press.
infancy through adolescence. Child Development, 53, 841-864.
Damon, W., & Hart, D. (1986). Stability and change in children's self-
understanding. Social Cognition, 4, 102-118.
Damon, W., and Hart, D. (in preparation). The development of self-understanding.
New York: Cambridge University Press.
DeBoeck, P. (1985). H1CLAS Computer Program: Version 1.0. Katholique Uni-
versiteit Leuven, Psychology Department, Leuven, Belgium.
DeBoeck, P., & Rosenberg, S. (1985). Hierarchical classes: Model and data analy-
sis. Unpublished manuscript, Katholique Universiteit Leuven, Leuven, Belgium.
Elkind, D., & Bowen, R. (1979). Imaginary audience behavior in children and
adolescents. Developmental Psychology, 15,38-44.
Epstein, S. (1981). The unity principle versus the reality and pleasure principles, or
the tale of the scorpion and the frog. In M.D. Lynch, A.A. Norem-Hebeisen, &
K. Gergen (Eds.), Self-concept: Advances in theory and research (pp. 27-37).
Cambridge, MA: Ballinger.
Gecas, V. (1972). Parental behavior and contextual variations in adolescent self-
esteem. Sociometry, 35, 332-345.
Grubb, P.D. (1986). The self as a mUltiplicity: An empirical analysis of identity
structure theory. Unpublished doctoral dissertation, Rutgers University, New
Brunswick, NJ.
Hart, D. (1986). Adolescent perceptions of past, present, and future selves. Manu-
script submitted for publication.
Hart, D., & Damon, W. (1986). Developmental trends in self-understanding.
Social Cognition, 4,388-407.
Hart, D., Lucca-Irizarry, N., & Damon, W. (1986). The development of self-
understanding in Puerto Rico and the United States. Journal of Early Adoles-
cence, 6,293-304.
Hart, D., Maloney, J., & Damon, W. (1987). The meaning and development
of personal identity. In T. Honess and K. Yardley (Eds) , Self and identity
(pp. 121-133). London: Routledge and Kegan Paul.
90 Daniel Hart
Harter, S. (1982). The perceived competence scale for children. Child Develop-
ment, 53, 87-97.
Hartup, W. (1983). Peer relations. In M. Hetherington (Ed.), Handbook of child
psychology (4th ed.) (Vol. 4, pp. 103-196). New York: Wiley.
James, W. (1950). The principles of psychology (Vols. 1 and 2). New York: Dover.
Joseph, C. (1985). Identity patterns of professional women. Unpublished master's
thesis, Rutgers University, New Brunswick, NJ.
Kuhn, M., & McPartland, T. (1954). An empirical investigation of self-attitudes.
American Sociological Review, 9,68-76.
Lecky, P. (1945). Self-consistency: A theory of personality. New York: Island Press.
Livesly, W., & Bromley, D. (1973). Person perception in childhood and adoles-
cence. New York: Wiley.
Martindale, C. (1980). Subselves: The internal representation of situational and
personal dispositions. In L. Wheeler (Ed.), Review of personality and social
psychology (pp. 193-218). Beverly Hills: Sage.
McGuire, W., & Padawer-Singer, A. (1976). Trait salience in the spontaneous
self-concept. Journal of Personality and Social Psychology, 33, 743-754.
McPhail, c., & Tucker, C. (1972). The classification and ordering of responses to
the question "Who am I?". The Sociological Quarterly, 13,329-334.
Mead, G.H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Montemayor, R., & Eisen, M. (1977). The development of self-conceptions from
childhood to adolescence. Developmental Psychology, 13,314-319.
Nowicki, S., & Strickland, B. (1973). A locus of control scale for children. Journal
of Consulting and Clinical Psychology, 40, 148-154.
O'Brien, R., & Kaiser, M. (1985). MANOVA method or analyzing repeated mea-
sures designs: An extensive primer. Psychological Bulletin, 97,316-333.
Rosenberg, M. (1979). Conceiving the self. New York: Basic.
Rosenberg, S., & Gara, M. (1985). The multiplicity of personal identity. In P.
Shaver (Ed.), Review of personality and social psychology (Vol. 6, pp. 87-113).
Beverly Hills: Sage.
Secord, P., & Peevers, B. (1974). The development and attribution of person con-
cepts. In T. Mischel (Ed.), Understanding other persons (pp. 87-113). Oxford:
Blackwell.
Smollar, J., & Youniss, J. (1985). Adolescent self-concept development. In R.
Leahy (Ed.), The development of self (pp. 247-265). New York: Academic
Press.
Volpe, J. (1981). The development of concepts of self: An interpersonal perspec-
tive. In J. Meacham & N.R. Santilli (Eds.), Social development in youth: Struc-
ture and content (pp. 131-144). Basel, Switzerland: Karger.
Youniss, J. (1980). Parents and peers in social development. Chicago: University of
Chicago Press.
5
The Embedded Self:
I and Thou Revisited
RUTHELLEN JOSSELSON
Are we not formed, as notes of music are,

For one another, though dissimilar.
-Percy Bysshe Shelley
Psychoanalytic developmental theory takes as its premise that the central
thrust of human development is movement from a state of dependence and
merger to a state of independent, differentiated selfhood. When we think
about the self, or identity, we are inclined to envision a person standing
alone, somehow being what he or she is, apart from all others. Our quest as
theorists has similarly been for a conceptualization of selfhood as a purely
internal function, as though one can have selfhood, or identity, indepen-
dent from embeddedness in a social matrix.
Bakan (1966) points out that it is in the nature of the agentic side of
human nature to separate itself, often forcefully, from its communion
aspect. Thus, the agentic view of human development stresses the attain-
ment of agentic forms: self-assertion, mastery, self-expansion, distinction,
and separateness. Communion, in its striving for contact, union, coopera-
tion, and overcoming of separateness tends to be repressed and disowned
by our agentic theories.
The effort in developmental psychology has been to view human growth
as proceeding from dependence to autonomy and to account for the proc-
esses by which more and greater aspects of experience are taken over by
the self. Individual developmental history is a movement from merger to
separateness. The metaphorical picture of the "lone self," has, however,
been contradicted by careful study of the phenomena. Human beings are
inexorably embedded in and with others, even in defining their selfhood.
Two areas of study have particularly highlighted the difficulties and com-
plexities that any theory of the self or of identity must take into account.
One has been the effort to understand adolescent development. The other
has been the effort to map female development. Study of these populations
has resisted theoretical efforts to describe an increasingly separate self. In
adolescents and in women, the self does not precipitate out of social
embeddedness-the self precipitates within a social matrix. At the same
92 Ruthellen losselson
time, new theoretical approaches have been advanced that attempt to inte-
grate a study of the self with the importance of connection to others. This
chapter reviews the enigmas and then considers the theoretical approaches
available to conceptualize these phenomena.
Let us consider first the problem of adolescent development. Any
theory of adolescence must account for the most obvious phenomenon:
that adolescence is a period in which the child metamorphoses into an
adult, that is, moves away from dependence on parents to self-sufficiency
and independent life and also assumes adult expression of sexuality.
Theories have had the task of accounting for the internal reorganizations
of the self that permit the emergence of adult autonomy out of the con-
stellation of childhood dependence. Adolescence is thus seen as a total
reworking of the structure of the self, as the adolescent detaches and dis-
engages from the objects of childhood, rebels against parents in search of
autonomous will, becomes overinvolved with peers to support an impover-
ished ego, and invokes rigid defenses to shore up a now-fragile ego
organization-all this so that a new ego can be consolidated at a great
distance from the original objects (Blos, 1962; A. Freud, 1958). These for-
mulations make a good deal of intuitive sense, except for the fact that none
of it happens phenomenologically, at least not to the majority of adoles-
cents. After 20 years of careful, thorough research into normal populations
of adolescents, researchers have simply failed to find consistent evidence of
detachment, disengagement, rebellion, overinvolvement with peers, ego
weakness, rigid defenses, or other manifestations of Sturm und Orang
among healthy, well-developing adolescents (Douvan & Adelson, 1966;
Offer, 1969; Offer, Ostrov, & Howard, 1981).
What has been found is that the vast majority of adolescents maintain
harmonious, loving relationships with parents throughout adolescence,
that those adolescents showing the highest self-esteem and most mature
functioning tend to have the most interrelationship with their parents, that
peers take on added importance but do not supercede the influence of
parents in important decision making-in short, that attachment to parents
continues unabated throughout adolescence and into adulthood (Fischer,
1986; Offer & Offer, 1975; Symonds & Jensen, 1961).
A similar conundrum had obtained in the study of women. While most
developmental theories, from Freud until fairly recently, purported to be
universal theories, it is becoming increasingly clear that these theories illu-
minate phenomenology unique to men. Again, it is the development-as-
separation model that founders and falls. Development is seen to move
toward increasing autonomy, culminating in the image of the corporate
president, the man at the top, who runs his own show, beholden to no one,
independent, self-actualizing. Women, because they attain less clearly
bounded selfhood, caught up as they are in caring for and responding to
others, are seen implicitly if not explicitly to be somehow less mature.
Erikson's (1968) concept of identity has been problematic in this regard.
5. The Embedded Self: I and Thou Revisited 93
Having men in mind, his epigenetic chart stresses that the attainment of
identity is a precondition for the foray into intimacy. Only with a reason-
ably delineated sense of identity can one undertake the merging of identi-
ties that is intimacy without too much risk of the dissolution of the self.
Erikson's case histories, all men, depict the anguish of striving for a place
in the world that betokens identity. Generally, such striving involves a
quest for meaningful work and for the attainment of ideology to organize
experience, such that the person has a sense of competence and influence
in a world that "makes sense" to him. The intuitive logic of Erikson's pre-
sentation has the ring of truth. Yet when we have applied his theoretical
framework to the study of women, to investigate their identity, what
women tell us about is their relationships. Focus on issues of industry,
ideology, and autonomy are secondary, if they are to be found at all.
What is common to the effort to understand adolescence and the effort
to understand women is that the search for the "lone self" has led to the
incontrovertible evidence of the ways in which the self is bound to others.
Our effort to find the separate self is stymied by the fact of the embedded
self. The deepest, most organizing aspects of the self seem to include some
aspects of what Buber (1970) calls the "I-Thou." Adolescents grow with-
out detaching themselves from parents. Women are adult and have identity
in the context of a relational web.
Separation and Connection

Separation-individuation theory has become a dominant paradigm for our
new developmental psychology. Development is powered by increasing
separation-individuation, from earliest infancy to old age. From Mahler's
explication of the hatching of the infant into an experientially separate
small human being, the motif of separation-individuation has been im-
posed on later developmental stages. BIos (1967), for example, discusses
adolescence as a "second individuation phase." After a quiescent latency
period, where homeostasis obtains between parent and child, adolescence
brings with it a need to further differentiate. Individuation, in this second
phase, promotes autonomy from the parents. The boundaries of the self
are differentiated from parents and thereby highlighted and sharpened.
Both outer separation in the form of less dependence on parents and inner
separation in the form of less power to parental introjects take place. The
heightened sense of distinctness from others during adolescence bequeaths
to the young person the heightened sense that his choices are his own.
Separation and its companion, individuation, have become key concepts
for describing the development that takes place during adolescence. So
many theorists and researchers have adopted this paradigm that there is
already much terminological muddying (Josselson, 1980), with some peo-
ple using individuation to stand for intrapsychic events, and others under-
94 Ruthellen Josselson
standing individuation to describe aspects of family systems wherein the

adolescent is permitted separate selfhood. For our purposes at the mo-
ment, it doesn't matter in which sense we take this term. What is critical is
that agreement obtains that movement in the direction of autonomy, self-
direction, a sense of responsibility for the self, and disentanglement from
parental control, in short, separation of the self out of the more or less
enmeshed child-parent web is what adolescence is all about. Here again,
under the surface, is the familiar developmental assumption. Development
proceeds from dependence to autonomy. To grow is to separate.
As theorists, we then follow the course of this now "separated" indi-
vidual. Once it has served its presumed developmental function, the rela-
tionship that was separated from has no further interest to us. It vanishes,
or should. I recall a case conference at a college counseling center in which
a case was reported of a depressed student. Speaking of the patient's cur-
rent life, the therapist mentioned that the student called home every week
to speak to his parents. During the case discussion, the psychoanalysts in
the group commented on the importance of a failure to adequately sepa-
rate in this case, supported by the evidence of frequent phone calls home.
Presumably, theory predicts that an adequately separated or individuated
adolescent should have no further interest in his parents.
Much of our understanding of separation-individuation has come from
clinical consideration of patients in whom an inability to appropriately
separate from parents precludes further growth or results in psychopatho-
logical symptom formation. As a result, we have come to appreciate the
critical importance of separation for psychological health. But there is a
continuum that underlies separation and individuation: too much or too
little presents dangers.
The general trend toward misunderstanding of the separation concept
warrants rethinking and reflection. Psychological separation-individuation
does not require that relationships be obliterated in the interest of gaining
autonomy; rather, separation modifies relationship. Separation-individua-
tion is one side of a matrix that connects individuals. When we look at the
separation side, we see individuals moving away from someone. But when
we turn the matrix over to view its other side, we see the separating indi-
vidual revising, and thus preserving the relationship.
Mahler's (Mahler, Pine, & Bergman, 1975) concept of rapprochement
can be similarly viewed from both sides. In Mahler's original sense, rap-
prochement represents a refueling, where the individual returns to the re-
lationship to obtain more "fuel" for autonomy. Another way of looking at
this process is that the separating toddler is returning to mother to bring
her along. The toddler's rapprochement with mother may not be only for
what he or she needs from mother, but to engage and share with mother to
ensure their continuing connection.
Rapprochement is a powerful concept for understanding development
because it blends concern with autonomy in the context of relatedness. We
become selves within, not in spite of, relationship. Rapprochement is

about preserving bonds of relationship in the presence of increasing
autonomy. This is as true in adolescence as it is in infancy. The adoles-
cent, as much as the toddler, brings his new ideas and his new ways of
being home, to be recognized in the context of ongoing connection, to
bring the relationship up to date.
Attachment is not the opposite of separation-individuation-it is coinci-
dent with it. Among infants, a strong attachment to mother facilitates a
child's autonomy. It is the securely attached children who are most free to
explore the world (Ainsworth, Blehar, Waters, & Wall, 1978). Research
evidence has shown this to be equally true of adolescents. Mature, well-
adjusted, competent adolescents report strong attachment to and loving
relationships with their parents (Frankel & Dullaert, 1977; Hamid & Wyl-
lie, 1980; Offer & Offer, 1975). Similarly, other research has shown that it
is both individuality and connectedness that are adaptive to adolescent
development (Cooper, Grotevant, & Condon, 1983).
Attachment behavior, the observable signs of the affective experience of
relatedness, varies across the life span, even within a developmental stage.
It is easiest to observe in a controlled way among infants and young chil-
dren because the behavioral repertoire is more limited. As the child
matures, we become less clear about which behaviors are to be construed
as evidence of attachment, but the child becomes better able to verbalize
her affective experience of connection.
If one talks to adolescents (rather than having them respond to predeter-
mined questionnaires), their connectedness to others is immediately appa-
rent. The people whom they are attached to are the people they think
about and talk about, the people that are "with" them internally. It is
difficult for the early and middle adolescent to voice an opinion without
immediately and spontaneously contrasting it with a parent's viewpoint.
Even if it is only to complain about them, the adolescent has her parents
with her at all times (Josselson, Greenberger, & McConochie, 1977a,
1977b).
As the adolescent comes to experience more freedom about choices for
the self, she consolidates her individuality with increasing independence
from her parents. She may choose to do things differently. But this does
not imply the relinquishment of her bonds with her parents. Rather, it may
become all the more important to her that her parents accept the new
aspects of her selfhood. Much of the pain of adolescence is in this effort at
rapprochement, the fear of putting separation-individuation and rela-
tionship at odds. That is, the adolescent who fears parental disapproval for
a choice she may make worries that she may damage, irreparably, the rela-
tionship. At what point does autonomy put a valued relationship asunder?
How flexible is the other to whom we are attached to accompany us on our
way? Relationships and bonds can be damaged, and neither the adolescent
nor his or her parents want to break the bonds. The late-adolescent rap-
96 Ruthellen 10sselson
prochement phase is a dance of discovery about where the limits of the

bond are.
As the late adolescent forms new connections to other people, she con-
tinues to pull along the old attachments. New relationships, although they
are formed out of a separated and individuated part of the self, do not
annihilate the old ones. The adolescent brings her friends "home." This
blinging home is the essence of rapprochement and determines the course
that identity formation will take. When new investments are made in
adolescence, it is not a simple matter of whether parents approve or dis-
approve, but whether the adolescent can experience some continuity of the
self between his or her new context and the old.
Tolstoy (1961) offers a superb example of this in the adolescent character
of Kitty in Anna Karenina. Rejected by the man she had hoped to marry,
Kitty falls neurotically ill, unable to envision any longer a future for her-
self. Sent off to a European sanatorium to recover, Kitty undergoes what
we might now consider to be a Moratorium phase. Meeting people unlike
those she had ever encountered before, she becomes passionately attached
to two women who embody saintlike expressions of piety and devotion to
others. Following their model, Kitty, too, begins to devote herself to tak-
ing care of others. Unfortunately for Kitty, she does such a good job of
taking care of one man that his wife becomes jealous and sends her away.
This leads Kitty to question her own motives, to wonder if she were not
just pretending, and to ponder what her ideals and values really are. She
experiences an identity crisis. Still, she tries to hold to her new ideals, to
make them part of herself, and to put them before her father when he
arrives for a visit. Eagerly, she hopes to show him her new self.
But with her father's return, all the world in which she had been living was
completely changed for Kitty. She did not renounce all that she had learned, but
she realized that she was deceiving herself in imagining that she could be what she
wanted to be. She seemed to have recovered consciousness; she felt without hypoc-
risy or boastfulness the whole difficulty of remaining on the heights to which she
had wished to rise; in addition, she felt the heavy burden of that world of grief,
sickness and death in which she was living; the efforts she had made to force herself
to love it seemed appalling to her now, and she longed to get back quickly to the
fresh air, back to Russia .... (p. 245)
What Kitty took away from her experience was a heightened sense of
what she had been, an ability to value aspects of her life "at home" that
had become meaningless to her. Her father brought with him his awareness
of Kitty's continuous self. Without his saying anything, just his being who
he always was counterpointed for Kitty the falseness of her new commit-
ments. She returned to Russia "completely cured ... calm and serene." In
short, she had tested and formulated an identity through a process of rap-
prochement.
Identity must be experienced in a social context in that it represents the
individual's sense of what he or she stands for in the world. Of course, we

are free to choose the social context that we wish to respond to us-to an
extent. Affectional bonds function as givens and as such delimit the choices
one may make. Both in terms of internal and external reality, individuals
stray far from home only at some psychological peril.
But our developmental psychology has tended to hold the value of limit-
less possibility for selves that might be. As teachers and therapists, we
counsel adolescents to dream big. We assure them that they can be what-
ever they want to be. Kitty's realization, though, was that she could
not be whatever she wished to be. She was already a self, already had an
identity that resulted from her previous experiences that would not mesh
with the new idealized self she wished to become. She was free to explore
the new alternative self because she had deposited her old self in the keep-
ing of her father, safe in the knowledge that he would hold it for her until
she knew if she would be able to live without it or not. This sort of holding
of old selves is what so many adolescents need from their parents, indeed
what so many even ask for from their peers. Like a comfortable old pair of
shoes, old selves must have somewhere to rest until the new ones fit as
well. Autonomy grows only in the context of connection.
As adolescents take increasingly more adult responsibility in the world,
it also becomes important to them to revise their relationships with pa-
rents, more as "man-to-man" or "woman-to-woman." Bonds change from
dependence to interdependence, from submission and rebellion to inter-
relatedness. As selfhood becomes more defined, relationships deepen and
differentiate. Connection grows in the context of autonomy.
What happens to adolescents who do not maintain or who never had
good enough connection with their parents, where there is no meaningful
relationship to separate within? These adolescents seem to be particularly
vulnerable to what Erikson (1968) calls identity diffusion and Kohut (1977)
calls the narcissistic fragmentation of the self. Without meaningful early
bonds of relatedness, there comes a concommitant failure of internaliza-
tion, such that the young person has few important introjects to sift during
adolescence. As a result, they remain at the mercy of their impulses,
resigning themselves to a world that makes sense only as fleeting pleasure
and pain. Such people are in search of someone or something to make
sense for them rather than to make sense to them. They are available to
be claimed by something, if they are fortunate to be in the wake of some
powerful other or idea that promises to structure reality for them.
A follow-up study of eight identity-diffuse college senior women showed
that all of them either remained diffuse in identity 10 years later or had
found a totalistic solution to identity (Josselson, 1987). That is, those who
had settled identity issues by age 32 had done so by subsuming themselves
to someone or some institutional structure that settled all questions of how
to live. At its extreme, one woman had joined an Eastern religious cult
where her spiritual leader decided all matters in her life. Looking at the
early psychodynamic histories of these identity-diffuse women, it was clear

that these were women who had not formed strong relational ties before
adolescence. Having parents who demanded or expected little of them,
there was little to internalize. In contrast to the women who were not diffuse
in identity, these women had little emotional sense of their families. They
had psychologically disconnected long ago. Without the context of related-
ness, separation-individuation, and with it, identity formation, could not
occur (Josselson, 1973).
Separation-individuation, then, is not a moving away from but an elabo-
rate pas de deux in which the developing individual moves in such a way as
to attempt to effect the degree of autonomy that she needs within the form
of relatedness that she wishes. Adolescence is a long period of negotiation
between an adolescent and her important inner and outer objects to find a
balance that is mutually tolerable. Separation-individuation implies con-
tinued, renewed, often strengthened, but revised connectedness. A more
clearly delineated sense of self makes new forms of relatedness possible.
Because separation-individuation and relatedness are two sides of the
same matrix, to focus on one at the exclusion of the other distorts the
understanding of the process. But we have more words to describe auton-
omous, independent functioning, and few terms for relatedness or attach-
ment. Every psychological theoretician who has confronted this problem
has been moved to comment on the lack of terminology for this most im-
portant aspect of human existence.
Separation in the Study of Women

The search for the autonomous self as the hallmark of maturity and identi-
ty equally washed ashore in the study of women's development. From the
time that researchers began to pay serious attention to females as different
developmentally from males, they were bedeviled by the dominance of
interpersonal focus among women. After studying thousands of adoles-
cents, Douvan and Adelson (1966) concluded,
"What we did not anticipate was the force of the sex variable, the extent to which it
defines and shades all aspects of the developmental crisis .... The key terms in
adolescent development for the boy in our culture are the erotic, autonomy and
identity. For the girl the comparable terms are the erotic, the interpersonal and
identity (p. 347)."
Where autonomy is a developmental organizer for boys, the researchers

found interpersonal focus for girls.
Freud recognized that women are more embedded in relationships and
regarded them, as a result, as morally inferior. Gilligan (1982), after study-
ing the way in which women approach moral dilemmas, argued that
women construct the world differently from men. "Sensitivity to the needs
of others and the assumption of responsibility for taking care lead women
to attend to voices other than their own and to include in their judgment
other points of view" (p. 16). Because women define themselves in a con-
text of relationship, a developmental orientation that equates growth with
autonomy will automatically relegate women to lower rungs of develop-
ment. Swidler (1980), examining cultural myths of adulthood, points out
that American culture finds it hard to make the achievement of adult com-
mitment, fidelity, intimacy, and care seem meaningful and heroic. With the
rise of feminist scholarship, women have presented a difficult challenge to
the autonomy-separation-achievement model of adult development.
The effort to investigate identity in women similarly finds that women
define themselves in the context of important relationships, even women
who are high-achieving in careers (Cox, 1970; Josselson, 1987). They
experience themselves as most deeply themselves in relation to those
they love and for whose well-being they take responsibility. This has led
theorists and researchers to wonder if, perhaps, the Eriksonian identity
and intimacy stages are merged for women (Lewittes, 1982; Miller, 1984;
Morgan, 1982).
What Erikson had in mind in the intimacy stage was the making of a
committed heterosexual relationship. Following his own logic, he later
considered the possibility that a woman delays identity consolidation until
the arrival of the man by whom she wishes to be sought. Erikson thought
he could include women in his theory by this slight modification. But
Erikson's amendment misses the point. Intimacy, or interpersonal
development, among women is identity and resides not in the choice of a
heterosexual partner, but in the development, differentiation, and mas-
tery of ways of being with others (not just men) that meet her standards for
taking care, that connect her meaningfully to others, and that locate her in
an interpersonal network.
Female development proceeds on an interpersonal track that is not rep-
resented in Erikson's scheme. Consider the industry stage, for example, a
stage devoted to skill development and the growth of competence. An im-
portant area of competence for girls, however, is the art of "getting along,"
of discovering the attributes, attitudes, and behaviors that are valued by
others, that promote her desirability as someone for others to engage with,
and that increase her capacity for harmony and pleasure in relating to
others.
The vast differences in women's ways of making sense of the world call
for a redefinition of identity in which it is possible to conceive of a self
intertwined with others. Miller (1976) points out that "women develop in a
context of attachment and affiliation with others [such that] their sense of
self becomes very much organized around being able to make and then to
maintain affiliations and relationships" (p. 83). For women, loss of an im-
portant relationship is often experienced as a loss of the sense of self. In
order to effect a redefinition of identity that will accurately encompass
100 Ruthellen losselson
women, a major shift must be made from separation to relatedness as the

basis for self-experience (Surrey, 1984). Maturing then becomes a progres-
sion toward the deepening capacity for relationship, where assertion and
autonomy are subordinate aspects of the self. This turns current concep-
tions of identity inside out. In the major developmental studies of men
(Levinson et aI., 1978; Vaillant, 1977), relationship is clearly the subtheme
or counterpoint to the real developmental story of advances in autonomy
and personal achievement. To turn this around and place relatedness at the
center of developmental psychology is so radical that we can barely begin
to conceptualize it.
Chodorow (1978) maintains that because women are mothered by some-
one of the same sex, they never individuate as much as men. In the place of
greater individuation, however, comes greater capacity for empathy, sensi-
tivity, and responsiveness to others. Far from viewing the path of women's
development as lacking in separation, Miller (1976) concludes that the
greater capacity for relatedness in women is a developmental achievement
higher than autonomy, representing a different and more advanced ap-
proach to living.
New Theoretical Outlooks

Recent theoretical advances in the understanding of the growth of the self
have taken more account of the importance of relatedness. In these
theories lie the bellwether of a fundamental revision of our understanding
of human growth. To adolescents and women, relationship is so compelling
an aspect of existence that we, as theorists, cannot mistake it for something
else. But it may be just as important in the development of men, of all
human beings, where phenomena of connection and bonding may have
lain hidden in the crevices of our imperfect language. Perhaps develop-
ment is not a path from dependence to autonomy, but a movement to
increasingly differentiated forms of relating to others. Perhaps autonomy is
merely a form of relatedness.
Fairbairn (1952) and the British object-relations school radically revised
Freud's central idea that objects are necessary for drive gratification, that
drive propels social life. Fairbairn's assumption was that human need for
other humans is independent from drive, that humans are born object
seeking, and that object ties are gratifying in and of themselves. Harlow's
famous experiments with rhesus monkeys provided experimental evidence
to confirm this view. Monkeys raised with surrogate, milk-dispensing
"mothers" grew up to severe disturbances, while monkeys deprived of a
mother but raised with peers did not (Harlow, 1974).
In classical psychoanalytic reasoning, tied to a drive-reduction frame-
work, the human adult is object seeking for biological need gratification.
From the object relations point of view, in which the quest for object re-
latedness is innate, people need each other because human contact is itself
gratifying. Sullivan made the interpersonal world central to his understand-
ing of the self-system. He understood that" ... loneliness in itself is more
terrible than anxiety" (1953, p. 261). The need to seek relief from loneli-
ness drives one to seek proximity and intimate exchange with other fellow
humans, a spur to development more powerful than libido.
Here is where we encounter the language problem. That people need
each other, seek each other out, and feel gratified by contact with others is
taken for granted but is experienced in ways not well articulated. The
words that our language makes available to describe the vagaries of deeply
sensed "object ties" are abstruse and vague, words like attachment, affilia-
tion, connectedness, communion, bonding, or that most ill-defined word of
all, love. These words are affectively connotative and we intuitively recog-
nize their importance. They suggest warmth and diffuse pleasure. (Words
such as intimacy and desire are stronger but, being closer to the passions,
are also closer to the drives and carry more anxiety.) When we think of
relatedness, we think of the well-being of the self. When we feel attached,
we feel a part of, not as in merger, but as in having a place in something
larger than ourselves. We feel not alone.
Bowlby's (1969,1973) extensive work on attachment has, more than any
other, developed this "other side" of human experience and has laid a
foundation for beginning to trace the history of the self as it grows in rela-
tion to others.
"Intimate attachments to other human beings are the hub around which a person's
life revolves, not only when he is an infant or a toddler or a schoolchild but
throughout his adolescence and his years of maturity as well, and on into old age.
From these intimate attachments a person draws his strength and enjoyment of life
and through what he contributes, he gives strength and enjoyment to others (1969,
p.442)."
Bowlby's work, relying on experimental evidence, shows clearly that

attachment and autonomy are not opposed. Self-reliance is most evident in
a context of secure attachment. If you wish someone to be autonomous,
see that they are securely attached. Thus, autonomy is not the antithesis of
relatedness. Rather, autonomy is a property of it. When we feel securely
embedded with another, we feel freest to express ourselves. And we feel
certain that, having expressed ourself, our connections to others will be
enriched. The obverse here is also our certainty that our other's self-
expression will enhance us.
Attachment, in Bowlby's formulation, is not only a constant during the
life cycle, but has important consequences for the formation of the self.
The self is experienced as an object of others' attachment. Both our valua-
tion and our experience of our selves is rooted in what we mean to others.
After childhood dependence on others for satisfaction of needs that the
child cannot gratify himself, attachment behavior persists. At this point, it
is no longer the continued presence of the object of attachment that is

critical, but the availability and responsiveness of the object.
"Confidence that an attachment figure is, apart from being accessible, likely to be
responsive can be seen to turn on at least two variables: a) whether or not the
attachment figure is judged to be the sort of person who in general responds to calls
for support and protection; b) whether or not the self is judged to be the sort of
person towards whom anyone, and the attachment figure, in particular, is likely to
respond in a helpful way. Logically these variables are independent. In practice
they are apt to be confounded. As a result, the model of the attachment figure and
the model of the self are likely to develop so as to be complementary and mutually
confirming (1973, p. 204)."
That is, the unloved person is likely to decide that he is unlovable, the
unwanted person likely to view himself as undesirable.
Winnicott (1965) stressed that "good enough" object relations are neces-
sary for the sense of self to develop. Kohut has similarly come to the
ongoing importance of others for the definition of the self. We live in a
world of selfobjects, says Kohut (1977), where the selfobject is an aspect of
another person that serves to maintain our sense of ourself. The self is not
only enriched, but its very existence is stabilized by the existence of others
who confirm the experience of the self. We need others to be ourselves.
Our experience of selfhood rests on a sense of fit between self and the
social world, on the expectation that the environment will, at least much of
the time, be in tune with us.
In Kohut's depiction of selfobjects lies the concept of the importance of
others for validation of the self. And what is most in need of validation are
those aspects of the self that lie nearest the core, the nuclear self, in
Kohuts' terminology. Others, then, can come to feel like part of the self,
although we may also recognize their separateness. This blurring of boun-
daries is an aspect of psychological health rather than pathology. The
cohesive self develops out of the ground of responsive relationships subjec-
tively experienced as soothing and meeting the self's needs for stabilizing.
Even as adults, people are vulnerable to disruptions in the sense of self
when placed in foreign environments where formerly reliable aspects of the
self go unacknowledged (Wolf, 1980). Looking at clinical data from his
extensive experience in treating adolescents, Masterson (1985) similarly
concludes that the "real self" grows as a function of adequate responsive-
ness from and sharing with others.
A number of writers, then, have discussed the need for object relations
for the self to develop, to firm, and to maintain. Metaphysically, however,
this view assumes that the purpose of objects is to ensure and regulate the
self. In psychoanalytic terms, objects are necessary in order to provide
good introjects. Perhaps, however, the converse is also valid. Perhaps in-
creasingly differentiated selfhood occurs in order to promote ever more
complex and rewarding interrelationships with others.
This latter construction is central to Stern's (1985) reformulation of in-

fant development. Stern inverts Mahler's way of looking at infants. Mahler
takes as a premise that infants are born "merged"; therefore, development
entails separation and individuation. Stern, by contrast, posits that infants
are born already separate with a preformed sense of self. The developmen-
tal task, then, in his view, is the achievement of relatedness. Going beyond
Winnicott, Fairbairn, and Kohut, Stern proposes that others do not merely
promote development of selfhood, but provide objects for increasingly
complex "intersubjective union." Needs for sharing of experience may be
basic and primary. If the fact of separation rather than the hypothesis of
merger defines the starting point, development becomes a process of dis-
covering new forms of being-with-others. Stern stresses the importance of
such interpersonal events as attunement and emotional sharing as develop-
mental necessities and achievements for the developing self.
Kegan's (1982) conception of the evolving self takes a dialectical move-
ment between autonomy and relatedness as its basis for increasing dif-
ferentiation of the self. The self grows within a culture of embeddedness,
where others provide reciprocity, gratification, and impetus for continued
growth. Kegan cautions that developmental psychology must equally re-
spect both distinctness and inclusion, and attempts a model of develop-
ment where maturity resides in interweaving these threads. Stern's model,
however, goes beyond oscillating sequential phases of work on self and
relational issues. He believes that separation and relatedness are concom-
mitant and equal developmental lines.
Existential writers (Fromm, 1963; Maslow, 1968; and others), whose
work has remained unintegrated with developmental psychology, have
brought much wisdom to bear on the role of relatedness in human life.
Starting from the premise of the fact of existential isolation, relationship is
a means of denying and overcoming the terrifying fact of separateness. By
being "with" others, we can forget, even for a moment, that we are ulti-
mately and irreparably alone.
In his reflections on existential aloneness, Yalom (1980) describes a schi-
zoid, interpersonally isolated man who asks during a therapy session,
"What are people for, anyway?" It might be instructive to ponder how our
current developmental psychology answers this question. As our para-
digms tell it, people are for the satisfaction of need. For infants, people are
for providing physical sustenance and emotional responsiveness such that
the infant develops an expectation of its needs being satisfied and an
awareness of its own separate existence. For children, people are "for"
protection and socialization, for teaching the child what he or she will need
to know in the culture. For adolescents, parents are "for" having some-
thing to separate from; peers are people to help the adolescent do this. For
adults, people are for sexual gratification and ensuring reproduction.
Within developmental psychology, we do not yet have an adequate
theory of human bondedness, including friendship or the varieties of
intimacy. Nor can we conceptualize the ways in which such relatedness

makes us feel most ourselves. One of the most interesting of human phe-
nomena can be observed at airports. When I pass through the landing gate,
there is a sea of faces, all eagerly scanning the faces of the passengers,
passing over mine fleetingly, disinterested. It is not I who am sought. One
by one the faces light up, all in exactly the same way. It is a look of joy and
the warm pleasure of attachment. There is a broad smile and a special
gleam in the eye, reciprocated by the person toward whom the look is
directed. When that special gleam is meant for me, I have a heightened
sense of myself and of my meaning in the world. We will have an adequate
psychology of the self when we can conceptualize, among other things, the
way in which people are "for" gleams in the eye.
The psychological study of groups, both psychotherapy groups and fami-
lies, has fostered our understanding of self experience in a social matrix.
We have come to understand the self-destroying dangers of enmeshment,
where a self may be so undifferentiated from others that self-experience is
all but nonexistent. At the other extreme are the dangers of isolation, a
distance from others so great that self-experience becomes empty. Most
social experience takes place somewhere in between.
In the social world there is always a dialectic between needs for separ-
ateness and needs for closeness. The course of the growth of the self lies in
eternal tension between needs for inclusion and needs for autonomy. Anxi-
ety waits at each pole: at the one, the dread of merging and identity loss; at
the other, the fear of exclusion and aloneness. Human development is an
ongoing effort to affect a balance between these needs and and between
these fears. A theory of the self, or of identity, must be a theory of such a
balance, must account for the ways in which the self remains poised be-
tween self-expression and relatedness, between the need for self-assertion
and social involvement. A theory of the self must, therefore, be inter-
woven with a theory of relationship.
REFERENCES
Ainsworth, M.D., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of
attachment. Hillsdale, NJ: Erlbauffi.
Bakan, D. (1966). The duality of human existence. Boston: Beacon Press.
Bios, P. (1962). On adolescence. New York: Free Press.
Bios, P. (1967). The Second individuation process of adolescence. Psychoanalytic
Study of the Child, 22, 162-86.
Bowlby, J. (1969). Attachment and loss: Vol. I. Attachment. London: Hogarth
Press.
Bowlby, J. (1973). Attachment and loss: Vol. II. Separation. New York: Basic
Books.
Buber, M. (1970). I and thou. New York: Scribner.
Chodorow, N. (1978). The reproduction of mothering. Berkeley: University of
California Press.
Cooper, c.R., Grotevant, H.D., & Condon, S.M. (1983). Individuality and con-
nectedness in the family as a context for adolescent identity formation and role-
taking skil!. In H.D. Grotevant & C.R. Cooper (Eds.), Adolescent development
in the family. San Francisco: Jossey-Bass.
Cox, R.D. (1970). Youth into maturity. New York: Mental Health Materials
Center.
Douvan, E. & Adelson, J. (1966). The adolescent experience. New York, Wiley.
Erikson, E. (1968). Identity, youth and crisis. New York: Norton.
Fairbairn, W.R.D. (1952). Psychoanalytic studies of the personality. London:
Tavistock Publications.
Fischer, L.R. (1986). Linked lives. New York: Harper & Row.
Frankel, J., & Dullaert, J. (1977). Is adolescent rebellion universal? Adolescence,
12,227-236.
Freud, A. (1958). Adolescence. Psychoanalytic study of the child, 16, 225-278.
Fromm, E. (1963). The art of loving. New York: Bantam Books.
Gilligan, C. (1982). In a different voice. Cambridge: Harvard University Press.
Hamid, P.N., & Wyllie, A.J. (1980). What generation gap? Adolescence, 15, 385-
391.
Harlow, H.F. (1974). Syndromes resulting from maternal deprivation: Maternal
and peer affectional deprivation in primates. In J.H. Cullen (Ed.), Experimental
behavior: A basis for the study of mental disorders. New York: Wiley.
Josselson, R. (1973). Psychodynamic aspects of identity formation in college
women. Journal of Youth and Adolescence, 2(1),3-52.
Josselson, R. (1980). Ego development in adolescence. In J. Adelson (Ed.), Hand-
book of adolescent psychology. New York: Wiley.
Josselson, R. (1987). Finding herself: Pathways to identity development in women.
San Francisco: Jossey-Bass.
Josselson, R. Greenberger, E., & McConochie, D. (1977a). Phenomenological
aspects of psychosocial maturity in adolescence: Part I. Boys. Journal of Youth
and Adolescence, 6, 25-56.
Josselson, R., Greenberger, E., & McConochie, D. (1977b). Phenomenological
aspects of psychosocial maturity in adolescence: Part II. Girls. Journal of Youth
Kegan, R. (1982). The evolving self. Cambridge: Harvard University Press.
Kohut, H. (1977). The restoration of the self. New York: International Universities
Press.
Levinson, D., et al (1978). The seasons of a man's life. New York: Knopf.
Lewittes, H.S. (1982). Women's development in adulthood and old age. Interna-
tional Journal of Mental Health, 11 (1-2), 115-134.
Mahler, M., Pine, F., & Bergman, A. (1975). The psychological birth of the human
infant. New York: Basic Books.
Maslow, A. (1968). Toward a psychology of being. New York: Bantam Books.
Masterson, J. (1985). The real self. New York: Brunner/Maze!.
Miller, J. B. (1976). Toward a new psychology of women. Boston: Beacon Press.
Miller, J.B. (1984). The development of women's sense of self. Work in Progress.
Wellesley: Stone Center. Working Paper Series.
Morgan, E. (1982). Toward a reformulation of the Eriksonian model of female
identity development. Adolescence, 17, 5, 199-21l.
Offer, D. (1969). The psychological world of the teenager. New York: Basic Books.
Offer, D., & Offer, J. (1975). From teenage to young manhood. New York: Basic
Books.
Offer, D., Ostrov, E., & Howard, K.I. (1981). The adolescent: A psychological
self-portrait. New York: Basic Books.
Stem, D. (1985). The interpersonal world of the infant. New York: Basic Books.
Sullivan, H.S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Surrey, J. (1984). The self-in-relation. Work in Progress. Wellesley: Stone Center
Working Papers Series.
Swidler, A. (1980). Love and adulthood in American culture. In N. Smelser & E.
Erikson (Eds.), Themes of work and love in adulthood. Cambridge: Harvard
University Press.
Symonds, P.M., & Jensen, A.R. (1961). From adolescent to adult. New York:
Columbia University Press.
Tolstoy, L. (1961). Anna Karenina. New York: New American Library.
Vaillant, G. (1977). Adaptation to life. Boston: Little, Brown.
Winilicott, D. (1965). The maturational processes and the facilitating environment.
New York: International Universities Press.
Wolf, E.S. (1980). On the developmental line of selfobject relations. In A. Gold-
berg (Ed.), Advances in self psychology. New York: International Universities
Press.
Yalom, I.D. (1980). Existential psychotherapy. New York: Basic Books.
Part II Ego
6
The "New Look" at the Imaginary
Audience and Personal Fable:
Toward a General Model of
Adolescent Ego Development
DANIEL K. LAPSLEY AND KENNETH RICE
In an often-cited passage, Piaget has remarked, "A day will come when the
psychology of cognitive functions and psychoanalysis will have to fuse in a
general theory which will improve both through mutual correction, and
starting right now we should be preparing for that prospect by showing
the relations which could exist between them" (cited in Noam, Kohlberg,
& Snarey, 1983, p. 59). As if to redeem Piaget's prophecy, a number of
papers have appeared over the years that have attempted to explore the
common ground between (orthodox and revisionist) psychoanalytic and
structural developmental theory (e.g., Greenspan, 1979; Lester, 1983;
Shapiro, 1963; Wolff, 1960). As Noam et al. (1983) point out, this compar-
ative work has not led to any systematic integration of the two theories.
What precludes a full integration is the fact that the two theories are beset
by important paradigmatic differences. There is an emerging consensus,
however, that some rapprochement is necessary in order to further our
understanding of "developmental psychopathology" (Noam, this volume;
Noam, 1986; Noam et aI., 1983; Selman, 1980). Hence, the search for
meaningful integrative linkages has entered a new phase. According to
Noam et al. (1983), theoretical rapprochement will be better served not by
focusing on Piaget's theory, but rather on neo-Piagetian social-cognitive
theories, which address more forthrightly psychological concerns (e.g., ego
development, object relations) of interest to the psychoanalytical theorist.
In their view social cognition is the "missing link" that can better bridge
these two dominant developmental paradigms. They write:
Integrations and theoretical synthesis between psychoanalytic ego psychology, self
and object relations theory, and Piagetian ego models have come into closer reach.
Piaget's vision of a general psychology that would one day integrate psychoanalysis
and genetic epistemology is still far from realization, though by [focusing] on the
self and the ego, in combination with our cumulative knowledge in social cognition,
a greater synthesis with psychoanalytic thought will be attained (Noam et al., 1983,
p.127).
In this chapter we explore the possibilities of a "greater synthesis" be-

tween social-cognitive and neopsychoanalytic theories of adolescent ego
110 Daniel K. Lapsley and Kenneth Rice
development. This synthesis should be capable of providing (a) a common

theoretical language for describing both the development of self-under-
standing in adolescence and the ego process of separation-individ-
uation, (b) a developmental grounding for certain individual differences
constructs (e.g., self-monitoring) that have not before been implicated
in adolescent ego development, and perhaps more importantly, (c) a unified
account not only of the common affective reactions that accompany normal
adolescent ego development, but also the psychopathological reactions
of impaired development as well.
Our point of departure is the "new look" at the imaginary audience and
personal fable constructs developed by Lapsley and Murphy (Lapsley,
1985; Lapsley & Murphy, 1985). The imaginary audience is the adoles-
cent's belief that he or she is the object of others' attention and evaluation.
The personal fable is the belief in one's uniqueness, invulnerability, and
omnipotence. These twin constructs have considerable heuristic power for
explaining typically observed adolescent behaviors. Self-consciousness,
show-off behavior, and the preoccupation with shame, shyness, and
embarrassment all seem to reflect the tendency to construct imaginary au-
diences. The adolescent's heightened sense of personal agency, idealism,
and indestructability (as evidenced, for example, in risk-taking behavior)
implicate the personal fable. Although these constructs are useful for
understanding the normal adolescent experience, Elkind (1985) has argued
that they can also be applied with profit to conceptualize certain clinical
phenomenon as well. Indeed, Elkind (1985) suggests that a suitable theory
of adolescent social-cognitive development must have sufficient generative
power to account for the clinical case.
The chapter proceeds in the following manner. We first argue, following
Noam (1986), that attempts to phrase integrative models of development
require a consideration of additional concepts to describe the rhythm of
development, concepts that have not been exploited by orthodox
structural-developmental theorists. In particular we argue that the study of
recapitulated patterns of development has considerable descriptive power.
To demonstrate the utility of this strategy, we examine three recapitulation
hypotheses. We first briefly review the ego separation-individuation pro-
cess in infancy, and show, following losselson (1980), how it is recapitu-
lated in the second phase of separation-individuation in adolescence (Bios,
1962). Next, we show how the sequences of self-understanding in infancy,
as revealed by the various mirror recognition studies, is recapitulated in the
adolescent levels of self-understanding described by Selman (1980) and
Broughton (1978). Third, we extend the recapitulation heuristic to a con-
sideration of the "grandiose self," as described by object-relations theory
(broadly defined, see Greenberg & Mitchell, 1983). We particularly attend
to the requirement of good-enough parenting to mirror, in phase-
appropriate ways, the infant's normal grandiosity and narcissism, so as to
encourage the nascent infant ego to take over for itself the soothing func-
6. "New Look" at the Imaginary Audience and Personal Fable 111
tions performed by the external object, a process called transmuting inter-

nalization. We argue that adolescent ego development also involves a
transmuting internalization process with respect to parental introjects, and
that the narcissism of the grandiose self of infancy is recaptitulated as im-
aginary audience and personal fable ideations in adolescence. We suggest
that a consideration of the three recapitulation themes involving
separation-individuation, self-understanding, and narcissism may provide
the grounding of a general model of adolescent ego development. We con-
clude with some reflections on good-enough parenting in adolescence, and
on how narcissistic personality and borderline disturbances can be under-
stood in light of the present synthesis.
Recapitulation and Rhythm of Development

Most structural developmental stage theories have at least a tacit commit-
ment to variants of the orthogenetic principle (Werner, 1957) that develop-
ment proceeds from a state of globality and lack of differentiation to a state
of differentiation and hierarchical integration. Progress toward the de-
velopmental apex is conceived to be linear and cumulative, yielding greater
adaptive capabilities with each successive acquisition in the sequence. This
understanding of the rhythm of development, however, is ill-suited to fully
account for clinical phenomena. According to Noam (1986, p. 32), "There
are too many simultaneous 'ego states', repetitions and breakthroughs of
primitive thought to be able to organize the material around the most ma-
ture stage alone." For this reason the orthogenetic understanding of de-
velopment must be supplemented with additional concepts, concepts that
take into account the repetitive nature of psychopathology (Noam, 1986).
The psychoanalytic tradition suggests such concepts as fixation, regression,
transference, encapsulation, and recapitulation. A model of developmen-
tal psychopathology so integrated recognizes that the progressive develop-
ment of the ego can, on the one hand, transform the meaning of early
psychic experiences, "thaw out" the basic faults incurred in early child-
hood, and integrate earlier vulnerabilities with later developmental experi-
ences. Such a model would also recognize, on the other hand, that aspects
of the self can become arrested, encapsulated, even though the overall
structures of the self may continue to develop throughout the life span
(Noam, 1986).
Noam (1986; this volume) has shown how the encapsulation of develop-
mental experiences can aid in our understanding of borderline personality
conditions. In this chapter we are interested in exploring the heuristic pow-
er of the recapitulation concept. Recapitulation describes the continuity,
persistence, or repetition of a structural form or pattern for a given
developmental process. By this definition we do not mean to imply that
developmental issues (e.g., attachment, separation-individuation) that
arise in later parts of the life span are mere recapitulations of early psychic
experiences. We do not believe, for example, that adolescents are merely
reworking fixated patterns of infant separation-individuation. Rather, we
argue that while adolescent separation-individuation is governed by "later
psychological organizations" (Noam, 1986, p. 25), the form of the process
in adolescence is structurally similar to the process in infancy.
RECAPITULATION AND PSYCHOANALYTIC THEORY

Although the concept of recapitulation has not been exploited by structural
developmentalists, it has been given a more prominent role in psycho-
analytic theory (e.g., A Freud, 1946, p. 152). A recapitulated pattern of
development can be seen in the similarity between ego ideal formation
in adolescence, and superego formation in early childhood (BIos, 1962).
A child in the phallic stage partakes of the magnificence of his or her
parents by identifying with them, bringing the superego into play. In con-
trast, in early adolescence, the genital child partakes of perfection by
identifying with an idealized friend, which lays the groundwork for ego
ideal formation. BIos (1962) also speaks of a recapitulated pattern of
defensive amnesia between the two age periods.
Usually, memories of the adolescent period become vague at the close of adoles-
cence, buried under an amnestic veil. Facts are remembered well, but the affective
side of the experience cannot be clearly recalled. Repression takes over at the
decline of the resuscitated oedipal complex as it once did before the close of the
oedipal period (BIos, 1962, p. 116).
More recently, Josselson (1980) has made explicit the claim by BIos
(1979) that adolescence constitutes the second phase of separation-
individuation, by showing how the stages of the first phase in infancy, as
described by Mahler (Mahler, Pine & Bergman 1975), are recapitulated in
the second. She writes:
The dynamics of adolescent individuation are strikingly parallel to the process de-
scribed by Mahler and her associates. Although the outcome of the early individua-
tion phase is structuralization, it is not unlikely that the massive structural modifica-
tion that takes place in adolescence repeats the process of the original structure
formation (Josselson, 1980, p. 193).
Mahler's description of the phases (e.g., symbiosis, differentiation, practic-
ing, rapprochement, consolidation) of "psychological birth" of the infant
are well-known. According to Josselson (1980), themes involving sym-
biosis, differentiation, and practicing are recapitulated during the latency
period. Symbiosis is reflected in the child's emotional dependence on pa-
rents; in the fact that the child's superego is composed of parental intro-
jects, in the derivation of self-esteem through parental approval, and in the
belief in parental omnipotence. Differentiation is seen in the child's under-
standing that he or she is a separate person with unique attributes, though
as Josselson (1980) points out, this is more aptly described as a "symbiotic

differentiation," a sense of "we-ness." Finally, the development of inter-
personal and motor abilities during latency reflects the practicing element
of separation-individuation.
Of particular interest to us is the recapitulation in adolescence of the
rapprochement crisis of the infancy period. With the onset of puberty there
is a resurgence of drive that undermines the harmony of the latency period.
There is once again pronounced ambivalence over autonomy, an oscilla-
tion between separation and autonomy, closeness and flight-a sort of
"splitting" of the object world. As in the rapprochement crisis of early
childhood, "the adolescent may come suddenly to an awareness of separ-
ateness, to a realization of the meaning of psychological detachment and of
its negative aspects" (Josselson, 1980, p. 195). As in the shadowing and
clinging behavior of the toddler, the adolescent desires to restore the sym-
biotic unity, wishes to share ego experiences, and desires approval from
parents. Similar patterns of repudiation are also evident: by saying "No" to
parents, by irritating them, and by acting as if he or she had no parents
(Josselson, 1980). Hence, as in infancy, the rapprochement crisis of the
adolescent involves the simultaneous repudiation and reliance on parents,
the desire to be separate and the desire to share with parents without
engulfment by them. When the adolescent can successfully consolidate the
conflict between autonomy and dependence by developing a sense of indi-
viduality within the context of ongoing relationships (i.e., the "capacity to
be alone"; see Winnicott, 1965), the rapprochement crisis is brought to a
close.
As BIos (1979) points out, the individuation process is an attempt by the
adolescent to transcend infantile object ties, to gain distance from the in-
ternalized object, which is to say that individuation is primarily an intra-
psychic process. What is required is an end to the domination of the repres-
sive superego, an end to dependency on parental introjects for approval,
self-esteem, and standards of behavior. The initial move in the ego's
attempt to master the superego involves the de-idealization of parents,
a move that typically results in "mourning" reactions, and also a press
toward conformity (Steinberg & Silverberg, 1986). Optimally, the reorga-
nization of the ego-superego balance results in an ascendent ego that inter-
nally regulates self-esteem based on realistic content, and not on the ideal-
ized content of the superego. Hence, according to Josselson (1980).
The ego must dissociate self-esteem from both the environmental viscissitudes
personified by the parents and the valent self-representations of the superego.
This involves the construction of stable, reality-tested self-representations that can
withstand both archaic guilt and reality-related disapproval (Josselson, 1980,
p. 198).
What is crucial to adolescent separation-individuation, then, is that the

ego take over for itself the services previously provided by parental in-
trojects. In a progressive fashion, played out over a number of years, the

ego must wrest control from the superego in the matter of self-esteem
regulation and in the regulation of guilt. Object-relations theory describes
a similar process operating in infancy, where the vulnerable infant ego
progressively manages to take over for itself the functions provided by
the good-enough mother (i.e., the functions of soothing, holding, and
quieting). This "transmuting internalization" process, then, is immanent in
both the infancy and adolescence phases of separation-individuation, so
that the neopsychoanalytic account of adolescent ego development in-
volves not one but two recapitulation themes. There is first the recapitula-
tion of "phases" (Josselson, 1980) and also the recapitulation of the trans-
muting internalization process. We develop the latter theme more fully in a
later section.
In sum, neopsychoanalytic theory shows that the structure of the ego
development process is formally similar in infancy and adolescence. Both
phases of separation-individuation involve rapprochement crises, an
ambivalence of autonomy, mourning reaction, similar defensive maneu-
vers, and transmuting internalization. In the next section we continue our
development of the recapitulation heuristic by considering cognitive-
structural theories of self-development.
RECAPITULATION AND COGNITIVE DEVELOPMENTAL THEORY

As we have indicated, the recapitulation concept has been most frequently
used within the psychoanalytic tradition to explicate the developmental
process. The stragegy has not been exploited, however, by the cognitive
developmental approach. In order to demonstrate the unifying power of
this concept, we will need to show that cognitive and structural develop-
mental accounts of the self in both infancy and adolescence can also be
described in terms of a recapitulated pattern of development. We turn first
to the emergence of self-understanding in infancy.
Much contemporary research on the self is organized around the distinc-
tion made by James (1985) between the self as subject and self as object (for
reviews, see Damon & Hart, 1982; Harter, 1983). The self as subject is the
I as knower. It is the existential self, implicating knowledge of independent
existence, agency, and volition. The self as object is the Me as known. It
subsumes the categories and concepts, the special features and characteris-
tics, that are revealed by the knower. It says "Not only do I exist and have
agency, but 'I know' , in addition, the particular concepts that define Me."
According to Lewis and Brooks-Gunn (1979), infants develop expectencies
about their agentic control of the environment as a result of the contingent
feedback that they receive from kinesthetic and external (people and ob-
jects) sources. The existential self is evident in visual recognition (mirror)
studies when the infant engages in contingency play with the mirror image,
imitates the movements of the mirror image, engages in rhythmic (bounc-
ing, clapping) movements, and differentiates between contingent and non-
contingent videotaped representations of the self (Harter, 1983). After 15

months of age, infants show considerable facility at feature recognition.
They make mark-directed behaviors to spots of rouge placed on their nose,
indicating that a figural schema of the self has been violated. They also use
appropriate personal pronouns to their mirror image, can label the self,
and can distinguish self from other in pictorial representations. These abili-
ties are interpreted as evidence for the categorical self (Lewis & Brooks-
Gunn, 1979). The ordered acquisition of existential and categorical self-
understanding has been documented as well by Bertenthal and Fischer
(1978). Hence, the available research on self-understanding in infancy
shows that the existential and categorical aspects of the self constitutes a
developmental progression. The infant first develops an existential under-
standing of the self, "that I am," and then comes to understand the cate-
gorical features of the self, "what or who I am."
The existential to categorical progression of infant self-understanding is
also evident in the phases of self-understanding in adolescence, as de-
scribed by Selman (1980) and Broughton (1978). As Damon and Hart
(1982) point out, there is a noteworthy convergence in the findings pre-
sented by these independently derived sequences.
Virtually all researchers have found that, with development, adolescent self-
understanding shows an increasing use of psychological and social relational con-
cepts for describing the "Me," a more prominent belief in the "I's" agency and
volitional power, and a tendency toward integration of the disparate aspects of the
self into an internally consistent construct system (Damon & Hart, 1982, p. 855).
In Selman's (1980) Level 3, for example, his first stage in adolescence, the
adolescent conceives of the "mind" as an active processor, monitor, and
manipulator of experience. As such, the mind is endowed with volitional
powers and is conceived as a powerful mode of self-control. Broughton's
(1978) first level in adolescence (Level 3) describes the mind similarly as an
entity that is endowed with volitional characteristics, and which is inde-
pendent of the physical activities of the body. The "I" mentally processes
information, monitors and evaluates inner states, and "knows itself" in a
privileged way. The heightened sense of personal agency in young teen-
agers has been attributed to this level of self-understanding (Damon &
Hart, 1982; Lapsley & Murphy, 1985).
The limitations of early adolescent self-understanding is described some-
what differently by Selman (1980) and Broughton (1978). Selman (1980)
sees the limitation of Level 3 self-understanding as residing in the adoles-
cent's tacit understanding that there are some aspects of self-experience
that are beyond the volitional control of his or her "observing ego." This
understanding of limits is uncoordinated with and contradicts the agentic
understanding of the "I" as an active manipulator of experience. Brought-
on (1978) sees the lack of appreciation for the self's unique properties as
being the limitation of self-understanding in adolescence. Both limitations
are resolved by the emergence of the final stages in the Selman (1980) and
Broughton (1978) sequences. In Selman's Level 4, the late adolescent con-

strues the self as a unified system with conscious and unconscious levels
and boundaries. In Broughton's (1978) Broughton (1981) Level 4, the self
is seen as a system of distinct elements (e.g., real self, false self) that can
operate in either a concordant or divided fashion.
Apart from the subtle differences, what both theories describe is a se-
quence of self-understanding in adolescence that proceeds from an existential
understanding of the "I" as involving agency, volition, monitoring, and
evaluation, to a categorical understanding of particular characteristics, of
levels and boundaries (Selman), and of the systemic properties of self-
elements (Broughton). If different terminology is preferred to existential
and categorical to describe the progression of self-understanding in
adolescence, and if this is to highlight the theoretical nature of self-
consciousness in adolescence (Broughton, 1977) , we might suggest that
adolescent thought be described in terms of an ontological and epistemo-
logical self. In early adolescence there is a keen sense of the mental self's
existence, that it has being, and is endowed with volition and agency. This
is the problem of ontology, "that I am," that concerns the young adoles-
cent. The knowledge of self-boundaries, the legitimation of "what or who
I am," is the epistemological problem of late adolescence. Apart from ter-
minological preferences, however, what is clear is that the process of
self-understanding bears remarkable structural similarities in infancy and
adolescence, which we are describing as a recapitulated pattern of develop-
ment.
SUMMARY
We have now examined the recapitulation themes that are immanent to
theoretical accounts of self and ego development in infancy and adoles-
cence. We have shown how the first process of separation-individuation in
infancy is recapitulated in the second phase in adolescence, and how the
emergence of self-understanding proceeds from an existential (ontological)
understanding of the self to a categorical (epistemological) understanding
in both age periods. The recapitulation heuristic, then, seems to possess
general power to conceptualize self and ego process irrespective of particu-
lar theoretical traditions. We are now in a position to extend the recapi-
tulation strategy in order to show how imaginary audience and personal
fable ideations can be considered recapitulations of the narcissism and
grandiosity of infancy.
Narcissism and the Grandiose Self

INFANCY
The dialectical interplay between narcissism and sensitive mothering is a
common theme in object-relations accounts of infant ego development.
The narcissistic infant's self-fascination has to be "mirrored" back to the

infant in the "gleam in the mother's eye" (Kohut, 1971). The mother has to
acknowledge and implement the prideful exhibitionism and sense of
omnipotence of the infant, giving strength to the infant's vulnerable ego
(Winnicott, 1965) and confirming the child's self-esteem (Kohut, 1971) in
the process. But the child cannot rely forever on the ego sustenance pro-
vided by mother. Kohut (1971) suggests the term transmuting internaliza-
tion to describe the process "where the mother's capacity to reduce physi-
cal and psychological tensions in the infant are gradually taken over by the
baby through a manageable, bit-by-bit withdrawal of the mother's minis-
trations" (White, 1986, p. 151). In this way, through phase-appropriate
empathic failure, the baby learns to soothe, hold, nurture, encourage, and
set limits for itself, reflecting the internalization of the "holding" functions
of the good-enough caregiver (White, 1986).
The chronic failure of phase-appropriate, good-enough caregiving
results in grandiose (Kohut, 1971) or false (Winnicott, 1965) self-
organizations. According to Winnicott (1965), for example, the etiology of
the false self resides in the mother who is not good enough, who does not
meet the infant's gesture, who does not implement the infant's omni-
potence, but rather substitutes her own response, resulting in compliance
on the part of the infant. Through the inability of the mother to sense her
infant's needs, the infant gets seduced into compliance and becomes reac-
tive to the environment. The "not good-enough" infant lives falsely.
"Through this false self the infant builds up a false set of relationships, and
by means of introjections even attains a show of being real, so that the
child may grow to be just like. . . whoever at the time dominates the
scene" (Winnicott, 1960, p. 146). The false self is organized to protect and
hide the true self from exploitation, but the defense can be expected to
engender feelings of unreality and futility. For Kohut (1971) the "gran-
diose self' is a developmental arrest and fixation at primitive levels of gran-
diosity. This arrest occurs as the result of the failure of the environment to
provide empathic responses ("mirroring") to the exhibitionistic needs of
the child. For healthy development to occur, the caregiver must be sensi-
tive to the infant's need for admiration (Cooper, 1986). When this does not
occur, when the infant is instead confronted with disapproval or neglect,
the result is a distortion of self-development, leading to "later narcissistic
vulnerability because the grandiose fantasy becomes repressed and inac-
cessible to modifying influences" (Kohut, 1986, p. 67).
Modell (1986) argues that Winnicott's "false self" and Kohut's narcissis-
tic disorder refer to identical clinical types. Both can be traced to insensi-
tive, phase-inappropriate mirroring during the course of the transmuting
internalization process. Both clinical types are said to describe a distur-
bance of affects. For example, the narcissistic character can be considered as
a defense against the intrusion of the not good-enough caregiver. This is
accomplished by creating the illusion of self-sufficiency, which prevents the
vulnerable self from placing demands on the caregiver. Hence, narcissistic
organizations (grandiosity, precocious autonomy) can be seen as "false

self" or "as if" presentations that defend the integrity of the true self
(Modell, 1986; also, Miller, 1986). Narcissism can also be a defensive reac-
tion against the feelings of separateness and object loss that are experi-
enced during the course of separation-individuation. To counter separation
anxiety and the mourning reactions associated with object loss., the infant
ego recalls activities that yielded a maternal smile and feelings of symbiotic
fusion. "If the 'self-as-agent' can perform in a manner that elicits the inter-
nalized representation of a maternal smile, there is the illusion that the
mother is present and part of the self, leaving the toddler with a feeling of
safety" (Rothstein, 1986, p. 310). The experience of the "self smiling at
itself" reduces the anxiety of object loss, thereby preserving the illusion
that the ego can control the object. By having recourse to the internalized
representation of the maternal smile, the ego preserves the elation of the
omnipotent mother smiling at, and in symbiotic unity with, the self (Roth-
stein, 1986). However, if the object cannot be so controlled, a variety of
affective disturbances may result, including "narcissistic rage" (Kohut,
1971; White, 1986), a fixation on the false self (Miller, 1986), mourning
(White, 1986) and shame reactions (Morrison, 1986), and feelings of deple-
tion, despair, and emptiness (Kohut, 1972; Morrison, 1986).
This review of theories of infant narcissism reveals a number of impor-
tant themes. One theme involves the fact that the course of ego develop-
ment in infancy and early childhood involves normal phases of narcissism
and grandiosity. Another theme concerns the ability of good-enough care-
givers to effect the transmuting internalization process, and hence to re-
spond to the infant ego in phase-appropriate ways. A third theme concerns
the continuity between normal childhood narcissism and narcissistic distur-
bances. Narcissistic disturbances are the result of defective object relations
and a subverted transmuting internalization process. "The development
of pathological forms of narcissism is largely dependent upon the actual
failure of the environment to provide appropriate empathic responses to
the infant's needs" (Cooper, 1986, p. 135). Such forms may be described
in terms of self-inflation, grandiosity, shame reactions, self-depletion, de-
pression, and "as if" -false-self organizations. Secondary narcissistic in-
vestment can also be seen as a defensive reaction to separation anxiety and
to the mourning reactions of object loss during separation-individuation.
These themes, normal narcissism, transmuting internalizations, and the
various defensive strategies and affective reactions, will be ascendent
again during the second phase of separation-individuation. And the struc-
ture of this process will take familiar forms. Normal narcissism and
grandiosity will find expression as personal fable and imaginary audience
ideations (e.g., Lapsley, 1985; Lapsley & Murphy, 1985); transmuting inter-
nalization will involve the gradual relinquishment of parental introjections
as the basis of self-esteem regulation (BIos, 1962; Josselson, 1980), and this
will be accompanied by familiar affective reactions; and finally, patho-
logical narcissism, expressed as narcissistic personality and borderline
disorders, will be seen to result from the failure of the environment to pro-
vide empathic responsiveness to the adolescent's narcissistic strivings and
pursuit of ideals.
ADOLESCENCE
Adolescent narcissism is an explicit feature of promient neopsychoanalytic
theories of adolescent ego development. Bios (1962), for example, sees a
marked increase in narcissism in the adolescence proper phase, and he
distinguishes between narcissistic object choice, narcissistic defense, and
the transitory narcissistic stage. Narcissistic object choice describes a type
of friendship selection that is motivated by the desire of the adolescent to
possess an admired quality, by proxy, through the friendship. Narcissistic
defense describes the prolongation of the grandiose self as a narcissistic
disturbance in response to the inability of the adolescent ego to relinquish
the omnipotence of parental introjects. The phase of transitory narcissism
precedes heterosexual object finding, and is occasioned by the decathexis
of the internalized parent. The latter two constructs describe the continuity
between normal and pathological ego development, and will hence play an
important role in the present model.
The shift of cathexis in adolescence is key to understanding normal nar-
cissism during this period (BIos, 1962; Josselson, 1980). The decathexis of
parental introjects removes object representations as a source of narcissis-
tic gratification. This is initially experienced as impoverishment, isolation,
ambivalence, a feeling of void, or, more generally, as mourning reactions.
In a manner similar to the infancy period (Rothstein, 1986), separation
anxiety is compensated for by narcissistic self-inflation. In a sense, the
grandiose self of infancy is reactivated (Josselson, 1980) to resupply the
adolescent ego with the aliments necessary to maintain self-esteem, just as
the infant ego was nurtured by the mirroring of the good-enough caregiver.
The principal task is to maintain a hold on object relations in the face of the
decathexis of the object world, and to reestablish firm ego boundaries. The
narcissistic adolescent accomplishes the former task by means of object
relational ideation, and the latter by the "willful creation of ego states of a
poignant internal perception of the self" (BIos, 1962, p. 98), that is, by a
"self-observing ego" (BIos, 1962).
Object relational ideation describes the private fantasies that prepare
the adolescent for interpersonal transactions. They are "trial actions" that
allow the adolescent to assimilate, in manageable doses, the affective ex-
periences of social participation. The rich fantasy life of the adolescent is
crucial to the transmuting internalization process of this period. According
to Josselson (1980, p. 199),
The love of the omnipotent parent, previously structured in the superego, must be
replaced by love of the self or the possible self. Much of the adolescent fantasy
... is the enactment of the transmuted narcissism, visions of the self [italics added]
to replace the lost superego love."
Related to the use of fantasy is the propensity of adolescents to use people

in make-believe relationships. "These relationships lack a genuine quality;
they constitute experiences which are created for the purpose of disengage-
ment from early love objects" (BIos, 1962, p. 97). Individuals in these
relationships are perceived more as representations than as persons, and
they serve mostly the need for reassurance in the face of separation anxi-
ety. The "as if' quality of the make-believe relationship can be understood
as restitution for object loss and as an attempt to enrich an improverish ego
(BIos, 1962).
The keen perception of inner life by a "self-observing ego" is motivated
by the desire to reestablish firm ego boundaries. "Self-induced ego states
of affective and sensory intensity allow the ego to experience a feeling of
self and thus protect the integrity of its boundaries and its cohesion"
(BIos, 1962, p. 92). Two important consequences result from the action of
the self-observing ego. First, the adolescent is said to experience a height-
ened sense of personal uniqueness: "Nobody ever felt the way I do";
"Nobody has ever seen the world the way I do." Second, the adolescent
experiences a surge of personal agency, and a sense of invulnerability.
"This circumstance gives the individual a false sense of power which in
turn impairs his judgment in critical situations, often with catastrophic
consequences" (BIos, 1962, p. 1(0). The grandiosity that results from the
action of the self-observing ego should also be seen, like object relational
ideation, as a restitution for the object loss that occurs with the decathexis
of parental introjects. In a manner reminiscent of infant narcissism, the
experience of the "self smiling at itself [i.e., self-observing ego] undoes the
feeling of object loss implicit in the separation-individuation experience"
(Rothstein, 1986, p. 310).
A Synthesis
The crux of our synthesis can be stated as follows: (a) Imaginary audience
and personal fable constructions can be seen as normative features of the
transitory narcissistic phase of adolescent ego development, features that
are described by BIos (1962) as object relational fantasy (e.g., imaginary
audience) and the "self-observing ego" (e.g., personal fable). (b) The
grandiosity and narcissism that accompanied the first phase of separation-
individuation (in infancy) is recapitulated in the second phase (in adoles-
cence) as personal fable and imaginary audience ideations. With imaginary
audience ideation, the adolescent anticipates the reactions of others to the
self in real or imagined situations (Elkind, 1967, 1985; Lapsley & Murphy,
1985). This constitutes "object relational" ideation, the function of which
is to preserve a hold on object ties during the course of psychological separa-
tion from parents. The construction of "imaginary audiences," and "vi-
sions of the self," serves defense and restitutive functions. They constitute
"trial actions," "make-believe" relationships, that prepare the adolescent

for heterosexual intimacy. Hence, these private fantasies of the reactions
of others to the self allows the adolescent to prepare for and manage the
affective experiences of social interaction.
The personal fable describes the adolescent conviction that he or she is
unique, that one's thoughts and feelings are not shared by others, and that
the omnipotence of one's reflections should be sufficient to compel others
to submit to the adolescent's idealistic schemes. Elkind (1967, p. 1031)
describes it this way:
Perhaps because he believes he is of importance to so many people, the imaginary
audience, he comes to regard himself, and particularly his feelings, as something
special and unique. Only he can suffer with such agonized intensity, or experience
such exquisite rapture. The emotional torments ... exemplify the adolescent's
belief in the uniqueness of his own emotional experience.
Bios (1962, p. 93) describes this "pesonal fable" using similar language: "It
as if the adolescent experiences the world with a unique sensory quality
that is not shared by others: 'Nobody ever felt the way I do'; 'Nobody sees
the world the way I do'." The adolescent sense of indestructability is
another feature of the personal fable. "This impairs the adolescent's judg-
ment in critical situations since it provides a false sense of power" (Bios,
1962, p. 100). Personal fable ideation is identical to the narcissistic restitu-
tion strategy described by Bios (1962), where the keen perception of inner
life, "the willful creation of ego states of a poignant internal perception of
the self" (Bios, 1962, p. 98), leads to a heightened sense of uniqueness,
indestructability, and personal agency-which impairs adolescent judg-
ment in critical situations.
Although Elkind's (1967) account of the imaginary audience and person-
al fable and Bios's (1962) description of transitory narcissism seen to refer
to identical phenomena, a more complete synthesis is possible when the
imaginary audience and personal fable are interpreted in terms of the "new
look" proposed by Lapsley and Murphy (1985). The new look suggests that
the onset and decline of these twin constructs from early to late adoles-
cence can be accounted for by reference to Selman's (1980) levels of self-
understanding (Levels 3 and 4) in adolescence, as described earlier. For
example, at Level 3, the mind is conceived to be a monitor, processor, and
manipulator of experience. According to Selman (1980, pp. 104-105),
What appears new and striking at Level 3 is a belief in the observing ego-that is,
the self-aware self as an active agent. This concept of active agency strikes us
as ... critical for a child's feeling of having some control over his own thoughts and
feelings. For the Level 3 child, the mind (or ego) is now seen as playing an active
moderating role between inner feelings and outer actions.
The self-reflective monitoring properties of the observing ego, the belief in

the power of the mind to manipulate experience, accounts for the adoles-
cent tendency to anticipate the reactions to the self in real or imagined

situations (Lapsley & Murphy, 1985). The self-reflective properties of the
mind permit the construction of object relational fantasies, visions of the
self, trial actions, make-believe relationships-imaginary audiences. This
stage also provides more suitable language for an integrative understand-
ing of the personal fable. Self-focused attention sets off the self as figure
from the background of other perspectives, highlighting elements of indi-
viduality in the process. This is comparable to the function of the "self-
observing ego" in BIos's (1962) theory, that is, to establish a firm sense of
ego boundaries "by willful creation of ego states of a poignant perception
of the self" (p. 98), which also results in a heightened sense of individual-
ity. The second element of the personal fable, the sense of personal agency
and indestructability, is also explained by Selman's (1980) Level 3. "The
increased self-consciousness of adolescents, who now conceive of the mind
as an active processor and manipulator of experience, may now see the
power of self-reflection as the capacity or condition for exerting power in
other circumstances as well" (Lapsley & Murphy, 1985, p. 213). The
heightened early adolescent understanding of the agentic and volitional
aspects of the "I," which are described so well in the sequences of self-
understanding (Selman, 1980; Broughton, 1978), readily accounts for
personal fable ideation, and is congruent with the "self-observing ego"
explanation offered by BIos (1962). Hence, the observing ego features
of Level 3 self-understanding provides an explanation of the imaginary
audience and personal fable constructs, and is compatible with the self-
observing ego explanation offered by BIos (1962).
Lapsley and Murphy (1985) account for the diminution of these idea-
tions in late adolescence by the appearance of the next stage (Level 4) in
Selman's (1980) sequence. At this level the young adult can coordinate all
possible third-party perspectives, an ability that obviates the need to im-
agine the reaction of others to the self. Because the adolescent can now
locate the self in the larger matrix of social relations, and because he or she
can coordinate the observing egos of others, social cognitions can be more
faithful attuned with reality (Lapsley & Murphy, 1985). The notion of per-
sonal agency, individuality, and indestructability also diminishes at this
stage. At Level 4 the late adolescent is aware that the observing ego is
powerless to control unconscious motivations, which highlights the limits
of self-reflective self-control. This understanding diminishes the sense of
personal agency. And because the adolescent can coordinate third-party
perspectives, he or she can realize the power and limitations of the observ-
ing egos of others, thereby decreasing the sense of personal uniqueness.
"The self is no longer isolated as figure from the ground of other perspec-
tives. Rather ... there is not only an awareness of multiple figures, but
also an awareness that the self can be located within the ground of other
perspectives (Lapsley & Murphy, 1985, p. 214)
SUMMARY
Thus far we have shown that the second phase of separation-individuation
also involves a transmuting internalization process. Crucial to this process
is the decathexis of object representations, which is experienced as am-
bivalence over autonomy and as object loss. As restitution, the adolescent
passes through a transitory phase of narcissism, which involves object re-
lational ideation and the self-reflective monitoring of inner states (the self-
observing ego). We argue that these twin features of adolescent narcissism
are identical in form and function to the imaginary audience and personal
fable constructs. These constructs have been interpreted (Lapsley & Mur-
phy, 1985) as being outcomes of the development of self-understanding
(e.g., Selman, 1980). Within the present synthesis we can argue that object
relational (the imaginary audience) and personal fable ideation, as prob-
lems of self-understanding (social cognition), are outcomes of the trans-
mutation of narcissism process of adolescent ego development. Implicit in
this summary are the three recapitulation themes. Adolescent ego develop-
ment, like development during early childhood, can be described with
reference to (a) the transmutation process involved in separation-
individuation, (b) the agentic (ontological) to categorical (epistemological)
shift in self-understanding, and (c) the normal manifestations of transitory
narcissism and grandiosity.
NARCISSISTIC DISTURBANCES
The recapitulation heuristic also sheds light on the continuity between
transitory narcissism and the narcissistic defense (BIos, 1962). The nature
of narcissistic disturbances during the infancy period is described in terms
of a fixation of the grandiose (Kohut, 1971) and false self (Winnicott,
1965). This fixation occurs in response to insensitive, not good-enough
caregiving that fails to mirror the child's need for admiration. Adolescence
provides a second opportunity to resolve the basic fault incurred in early
childhood (Balint, 1957; Giovacchini, 1979). The recapitulation heuristic
can be used to generate hypotheses on what to look for during this period.
It would seem clear, for example, given the heuristic, that the rhythm of
ego development in adolescence would require a formally similar pattern
of response from the environment as was required in the first phase of
psychological individuation. It would require that parents, siblings, and,
perhaps more importantly, peers be empathically responsive to the adoles-
cent's need for admiration. The requirements of good-enough care giving
remain the same as in early childhood: to be emotionally available during
the course of the adolescent's ambivalence over autonomy; to "mirror"
narcissistic and exhibitionistic strivings; to effect the transmutation of
narcissism by withdrawing, in phase-appropriate ways, through a gradual
selectivity of response, the mirroring support, thereby channeling the

adolescent's narcissistic needs in realistic directions (White, 1986). Hence,
on the one hand, parents must be empathic, sensitive, and emotionally
available; on the other, they must allow the adolescent to individuate, to
develop inner resources, and to regulate self-esteem in light of optimal
frustration. As Mahler (Mahler et aI., 1975, p. 79) points out in the case of
early childhood, the parents' emotional willingness "to let go of the
toddler-to give him, as the mother bird does, a gentle push, an en-
couragement toward independence-is enormously helpful. It may even
be the sine qua non of normal (healthy) individuation." There is no reason
to doubt that a similar "gentle push" is required in adolescence in order to
advance the individuation process.
Impaired ego development in adolescence can also be expected to result
in a consolidation of the grandiose or false self. Indeed, we argue that the
failure to transmute narcissism builds the clinical foundation for narcissistic
personality (Bleiberg, 1984; Kohut & Wolf, 1986) and borderline condi-
tions (Kernberg, 1963; Simon, 1984) (see Cooper, 1986, and Meissner,
1986, on differential diagnosis). The diagnostic criteria for narcissistic per-
sonality disorder, from the DSM-III (American Psychiatric Association,
1980), are reported in Table 6.1. What is plainly evident is that the di-
agnostic features of narcissistic personality disorder describe exaggerated
imaginary audience and personal fable ideation. Pronounced object re-
lational ideation is seen in Criterion B (preoccupation with success and
power fantasies) and C (exhibitionism, requiring constant attention and
admiration). It is also seen in Criterion D, particularly when the feelings of
shame and humiliation are understood as reactions to an "audience"
(Elkind, 1967). An exaggerated personal fable is seen in Criterion A
(grandiose esense of self-importance or uniqueness), and in at least some
of the disturbances in interpersonal relationships (e.g., entitlement, ex-
ploitativeness). Bleiberg (1984) argues that narcissistic children define
themselves with reference to the false self, and assume "as if" identities.
She writes:
Feelings of unreality pervade their self-experiences, and often they spend their lives
acting out a role. It matters little whether the role is ... a chameleonic perfor-
mance in which these children carefully monitor the environment and adapt the
most convenient identity. The identity of these children ... are not based on an
internal or core sense of identity, but on their perception of what others expect
from them or what they believe will gain them admiration or advantage. (Bleiberg,
1984, p. 505)
The pronounced chameleonic adaptation to the environment is also a

feature of the Group III borderline category described by Grinker, Wer-
ble, and Dryer (1968). These patients await cues from others, assuming "as
if" roles in response to the expectations of those with whom they are with.
As Simon (1984) points out, adolescents with a Group III disturbance may
TABLE 6.1. Diagnostic criteria for narcissistic personality disorder.

A. Grandiose sense of self-importance or uniqueness, (e.g., exaggeration of achievements
and talents, focus on the special nature of one's problems)
B. Preoccupation with fantasies of unlimited success, power, brilliance, beauty, or ideal love
C. Exhibitionism: the person requires constant attention and admiration
D. Cool indifference or marked feelings of rage, inferiority, shame, humiliation, or
emptiness in response to criticism, indifference of others, or defeat
E. At least two of the following characteristics of disturbances in interpersonal relationships:
1. Entitlement: expectation of special favors without assuming reciprocal responsibilities
(e.g., surprise and anger that people will not do what is wanted)
2. Interpersonal exploitiveness: taking advantage of others to indulge own desires or
for self-aggrandizement; disregard for the personal integrity and rights of others
3. Relationships that characteristically alternate between the extremes of over-
idealization and devaluation
4. Lack of empathy: inability to recognize how others feel (e.g., unable to appreciate the
distress of someone who is seriously ill)
Note. From Diagnostic and Statistical Manual of Mental Disorders (3rd ed., p. 317), American
Psychiatric Association, 1980, Washington, DC: American Psychiatric Press. Reprinted by
permission.
behave in ways that are similar to normal adolescents, thus escaping

psychiatric attention.
We wish to make two general points. First, we have argued that transi-
tory adolescent narcissism, as a normal phase of adolescent individuation,
is expressed as imaginary audience and personal fable ideation. However,
narcissistic disturbance can also be subsumed under the present general
framework. Narcissistic personality disorder, as one example, is expressed
largely in terms of an exaggerated, pronounced, and persistent tendency to
construct personal fables and object relational ideation. The narcissistic
disturbance, as it does in early childhood, reflects an alienation of the core
self, and the ascendence of the false self.
Our second point regards the chameleonic existence of false-self pre-
sentations. We are struck by the conceptual similarity between this feature
of the narcissistic disturbance, and the self-monitoring construct developed
by Snyder (Snyder, 1979; Snyder & Cantor, 1980; Snyder & Gangestad,
1982). Self-monitoring is an individual differences orientation to how one
presents and guides one's behavior in social contexts. High self-monitors
carefully manage the images of the self that they present in social situa-
tions. They are like chameleons, striving to be the type of person called for
by a given situation. High self-monitors are hypersensitive to interpersonal
cues so as to always be ready to respond in situation ally appropriate ways.
Indeed, their behavior shows considerable situation-to-situation specific-
ity. Low self-monitors, on the other hand, are (a) less responsive to the
contingencies of the situation; (b) more likely to project their own core
identities and personal dispositions in social situations, and (c) more likely
to guide their behavior on the basis of "inner" sources of information. We

would argue that self-monitoring reflects not only a dimension of individual
differences, but that it also has a developmental history that is grounded in
the general process of adolescent ego development. The high self-monitor
lives a false-self existence. It is an existence predicated on "not good-
enough" responses to transitory adolescent narcissism, and it reflects a
deficit in the transmuting internalization functions of ego individuation.
Conclusion
In this chapter we extend Lapsley and Murphy's (1985) "new look" at the
imaginary audience and personal fable constructs. This view was criticized
by Elkind (1985), who doubted whether the "new look" would have suf-
ficient generative power to account for clinical phenomena. In response,
Lapsley (1985, p. 235) argued that "interpersonal understanding, and the
emergence of Level 3 perspective-taking abilities, may be at the heart of
ego development in early adolescence, so that the imaginary audience and
personal fable ideations can be seen as an integral part of the ego develop-
mental process of separation-individuation." This chapter is an attempt to
"make good" on this expectation. Lapsley and Murphy (1985) showed how
the personal fable and imaginary audience can be understood as outcomes
of social-cognitive development. In this chapter these constructs are seen
to take on new meaning for adolescent development. They are the most
visible expressions of adolescent narcissism. They are an outcome of social-
cognitive development, and a marker of the transmuting internalization
process. Indeed, the three recapitulation themes (e.g., transmuting inter-
nalization, self-understanding, narcissism) are anchored by these twin
constructs.
It should be clear that what we describe here is not yet a general theory
of adolescent ego development. Our aim was more modest. We explore
the relations that might exist between neo-Piagetian and neo-Freudian
accounts of object relations and self-development in adolescence, using
the "new look" as a point of departure, and recapitulation as a unifying
heuristic. Future research will need to demonstrate the unifying power of
the imaginary audience and personal fable, that is, their relation to levels
of self-understanding and to adolescent narcissism. The developmental
trajectory of these constructs, particularly as a function of caregiving
styles, will also need to be addressed. Additional research questions
include how the transmutation of narcissism is expressed in terms of in-
dividual differences and in patterns of psychopathology. Fortunately, the
availability of models of transmuting internalization (Benjamin, 1979;
Goldberg, 1986), and of recently designed assessments of ego individua-
tion (Levine, Green, & Millon, 1986) and adolescent narcissism (Raskin &
Hall, 1981) should facilitate the type of integrative research suggested by
the present model. Such research should take us some steps further in the
development of a general theory of adolescent ego development.
REFERENCES
American Psychiatric Association. (1980). Diagnostic and statistical manual of men-
tal disorder (3rd ed.). Washington, DC: American Psychiatric Press.
Balint, M. (1957). Problems of human pleasure and behavior. London: Hogarth.
Benjamin, L.S. (1979). Structural analysis of differentiation failure. Psychiatry, 42,
1-23.
Bertenthal, B., & Fischer, K. (1978). Development of self-recognition in the in-
fant. Developmental Psychology, 14,44-50.
Bleiberg, E. (1984). Narcissistic disorders in children. Bulletin of the Menninger
Clinic, 48,501-517.
Bios, P. (1962). On adolescence. New York: Free Press.
Bios, P. (1979). The adolescent passage. New York: International Universities
Press.
Broughton, J. (1977). 'Beyond formal operations": Theoretical thought in adoles-
cence. Teachers College Record, 79,87-96.
Broughton, J. (1978). The development of the concepts of self, mind, reality, and
knowledge. In W. Damon (Ed.), Social cognition (pp. 75-100). San Francisco:
Jossey-Bass.
Broughton, J. (1981). The divided self in adolescence. Human Development, 24,
13-32.
Cooper, A. (1986). Narcissism. In A. Morrison (Ed.), Essential papers on narciss-
ism (pp. 112-143). New York: New York University Press.
Damon, W., & Hart, D. (1982). The development of self-understanding from in-
fancy through adolescence. Child Development, 53, 841-864.
Elkind, D. (1967). Egocentrism in adolescence. Child Development, 38, 1025-
1034.
Elkind, D. (1985). Egocentrism redux. Developmental Review, 5, 218-226.
Freud, A. (1946). The ego and the mechanisms of defense. New York: International
Universities Press.
Giovacchini, P. (1979). The borderline aspects of adolescence and the borderline
state. Adolescent Psychiatry, 6,320-338.
Goldberg, A. (Ed.) (1986). Progress in self psychology (Vol. 2). New York: Guil-
ford.
Greenberg, J., & Mitchell, S. (1983). Object relations in psychoanalytic theory.
Cambridge, MA: Harvard University Press.
Greenspan, S. (1979). Intelligence and adaptation: An integration of psychoanalytic
and Piagetian developmental psychology. New York: International Universities
Press.
Grinker, R.R., Werble, B., & Drye, R. (1968). The borderline syndrome: A be-
havioral study of ego functions. New York, Basic Books.
Harter, S. (1983). Developmental perspectives on the self-system. In P. Mussen
(Ed.), Handbook of child psychology: Vol. IV. Socialization, personality, and
social development (E.M. Hetherington, Vol. Ed.), (pp. 275-385). New York:
Wiley.
James, W. (1985). Psychology: The briefer course. Notre Dame, IN: University of
Notre Dame Press.
Josselson, R. (1980). Ego development in adolescence. In J. Adelson (Ed.), Hand-
book of adolescent psychology (pp. 188-210). New York: Wiley.
Josselyn, I. (1954). The ego in adolescence. American Journal of Orthopsychiatry,
24,223-237.
Kernberg, O. (1963). The diagnosis of borderline conditions in adolescence.
Adolescent Psychiatry, 6,289-319.
Kohut, H. (1971). The analysis of the self. New York: International Universities
Press.
Kohut, H. (1972). Thoughts on narcissism and narcissistic rage. Psychoanalytic
Study of the Child, 27, 360-399.
Kohut, H. (1986). Pathological forms of self-esteem regulation. In A. Morrison
(Ed.), Essential papers on narcissism. New York: New York University Press.
Kohut, H., & Wolf, E. (1986). The disorders of the self and their treatment: An
outline. In A. Morrison (Ed.), Essential papers on narcissism (pp. 175-196).
New York: New York University Press.
Langer, J. (1982). Dialectics of development. In T. Bever (Ed.), Regressions in
mental development. Hillsdale, NJ: Erlbaum.
Lapsley, D. (1985). Elkind on egocentrism. Developmental Review, 5, 227-236.
Lapsley, D., & Murphy, M. (1985). Another look at the theoretical assumptions of
adolescent egocentrism. Developmental Review, 5, 201-217.
Lester, E. (1983). Separation-individuation and cognition. Journal of the American
Psychoanalytic Association, 31, 127-155.
Levine, J., Green, c., & Millon, T. (1986). The separation-individuation test of
adolescence. Journal of Personality Assessment, 50,123-137.
Lewis, M., & Brooks-Gunn, J. (1979). Social cognition and the acquisition of self.
New York: Plenum.
Mahler, M., Pine, F., & Bergman, A. (1975). The psychological birth of the human
infant. New York: Basic Books.
Meissner, W. (1986). Narcissistic personalities and borderline conditions. In A.
Morrison (Ed.), Essential papers on narcissism (pp. 403-437). New York: New
York University Press.
Miller, A. (1986). Depression and grandiosity as related forms of narcissistic dis-
turbances. In A. Morrison (Ed.), Essential papers on narcissism (pp. 323-347).
Modell, A. (1986). A narcissistic defense against affects and the illusion of self-
sufficiency. In A. Morrison (Ed.), Essential papers on narcissism (pp. 293-307).
Morrison, A. (1986). Shame, ideal self, and narcissism. In A. Morrison (Ed.),
Essential papers on narcissism (pp. 348-372). New York: New York University
Press.
Noam, G., Kohlberg, L., & Snarey, J. (1983). Steps toward a model of the self. In
B. Lee and G. Noam (Eds.), Developmental approaches to the self (pp. 59-141).
New York: Plenum.
Noam, G. (1986). Borderline personality disorders and the theory of biography and
transformation (Part 1). McLean Hospital Journal, 11, 19-43.
Pulver, S. (1986). Narcissism: The term and the concept. In A. Morrison (Ed.),
Essential papers on narcissism (pp. 91-111). New York: New York University
Press.
Raskin, R., & Hall, C. (1981). The narcissistic personality inventory: Alternative
form reliability and further evidence of construct validity. Journal of Personality
Assessment, 45, 159-162.
Rothstein, A. (1986). The theory of narcissism: An object-relations perspective. In
A. Morrison (Ed.), Essential papers on narcissism (pp. 308-320). New York:
New York University Press.
Selman, R (1980). The growth of interpersonal understanding: Developmental and
clinical analysis. New York: academic Press.
Selman, R, & Yando, R (Eds.) (1980). Clinical developmental psychology. New
directions for child development. Vol. 7. San Francisco: Jossey Bass.
Shapiro, R (1963). Adolescence and the psychology of the ego. Psychiatry, 26,
77-87.
Snyder, M. (1979). Self-monitoring processes. In L. Berkowitz (Ed.), Advances in
experimental social psychology (Vol. 12, pp. 85-128). New York: Academic
Press.
Snyder, M., & Cantor, N. (1980). Thinking about ourselves and others: Self-
monitoring and social knowledge. Journal of Personality and Social Psychology,
39,222-234.
Snyder, M., & Gangestad, S. (1982). Choosing social situations: Two investigations
of self-monitoring processes. Journal of Personality and Social Psychology, 43,
123-135.
Steinberg, L., & Silverberg, S. (1986). The vicissitudes of autonomy in early
adolescence. Child Development, 57, 841-851.
Werner, H. (1957). The concept of development from a comparative and organis-
mic point of view. In D. Harris (Ed.), The concept of development (pp. 125-148).
Minneapolis: University of Minnesota Press.
White, M.T. (1986). Self-relations, object relations, and pathological narcissism. In
A. Morrison (Ed.), Essential papers on narcissism (pp. 144-164). New York:
New York University Press.
Winnicott, D. (1965). The maturational processes and the facilitating environment.
Wolff, P. (1960). The developmental psychologies of Jean Piaget and psycho-
analysis. New York: International Universities Press.
7
Integrity and Aging: Ethical,
Religious, and Psychosocial
Perspectives
F. CLARK POWER, ANN R. POWER, AND JOHN SNAREY
Personality theorists have for many years dealt with the question of how
adults successfully cope with the somatic, psychological, and social losses
that are often characteristic of the final age period of the life cycle. Some
have proposed that "successful" aging depends upon developing new
strategies for adapting to the challenges of this period. Others have
stressed the role of personality traits developed much earlier in the life
cycle. Over 25 years ago, for example, Cumming and Henry (1961) touted
the highly controversial view that successful aging entails a withdrawal or
"disengagement" from social roles and relationships. Their critics argued
the opposite-that life satisfaction in old age requires maintaining a high
activity level by replacing lost roles and relationships (e.g., Lemon, Beng-
ston, & Peterson, 1972). Research by Neugarten, Havighurst, and Tobin
(1968), although somewhat supportive ofthe activity theory, indicated that
personality type is a more important determinant of life satisfaction. They
found that persons classified as "integrated" reported high life satisfaction
whether they were active or disengaged.
This debate illustrates the need to examine age-related personality
changes in the context of more stable features of the self. Furthermore, it
raises the deeper issue of whether measures of adaptation or life satisfac-
tion adequately assess "successful" aging. We side with John Stuart Mill
(1863/1967) who, in arguing for a hierarchy of values, noted: "It is better to
be a human being dissastisfied than a pig satisfied; better to be Socrates
dissastisfied than the fool satisfied" (p. 902).
In this chapter, we employ Erikson's notion of integrity in describing an
ideal of successful aging consistent with ethical and religious standards
of maturity. We elucidate the meaning of integrity by showing how it
Research for this paper was supported by a grant from the Mellon Foundation to
the Henry A. Murray Research Center for the Study of Lives. We wish to thank
members of the Center's staff for their help, and Richard Shulik for donating his case
studies to the Center. We also wish to express our gratitude to James Fowler for
sharing some of his case studies with us.
7. Integrity and Aging 131
includes both social age-period and structural stage characteristics within a

functional phase model of development. Through an analysis of how aging
adults address four component themes of integrity, we demonstrate some
of the similarities and differences between age-period and stage
approaches and suggest how they may dialectically contribute to a more
comprehensive understanding of aging individuals.
An Integrative Approach to Integrity and Aging

STAGE, AGE, AND PHASE MODELS OF EGO DEVELOPMENT
We analyze integrity and aging along the theoretical lines recommended by

Snarey, Kohlberg, and Noam (1983), who distinguish three types of
approaches to understanding ego development: structural stages (e.g.,
Kohlberg, 1984), social ages (e.g., Neugarten, 1977), and functional
phases (e.g., Erikson, 1962). Unfortunately, the concepts of stages, age
periods, and phases have been used confusedly, as if they were more or less
synonymous. Thus, we first summarize the distinctions between these three
types of models:
1. Stages focus on transformations in the process of thinking, and ages are
tied more directly to the content or reasoning, while phases clarify how
structural together to give rise to new developmental functional tasks.
2. Stage models emphasize development rather than aging, social age-
period models are more directly linked to chronological age, and phases
are based on the synchronization of structural development and social
aging.
3. Stages represent qualitative differences in development, but social ages
place greater emphasis on quantitative changes, while phases involve
both qualitative and quantitative changes.
4. Stages emphasize an active individual, age periods emphasize an active
society, and phases are characterized by new psychosocial tasks that
grow out of the interaction of the individual and social action.
S. Stages are unconscious cognitive structures, and age periods are based
on the conscious passage of time, while phases are preconscious.
6. Stages are hierarchical and age periods are not universally hierarchical;
phases are not hierarchical in terms of complexity, but they are in terms
of their ability to give order to or make sense of life.
7. Stages are invariant in sequence and social age periods vary in sequence
between cultural groups; later functional phases do subsume earlier
phases, but the specific sequence of adjoining phases may be reversed
or difficult to distinguish.
8. Finally, stagaes are structurally universal phenomena, age periods are
socially relative, and various elements of phases may be universal or
relative.
132 F. Clark Power, Ann R. Power, and John Snarey
A detailed summary of the characteristics and assumptions of stage, age,

and phase models is presented in Table 7.1; a more detailed presentation is
available in Snarey et al. (1983).
ERIKSON'S FUNCTIONAL PHASE MODEL OF EGO DEVELOPMENT

The concept of ego, according to Erikson, is a "term which in the scholas-
tics stood for the unity of body and soul, and in philosophy in general for
the permanency of conscious experience" (1964, p. 147). Erikson's model
of ego development, from its debut in Childhood and Society (1950) to its
most recent formulation in Vital Involvements in Old Age (Erikson, Erik-
son, & Kivnik, 1986), has elaborated eight phases ("stages" for Erikson)
across the life span (with their corresponding ego strength): 1) trust versus
mistrust (hope), 2) autonomy versus shame and doubt (will), 3) initiative
versus guilt (purpose), 4) industry versus inferiority (competence), 5)
identity formation versus identity confusion (fidelity), 6) intimacy versus
isolation (love), 7) generativity versus self-absorption (care), and 8) in-
tegrity versus despair (wisdom).
These phases describe changes in ego functioning that are precipitated
by internal psychological development and external social role expecta-
tions. The focus of each phase is upon the maturing person's self as he or
she experiences new sociocultural spheres and roles. During each new
phase, the ego must reestablish itself. These changes are defined as age-
related "crises," involving particular tasks or concerns in psychosocial
functioning. Each crisis represents the intersection of psychomotor, cogni-
tive, psychosexual, and interpersonal development, which challenges the
self with a new developmental task in its interaction with its environment.
These developmental tasks with which the self is sequentially preoccupied
have a biological base in the life cycle and a cultural base in society's role
expectations that are synchronized with human maturation. According to
Erikson's epigenetic principle, however, each task is also present through-
out the entire life cycle: "Earlier stages are not replaced but are absorbed
into a hierarchical system of increasing differentiation" (1974, p. 206).
While each of the eight concerns is present throughout the life cycle, each
concern intensifies to a crisis level in an ordered sequence. Thus, ego
identity formation is a constant activity at each phase, even though it is not
until adolescence that it becomes an overwhelming preoccupation.
The synchronization of social age change and structural stage develop-
ment underlies each phase of Erikson's model. Consider the phase of
industry versus inferiority. In most of the world's cultural groups, children
"join" society around ages 5 to 7 (in the United States they begin
elementary school). During this period, they also generally achieve the
stage of concrete operations in cognitive development. The interaction of
social age change and cognitive stage gives rise to the child's ego function
ing task, which is to become socially industrious and to derive a sense of
TABLE 7.1. Contrasting characteristics of structural stage, cultural age, and

functional phase conceptions of human development.
Structural stages Functional phases Cultural ages
1. Process versus content.
Stages are based on trans- Phases illuminate how both Cultural periods are tied
formations in cognitive structural processes and more directly to the content
structures-in the way the cultural contents function of reasoning rather than to
mind processes the content together to give rise to new the process. New cultural
about which it is thinking. functional tasks in the indi- contents or knowledge are
These patterned processes of vidual's development. introduced in each age
thinking are an integrated period.
set of mental operations that
account for how the indi-
vidual makes sense of, or
performs, operations on the
contents of his world.
2. Development verus aging.
Stages are not based on par- Phases are based on the syn- Periods, as times of stability
ticular ages, although it is chronization of structural and transition in the life cy-
generally possible to give development (stage change) cle, are critically linked to
modal age ranges for each and cultural aging (aging in age. All societies divide their
stage. Chronological age the context of cultural ex- membership into age cate-
does not guarantee a corre- pectations). Similar to ages, gories (e.g., infant, lap
sponding stage of develop- they are somewhat inevi- child, yard child, elder).
ment; some adults are table; the next phase comes These function as taxonomic
fixated at stages typical of in a maturational sequence. devices to organize the
children and a precocious Similar to stages, the suc- process of status and role
child may be more mature cessful resolution of later changes within the life cycle
than age would predict. tasks is partially dependent and to establish the person's
on the resolution of prior participation in society in a
developmental crisis. way that takes into consid-
eration maturation, physical
energy, and needs.
3. Qualitative versus quantitative.
Stages represent distinct Phases involve both qualita- Culturally defined periods
qualitative structural differ- tive and quantitative change. put much greater emphasis
ences in thinking about and For instance, as a result of on quantitative changes in
orienting one's self to the qualitative changes in cogni- age, mastery, performance,
world. A child's develop- tive structures and quantita- knowledge, rights, and re-
mental stage is not simply an tive changes in social status, sponsibilities. Puberty rites,
immature version of adult the individual is faced with a for instance, often involve
meaning making, but is a new developmental task in exposure to greater quanti-
general organizing tendency ego functioning. ties of stress knowledge than
that is truly different from a younger person is per-
adults. Stages are total ways mitted to experience.
of thinking, qualitatively
different from other stage
approaches to the same
tasks.
TABLE 7.1. (continued)

4. Active individual versus active society.
Stages are not simply the Phases are characterized by Age periods emphasize the
result of internal factors new psychosocial tasks that active instructional function
(inherited maturation) or grow out of the interaction of the external social
external factors (nurturing of individual and social setting-the individual is
environment); they are action. The paradox of how acted upon by society.
forms of equilibrium new culturally programmed Whereas stages emphasize
constructed out of the roles are fully self-chosen cognitive operations,
interaction between an individual achievements periods emphasize cultural
individual and his milieu. is illuminated in part by cooperations. Social roles
Within this interactive the functional phase syn- are the primary mechanism
exchange, however, stages chronization of individual by which the needs of the
tend to stress the activity of and social action. individual are met in the
the internal organism on the process of meeting the needs
external environment. of society. Rites of passage
Activity is typically between periods foster
portrayed as taking place separation of an individual
within the mind, and from a pervious social
cognitive operations are sphere and incorporate the
defined as interiorized person into a new social
generalizable actions. role.
5. Unconscious versus conscious.
Stages are unconscious Phases are, to varying Age periods are based on
cognitive structures. In the degrees, both conscious. An the conscious passage of
manner of a child who adolescent is aware to some time, consciously observable
speaks grammatically but degree of the need to make events, and consciously
does so without being appropriate education of acquired cultural knowledge
consciously aware of the work choices and these and skills. The person is
grammatical structures, the choices are seen as defining consciously aware ofthe
ego is not consciously aware the self, but the person is not contents of his or her
of the structures on which fully aware of the identity thoughts. Periods are thus
the ability to construct achievement dynamics that highly conscious in most
meaning depend, although are taking place. cultures and fairly conscious
an adult's ability to think in all. Even in U.S. society,
about his own thinking does where adult periods of the
bring a general awareness of life cycle are inadequately
thought structures. objectified, they can be
Developmental change and brought into public
the stability of a new stage consciousness fairly easily
are also not the result of (e.g., Sheehy, Levinson).
conscious choice.
6. Hierarchical versus nonhierarchical.
Stage sequences are hier- Later phases are more Age periods are relatively
archical. A higher stage is adequate than earlier nonhierarchical. It is difficult
constructed on the previous phases, not necessarily in to say that a later period is
stage, reintegrating it into a terms of complexity, but in more adequate than an
more highly differentiated, terms of their ability to give earlier one because
flexible, and complex stage. order to or make sense of attainment of and the
Later stages are more ade- one's life in a form that is adequacy of performance at
TABLE 7.1. (continued)

quate than earlier stages stable and meaningful. a particular age period are
since they include earlier distinct. Furthermore, the
stage patterns, resolve the period that society defines as
same problems better, and life's best time varies from
are more justifiable in terms culture to culture.
of the universal inclusiveness
of their ordering of
experience.
7. Invariance versus variation.
Stages are invariant; each The achievement of earlier Cultural ages vary in se-
stage develops out of the phases or the resolution of quence between cultures
previous one and a person previous crises serves as a and between subcultures
must progress up the hier- foundation for new phases. within a particular culture.
archy one step at a time, with- Later phases subsume The accepted sequence in
out skipping or reversing any earlier phases. The one sociocultural system
of the stages. Although an successful achievement of may be reversed or absent in
individual may become later tasks is partially but not another. Even such general-
fixated at a particular stage, completely dependent on ly defined periods as adoles-
or even regress, all for-ward the successful achievement cence or the elderly can be
progress requires an of earlier tasks. The specific absent where the culturally
invariant sequence of sequence of adjoining defined life cycle takes a
development in accord with phases may be reversed or person directly from late
the stage hierarchy. difficult to distinguish among childhood into adulthood
Environmental factors and some individuals in our or where a person does
innate capabilities may help culture and many individuals not research full adulthood
one person reach a given in some other cultures. until very late in life.
stage of development earlier
than another, but all people
go through the same stage
sequence.
8. Universal versus relative.
Stages are structurally uni- Phases are neither as Cultural age periods are
versal phenomena. Stage universal as stages nor as local road maps that predict
theories form an interna- relative as cultural ages. the contents one will be
tional road map predicting More plastic than the former concerned with during
the sequential development and firmer than the latter, various ages of life in that
of the ability to structure or phases address both the particular culture. Periods
make sense of the world. All commonality and unique- cannot be universal since
persons, regardless of their ness of personal experience they vary tremendously from
sociocultural setting, can be and developmental conflicts. one culture to another. The
expected to go through the relativity of age periods
same stages. The order of between cultures, however,
forward movement is also implies a general
universal, although uniformity within a culture
individuals raised in or subculture.
different environments will
no doubt progress through
the stages at varying rates.
Note. From Snarey, Kohlberg, and Noam (1983, pp. 328-330).

competence from that achievement. Or consider the phase of identity ver-

sus identity confusion. The achievement of formal operations in cognitive
development (Le., the ability to engage in hypothetical deductive reason-
ing) and the social requirement of learning to fill specific adult roles are
synchronized in the American adolescent's functional task of identity
achievement and the working out of what one should do with one's life
(Snarey et al., 1983). Here we also see why phase models tend to blur such
distinctions as conscious versus unconscious. Adolescents in turmoil are
mOre Or less aware that they are in a time of identity crisis, while they are
unconscious of the overall course of ego development, just as they do not
know that they are also moving to the second substage of formal opera-
tions.
Functional phases of development appear to be facilitated by the optimal
interaction Or matching of structural stage possibilities and social age re-
sponsibilities. As Erikson states, "The very process of growth provides
new energy even as a society offers new and specific opportunities accord-
ing to its dominant conception of the phases of life," (1968, p. 163). Thus,
"it is the joint development of cognitive and emotional powers paired with
appropriate social learning which enables the individual to realize the
potentialities" of each phase of development (1964, p. 225). Of course, as
Erikson also suggests, the ideal of coordinated interdependence and
optimum mutual activation between personality and society within each
psychosocial phase of development is never perfectly achieved. It follows
that phases of psychosocial functioning can be supported, fixated, Or
retarded as a result of sociocultural conditions. In particular, the phase
of integrity versus despair presents a turning point that can lead to a vital
adaptation to the cultural environment Or to an impaired maladaptation. If
the adaptation is positive, however, then the ego strength of wisdom
emerges.
INTEGRITY IN OLD AGE

The word integrity literally means "untouched" (in [not], tangere [to
touch]) and suggests structural and functional soundness, completeness
and comprehensiveness, purity, and authenticity. Erikson (1962) and Erik-
son et al. (1986) use it to describe the "syntonic" Or positive disposition in
aging and oppose it to despair, the "dystonic" disposition. While integrity
versus despair constitutes the final and fundamental option of human exis-
tence, integrity can only be lived in a dynamic balance with despair. Loss
and disintegration leading to death can give rise to regret, cynicism, and
hopelessness, but integrity demands that they be accepted consciously and
courageously.
Although senescence brings with it loss and decline, it can also bring
wisdom, the virtue of old age. With his colleagues, Erikson (Erikson et al.,
1986) defines wisdom as a "detached concern with life itself, in the face of
death itself' (p. 37) or as "involved disinvolvement" (pp. 50-51). In the

Greek philosophical and Judeo-Christian religious traditions, wisdom is a
comprehensive kind of knowledge that includes both theoretical and prac-
tical understanding. It entails the ability to use experience in grasping fun-
damental principles that integrate and give substance to all that one knows.
In religious terms, to be wise means to identify with and surrender to the
Creator who gives life its ultimate meaning and purpose. When one is truly
wise, then one will find true happiness and life itself:
Happy the one who obeys me,
and happy those who keep my ways, ...
For the one who finds me finds life,
and wins favor from the Lord;
But they who miss me harm themselves
and all who hate me love death. (Proverbs 9:33, 35-36)
The notions of integrity and its companion virtue, wisdom, presuppose
the development of a new sense of self or of identity. The many changes
that occur in old age, from gray hair and wrinkles to retirement and grand-
parenthood, demand a corresponding change in one's reflexive self-
awareness. Furthermore, cumulative life experience should contribute to a
more "seasoned" sense of self. On a more profound level, loss and im-
pending death call for an identity that includes as well as transcends all
temporal identities, an identity that is accepted as well as achieved. The
integrated identity emerges from a dual process of "reckoning" and "relin-
quishing." In reckoning, one evaluates the self-made identity-an identity
constructed out of decision and action throughout the life span. Relin-
quishing completes the process through a surrender of the self-made identi-
ty and a recognition of a deeper sense of the self as part of a larger whole.
INTEGRITY AS A SOCIAL AND COGNITIVE CONSTRUCT

Applying the functional phase notion of integrity to the empirical study of
aging adults requires a more careful analysis of the structural stage and
social age components of the issues raised at the end of the life cycle. From
a philosophical standpoint, these issues may bst be classified as "religious"
or metaphysical, in so far as they arise at the limits of existence and
rationality and involve what the theologian Paul Tillich (1952) calls onto-
logical anxiety-a fear about no particular object but about nothingness,
nonbeing, and meaninglessness. That anxiety comes to the fore when we
contemplate suffering, solititude, and death, the legacy of old age. Tillich
describes "courage" as the necessary response to these threats. Erikson
speaks of integrity and wisdom. This virtue language emphasizes the im-
portance of the will affirming being in the face of nonbeing. Yet courage,
integrity, and wisdom imply more than an emotivistic response to life's
ultimate problems.
In our view, questions about life's meaning and value, which arise in the
experience of ontological anxiety, have a cognitive core. They are ques-
tions about the "character" of the universe and its relationship to the hu-
man subject who is seeking meaning, truth, and happiness. Ordinarily we
assume that the universe supports our deepest human longings. We have a
tacit or unconscious faith in the meaning and worth of human existence
that makes our "ordinary" activities possible. For example, in trying to
lead a moral life, we attend to the dictates of conscience and take for
granted that acting morally makes sense. Nevertheless, in times of reflec-
tion or in times of crisis, we sometimes ask with Job, "Why be moral?".
This question leads to others: "Why be rational?"; "Why live?". Such
radical inquiry into the meaningfulness of the very foundations of our
humanity forces us to confront the limitations of human existence and to
wonder whether we are grounded in an Ultimate or Transcendent Pres-
ence. Thus, we can think of the inquiry itself as "religious," whether or not
the response makes an appeal to God.
Note that religious questions express concerns that are quite different
from our more mundane worries and problems. Usually we do not enter-
tain doubts about why we should live, but we frequently ask how we should
live. Religious questions are thus questions about the limits and boundaries
of life and thought (Tracy, 1975). Because they come at the edges of our
lives, we can refuse to think about them and still go about our everyday
business. Nevertheless, if we deny them or we fail to come to terms with
their import, our lives would be less than integrated or whole.
Research on faith and religious development by Fowler (1981) and Oser
(1980) indicates that most children and adults do ponder limit questions,
although many do not answer them by referring to the supernatural. Further-
more, their research shows that the responses individuals make to limit
questions vary in terms of their cognitive differentiation and integration
and may be assessed within a hierarchical stage framework. One of the
characteristics of higher stages of faith and religious development is the
ability to identify the specifically religious or metaphysical import of limit
questions. For example, at higher stages, individuals understand the differ-
ence between a moral justification for a particular decision and a meta-
physical justification for why one should be moral at all. At lower stages,
confusion is common. For example, younger subjects sometimes say that it
is important to obey because that is what God wants. On the other hand,
they frequently argue that it is important to be moral to stay out of trouble.
These responses indicate a failure to grasp the import of religious and mor-
al questions. While subjects at lower stages think about religious questions,
they have not sufficiently differentiated them from questions that arise in
other domains of life. Thus they do not appreciate just how radical they
really are.
Only at the higher stages of reasoning about limit questions do persons
fully comprehend the great crisis, summarized in Erikson's eighth phase of

integrity versus despair. That crisis is very much an ontological one. It
involves a final choice between two stark possibilities: to be or not to be.
The choice of integrity entails rejecting the false kinds of assurance that
deny finitude and death. It involves accepting the meaninglessness of life
when seen from a finite perspective and identifying with an infinite or cos-
mic perspective (Fowler, 1981; Kohlberg & Power, 1981). That infinite
perspective offers a new horizon of self-awareness that transforms previous
self-understandings by overcoming the dualities of subject and object,
active and passive, free and determined.
A Functional Window on Integrity and Aging

SHULIK'S RESEARCH ON STAGES IN AGING
Richard Shulik (1979) has conducted the most extensive study to date of
faith and moral reasoning in aging. He administered Fowler's Faith De-
velopment Interview to a sample of 40 men and women (ranging in age
between 60 and 86) and Kohlberg's Moral Judgment Interview and Loevin-
ger's Sentence Completion Test to subsamples of 26 and 22 subjects. He
found strong correlations among all three measures, indicating that all
three tap a structural "core." The faith scores were distributed somewhat
in the shape of a pyramid: 17 were between Stages 3 and 3/4,14 at Stage 4,
and 9 between Stages 4/5 and 5. The faith stages are summarized in Table
7.2. Shulik found no examples of Stage 6, although Kohlberg and Power
(1981) and Power, Power, & Snarey (1986) have rescored two at that stage.
Integrating these findings with Fowler's (1981) presentation of his cross-
sectional faith development data from subjects of all ages, we find that the
few subjects scored as Stage 6 are all older than 60. This provides some
support for Kohlberg and Power's (1981) speculation that the functional
crisis of aging induces development to Stage 6.
Shulik's finding of only a very small positive correlation between faith
stage and chronological age suggests that most aging adults' faith develop-
ment has stabilized. Nevertheless, it leaves open the possibility of function-
al change. Although Shulik did not attempt to assess this directly, he did
measure subjects' "age sense" or their reflective awareness of having
changed in old age. He rated responses to questions about the experience
of getting older on one of three levels of age sense:
Levell: No awareness of change
Level 2: Awareness of only external and circumstantial changes (e.g., re-
tirement)
Level 3: Awareness of internal changes in personality characteristics, tem-
perament, interests, philosophy, and intellectual style
TABLE 7.2. Faith stages by aspects.

Aspect
C. Form of moral
A. Form of logic B. Perspective judgment D. Boundsof
Stage (Piaget) taking (Selman) (Kohlberg) social awarne~s
I Preoperational Rudimentary Punishment-reward Family, primal

empathy others
(egocentric)
II Concrete oper- Simple perspec- Instrumental "Those like us"
ational tive taking hedonism (recip- (in familial,
rocal fairness) ethnic, racial,
class and religious
terms)
III Early formal op- Mutual inter- Interpersonal ex- Composite of
erations personal pectations and con- groups in which
cordance one has interper-
sonal relation-
ships
IV Formal opera- Mutual, with Societal perspec- Ideologically
tions (dichoto- self-selected tive, reflective compatible com-
mizing) group or class relativism or class- munities with
(societal) biased universalism congruence to
self-chosen norms
and insights
V Formal opera- Mutual with Prior to society, Extends beyond
tions (dialectical) groups, classes principled higher class norms and
and traditions law (universal and interests; disci-
"other" than critical) plined ideological
one's own vulnerability to
"truths" and
"claims" of out-
groups and other
traditions
VI Formf I opera- Mutual, with Loyalty, to being Identification
tions (synthetic) the common- with the species;
wealth of being trans-narcissistic
love of being
Note. From Fowler (1981, pp. 214-215).

Aspect
F. Form of world
E. Locus of authority coherence G. Symbolicfunction
Attachment/dependence Episodic Magical-numinous

relationships; size, power,
visible symbols of authority
Incumbents of authority Narrative-dramatic One-dimensional; literal
roles, salience increased by
personal relatedness
Consensus of valued group Tacit system, felt meanings Symbols miltidimensional;

and in personally worthy symbolically mediated, evocative power inheres in
representatives of belief- globally held symbol
value traditions
One's own judgment as Explicit system, concep- Symbols separated from

informed by a self-ratified tually mediated, clarity symbolized; translated (re-
ideological perspective; about boundaries and inner duced) to ideations; evoca-
authorities and norms must connections of system tive power inheres in mean-
be congruent with this ing conveyed by symbols
Dialectical joining of Multisystemic symbolic and Postcritical joining of irre-

judgment-experience pro- conceptual mediation ducible symbolic power and
cesses with reflective claims ideational meaning; evoca-
of others and of various ex- tive power inherent in the
pressions of cumulative reality in and beyond sym-
human wisdom bol and in the power of un-
conscious processes in the
self
In a personal judgment in- Unitive actuality felt and Evocative power of sym-
formed by the experiences participated unity of "One bols actualized through un-
and truths of previous beyond the many" ification of reality mediated
stages, purified of egoic by symbols and the self
striving, and linked by dis-
ciplined intuition to the
principle of being
The interview responses were almost evenly distributed across these levels.
More importantly, there was a very high positive correlation between faith
stage and age sense. Only 1 of the 14 subjects, who scored between Stages
3 and 3/4, was rated at the third level, compared to 6 of 9, who scored
between Stages 4/5 and 5. These results raise serious doubts about the
generality of a functional phase of integrity versus despair because the
experience of a crisis in aging seems to depend upon stage.
In defense of the functional model, Shulik's measure of age sense may
require more reflection than necessary. It is conceivable that aging adults
may deal with functional concerns without an explicit awareness of changes
in their thinking or personality. Furthermore, age sense should not in and
of itself be associated with a specific phase in the life span. Adolescents and
younger adults are equally capable of self-reflection and may have as much
need to reflect on age-related changes as older adults.
THEMES OF INTEGRITY
The Shulik study challenged us to demonstrate that there are ethical and
religious concerns common to the "aged" period of the life cycle but that
aged persons functioning at different stages will differ in the achievement
of integrity. We tried to assess these functional phase concerns by iden-
tifing four themes implicit in Erikson's functional description of integrity.
We then derived definitions for each theme through an analysis of the Faith
and Moral Judgment Interviews of a subsample of 9 of Shulik's subjects
who were scored between Stages 3 and 5. The four themes are aretaic
review, acceptance of limits, detachment, and holistic reflection. In the
following sections, we describe each one of them, note their relationship
to the master theme of integration, and provide examples from the nine
"construction cases."
Aretaic Review
The term aretaic is derived from the Greek term arete, commonly trans-
lated as "virtue." According to Aristotle, the arete of things or persons is
that state that disposes them to perform their function well (Irwin, 1985,
p. 430). For example, the arete of one's car will be its dependability, safety,
fuel economy, maneuverability, and so forth. The arete of persons will be
that which makes them good persons. Thus arete includes not only moral
qualities but nonmoral excellences as well. We wish to retain this very
broad meaning of arete in analyzing how aging adults evaluate the worth of
their lives. Questions of personal moral worth and value arise throughout
the life cycle, but we would expect them to present themselves with par-
ticular force and in a special way toward its end. To paraphrase Aristotle,
"Integrity requires both a complete life and complete virtue." Since the
better part of their lives have been lived and their characters have been
more or less determined, aging adults can sum up and evaluate who they
have become and, in a sense, will forever be. Such a retrospective evalua-
tion can bring with it deep-rooted happiness if their lives have, in fact, been
virtuous. As a stable and internal good, virtue, never finally secured until
the end of life, can withstand the onslaughts of senescence and misfortune.
We found many examples of aretaic review in the construction cases.
Typically, the aging subjects engaged in retrospective evaluation in re-
sponding to the theodicy issue: "Why do the just suffer while the unjust
prosper?" Mr. Wine maintains that justice brings its own rewards, or at
least it did for him: "I am honest, have good principles and they have
served me well." Another, Mr. Harris, struggles with the issue:
I, ah, I have tried to lead an honest life. Let me cite an instance that will give you a
good understanding of what I am. [Gives examples of refusing to buy a stolen
automobile battery and of refusing to cooperate in a scheme to avoid paying sales
tax.] Those [dishonest] people irritate me so much because I am one of the good
guys, and they made me into a sucker .... or so they think in their terms. Because
of my upbringing and the models presented to me by my mother and father I be-
came an honest man. And I will go so far as to say that I am proud of this fact.
Because money is never so terribly important to start with.
Honesty did not payoff for him as it seems to have for Mr. Wine. Never-
theless, he says he is "proud" of being honest.
Mr. Harris's response to the Job issue is not scored at a very high faith
stage. He answers the question, "Why be moral?" or "Why be honest?" by
referring to his upbringing and his devaluation of money, concerns that
would probably be scored as Stage 3 on Fowler's scale (see Table 7.2).
Note, however, that aretaic review and the other categories are not meant
to refer to a particular stage of reasoning but are "content" categories that
are, in principle, compatible with all of the stages of reasoning exhibited by
adult subjects. The category scheme allows us to ask whether the themes
we expect to find in the functional phase of identity versus despair are, in
fact, present. We may then ask the question of whether these themes are
articulated in qualitatively different ways. Mr. Harris's reply indicates the
significance of aretaic review for his very self-definition. He claims that his
acts of honesty will give us a "good understanding" of "what" he is.
In describing the function of aretaic review, we have stressed its impor-
tance as a means of maintaining a sense of self-esteem. A related function
is to utilize the past as a resource in facing the future. Mrs. Wilson illus-
trates such a function by remarking:
I figure I should make mine the best life I can from what I know about living in that
I try to do the right things and not the wrong things. For if I have to be here, I might
as well live as good a life as I can, and I think that this is the only way you can lead a
life that does not lead to total despair.
Before making this statement, Mrs. Wilson had been discussing her beliefs
about the meaninglessness of life, beliefs that have become stronger in her
old age. She summons the courage to face life in spite of its futility by
drawing on her past ("from what I know about living"), a past that has
taught her to be moral ("to try to lead as good a life as I possibly can").
Not all of the subjects in our construction case sample judge their lives as
virtuous or worthwhile. Some mention regrets and feelings of despair. Mrs.
Meyer notes several times on her interview how much she wishes that she
could undo her past failures:
I'm trying not to fear death because they say you shouldn't fear death. . . . It
seems final. . . . I think back and I wish things had been different. . . . I find I wish
I had talked with my mother and father about some things and had been closer to
them .... When my father was in the hospital I didn't go to visit him and I feel
badly because I wasn't more understanding.
Other subjects speak of failures to advance in their jobs, spend time with
their families, further their education, and so forth. However, unlike Mrs.
Meyer, they seem to have come to terms with their misgivings. As one
woman puts it, "There are a few things I regret but on the whole I would
not do it any other way."
Acceptance of Limits
The second component of integrity is the acceptance of limits. In old age,
life is nearly over, one's physical and mental powers are declining, and role
responsibilities are diminishing. What the self cannot be becomes more
salient than what the self can be. It is in this humble acknowledgment of
finitude that the elderly may manifest a special kind of tolerance of them-
selves and others. For example, Mr. Dawson confessed, "I was more judg-
mental and impatient when I was younger. Now I don't think that bothers
me any more." In the context of discussing how his family got along, Mr.
Smith pointed out:
When you get older, you could have had all the difficulties when you were young,
but when you get older you start growing together. You forget and you forgive.
You are understanding them better.
Subjects also seem to grasp, if only intuitively, the Socratic wisdom of
knowing their own ignorance. Mrs. Wilson admits that neither she nor
anyone else had the answers to life's most important problems: "Nobody
knows the why and wherefore of anything. All you can do is accept on faith
this much-well here we are, and, well, yes." Mr. Smith does not stress
universal ignorance as much as humility. He takes no credit for his own
intelligence but calls it and other talents "gifts." He concludes that growing
older has taught him "how little you do by yourself."
Detachment
The third of the categories is detachment. Detachment does not mean
withdrawal or disengagement, rather it means a distancing of one's self
from one's involvements so that one may see them for what they are. In the
major religious traditions, detachment from "wordly values and view-
points" is necessary for union with God. The Christian tradition recom-
mends a spiritual poverty as a way of being in the world but not of the
world. The Hindu classic, The Bhagavad Gita, stresses the importance of
the yoga of action (karmayoga)-renunciation in action. There Krishna
opposes the view that a contemplative withdrawal from action is the way to
God. Action is unavoidable, Nature and God are active. The spiritual life
demands the renunciation of the "fruits of action": "pleasure and pain,
profit and loss, victory and defeat" (11.38).
As we noted, detachment or "involved disinvolvement" is at the heart of
wisdom. Cumming and Henry (1961) seem to have had that same insight in
their disengagement hypothesis, in spite of their difficulty in operationaliz-
ing it. We found evidence for detachment in Mrs. Wilson's humorous
perspective of looking down on life, as if she were with God in heaven.
She says:
I get the biggest kick out of this universe and the people here. And I just know that
God is sitting up there holding himself and laughing at the way people are scram-
bling around down here.
We also identified detachment in Dr. Stallworth's description of a newly
achieved ability "to face varying circumstances with greater equanimity."
He elaborates the view that as he gets older he becomes less personally
involved in the outcomes of his deliberation which allows him to sort out
the relevant issues and ignore minor details:
I feel that I am at the point where I can distinguish between the major conse-
quences and the details that build up to them, that are irrelevant in the kind of a
situation where you're trying to briefly summarize what a situation is .... I'm not
so involved in the outcome one way or another. Or even if I put a fair amount of
work on something I don't care about the details of it anymore. I'm concerned with
the general conclusion.
Holistic Awareness
The last of the categories is holistic awareness. In his first formulation of
the concept of ego integrity, Erikson (1962) wrote: "It is a post-narcissistic
love of the human ego-not of the self-as an experience which conveys
some world order and spiritual sense, no matter how dearly paid for" (p.
268). Later, Erikson et al. (1986) use terms such as "spiritual personality"
or "all human and existential identity" to describe a transcendent ego. This
new identity is characterized by a heightened awareness that they describe
as "central rather than peripheral in space and time ... continuous rather
than scattered, indivisible rather than divided, inclusive rather than iso-
lated and excluded ... " (p. 52).
In assessing interviews for holistic awareness, we looked for explicit in-
dications that the subjects perceived their lives as an organic unity and saw
themselves as part of a larger world order. Dr. Stallworth speaks of seeing

his whole life as a "continuum" and of his personal identity as part of a
general "human equation." Although he is an atheistic scientist, he states
that there is a universal human need for some kind of philosophy of life
that maintains, provides a frame of reference for, and supports the self:
I think most people need a basic philosophic idea or thought to maintain them in
the course of their existence. And I feel that religion provides a frame of reference
or self-support. If one wants to say that everyone needs a philosophy of life and
amongst the philosophies of life there is organized religion, well, that can play the
same role for most people who have taken it up.
Miss Simpson, a subject discussed in considerable detail elsewhere (Kohl-
berg & Power, 1981), exhibits a consciously religious identification with
God and the cosmos:
We start by seeking a power greater than ourselves. I don't think anyone can fail to
realize that there is a power beyond themselves when they look out at the scene of
their own neighborhood to say nothing of the cosmos. I don't think it matters a bit
what you call this power, but it is within every mind and experience and makes us
aware of the oneness, not only of all people but of all life.
From the perspective of structural stage theory, the functional phase
themes are "content categories" that in principle may appear at any stage.
The stages are ways of thinking and problem solving that reconstruct
the same content categories on different levels of meaning. In the studies
that we present, the Kohlberg Moral Judgment and the Fowler Faith
Development Interviews were used to assess to what extent these themes
really did represent "contents" distinct from stage structure.
STUDY 1: ARETAIC REFLECTION AND MORAL JUDGMENT

In this study, two sample groups were selected: 20 aging subjects between
the ages of 60 and 86, and 20 adolescents and adults between the ages of 13
and 35. The subjects in both groups were nearly evenly divided between
males and females and represented diverse occupational, educational, and
religious backgrounds. Both groups were administered Kohlberg's stan-
dard moral dilemmas and asked questions designed to elicit prescriptive
moral reasoning about what the protagonist in the dilemma should do and
why. The question that originally motivated this study was whether there
were differences between subjects in these two age groups that could not
be explained by their stage of moral reasoning. A comparative analysis of
nine construction cases from each sample had indicated that the aging
adults were more likely to engage in aretaic reflection about their character
in the context of responding to the dilemmas than were the younger sub-
jects. Although subjects from both groups took the task of responding to
the dilemmas seriously, many of the aging subjects seemed equally con-
cerned to relate their resolution of the abstract dilemma to similar deci-

sions in their past.
As an illustration of this interest in aretaic reflection, we quote from the
interview of a 73-year-old man, who was administered the Heinz Dilemma,
which asks whether Heinz should steal a drug if it were the only means of
saving his wife's life. He responds:
I would say in desperation he did the right thing. I myself had experiences like that.
I remember there was a family by the name of Baxter. I had a grocery store. Mrs.
Baxter had nine or eleven children. . . and Mr. Baxter ran off somewhere. . . Here
she accumulated a debt of several hundred dollars, which was an awful lot in those
days. So this night I said I'll take a ride there (to see if I could collect some
money) ... I went to the door; a little kid opened it. I walked in the house and I
shall never forget what I saw. She was in bed with the grippe and the children were
running around with snotty noses. It was freezing in the house and complete dark-
ness. . . . So you know I went back to the store and opened up. . . I loaded my car
with groceries. . . and I took back some coal we used to have in the house and I
went back to the house .... When I can see the desperation of that man Heinz, to
break in [to steal the drug] is not so bad.
This subject goes on to recount another experience in which he managed to
procure a rare drug for his own wife who was quite ill. He concluded, "But
you see, I got it. I would have stolen it .... And yet my background and
my character would show that I have never disobeyed the law."
It is not unusual for subjects of all ages to describe how they would
resolve the Heinz Dilemma. Typically they do so either because they doubt
that they will live up to their ideals or because they resist the idea of mak-
ing a universal moral prescription (e.g., "I cannot say what Heinz should
do, I can only tell you what I would do"). This subject does more than
simply state what he would do if he were Heinz. He tells us what he did do
in somewhat similar circumstances. More importantly, he uses his review
of the past to demonstrate that he is a person of character and then he
works back from this evaluation of his character to his response to the
Heinz Dilemma. If all he did was to tell us about similar experiences in his
past, his reminiscence would have indicated a predictable but uninteresting
age-period difference. We would expect that because older people have
quantitatively more life experience, the dilemmas would be more likely to
remind them of similar events in their past than they would younger per-
sons. On the other hand, to engage in a moral appraisal of the past and to
use this appraisal as a way of responding to a moral dilemma suggests a
mode of response much closer to a functional sense of wisdom.
The results of the full-scale analysis of moral judgment interviews should
that 75% of the aging subjects engaged in some aretaic reflection, while
only 15% of the younger subjects did so. There was no relationship be-
tween the subjects' stage of moral reasoning and whether or not they en-
gaged in aretaic reflection, indicating that aretaic reasoning is a social age-
period theme that is independent of structural stage. The spontaneity of
this aretaic reflection on an interview focused on prescriptive judgments

about the morality of actions rather than persons gives further evidence
that this theme is of genuine concern to older subjects.
STUDY 2: THE FOUR INTEGRITY THEMES IN THE FAITH INTEGRITY
As in the previous study, we compared an aging group with an adolescent

and young-adult group. Twenty subjects from each group were adminis-
tered Fowler's Faith Development Interview. The interviews were rated
for the stage of faith development and for the salience of the four integrity
themes. The data indicate that the themes of aretaic reflection and the
acceptance of limits are more salient for the aging subjects than the younger
ones. One the other hand, the themes of holism and detachment appear to
be related to stage not age. Of the four subjects who expressed either one
or both of these themes, three were assessed as being Stage 4/5 or higher
on the faith interview. Although the aging subjects at all stages engaged in
aretaic reflection and demonstrated concern for accepting limits, the likeli-
hood that they would demonstrate a strong interest in those themes was
positively correlated with their stage of reasoning.
The finding that aging adults of all stages engage in aretaic reflection in
itself indicates not that they possess integrity, but only that they are en-
gaged in evaluative self-reflection. We may then ask whether there is any
relationship between positive judgments of self-worth and stage. In order
to assess self-worth, we classified each subject's aretaic reflection as either
positive, ambivalent, or negative. We discovered that there is a surprising-
ly strong relationship between stage and the self-worth judgment. The only
subjects who made clearly negative judgments of self-worth were scored
as Stage 3 on the Faith Development Interview. Subjects who made am-
bivalent judgments of self-worth scored in the middle range of stages
(Stages 3/4 and 4). All of the subjects at the highest stages (Stages 5 and 6)
expressed quite clear positive self-evaluations. This suggests that the
achievement of integrity in a functional phase sense is dependent upon
both social aging and stage development. The relationship between stage
and self-worth judgments may be based on the fact that the stages reflect
increasingly adequate modes of reflection on limitations and sources of
meaning and value. Subjects at the higher stages are better equipped to
deal with the losses that come at the end of the life cycle because they can
see them within a larger framework. At the lower stages without such a
framework these losses can become unbearable. For example, several of
the despairing subjects at Stage 3 mentioned that they were overwhelmed
at the loss of their spouses. In a sense, they had made their spouses their
centers of value and ultimate worth; and they were unable to envision a
meaningful existence without them.
Conclusion
In this chapter we relate functional phase, social age, and structural stage
models to the study of ego integrity. Through an analysis of Erikson's con-
cept of integrity, we identify four component themes. We find that two
seem to be dependent upon stage rather than social age, and that the other
two are functionally related both to social age and structural stage. In an
analysis of the judgments of self-worth, we also find a relationship between
stage and positive self-evaluation, indicating that stage development may
help aging adults to cope with the losses of later life and achieve integrity.
Our results indicate the fruitfulness of a dialectical approach to the de-
velopmental study of aging. Rather than assimilating one model within
another, we have tried to indicate some of the ways in which the social
aging and structural stage models interactively contribute to the under-
standing of the functional phase of integrity versus despair. Erikson's
functional phase model provides a more inclusive framework for the study
of ethical, religious, and psychosocial development in aging; but the model
does not capture some of the stage-related differences in dealing with the
conflicts of each age period. His idealized conceptions of integrity and wis-
dom presuppose relatively rare structural development, as we discovered
in pilot research showing that only those at the highest stages of faith
articulated themes of detachment and holism. A more sensitive and prom-
ising approach to the empirical study of ethical and religious development
in aging seems to call for the dialectical interplay of both stage, age-period,
and phase models.
REFERENCES
Cumming, E., & Henry, W.E. (1961). Growing old. New York: Basic Books.
Erikson, E. (1950). Childhood and society. New York: Norton.
Erikson, E. (1962). Childhood and society (2nd ed.). New York: Norton.
Erikson, E. (1964). Insight and responsibility: Lectures on the ethical implications of
psychoanalytic insight. New York: Norton.
Erikson, E. (1968). Life cycle. In International encyclopedia of the social sciences
(pp. 286-292). New York: Macmillan.
Erikson, E. (1974). Dimensions of a new identity. New York: Norton.
Erikson, E., Erikson, J., & Kivnik, H. (1986). Vital involvement in old age. New
York: Norton.
Fowler, J. (1981). Stages of faith: The psychology of human development and the
quest for meaning. San Francisco: Harper & Row.
Irwin, T. (1985). Notes. In Aristotle, Nicomachean ethics. Indinanapolis: Hackett.
Kohlberg, L. (1984). Essays on moral development: Vol. II. The psychology of
moral development. New York: Harper & Row.
Kohlberg, L., & Power, C. (1981). Moral development, religous thinking, and the
question of a seventh stage. Zygon, 16,203-260.
Lemon, B.W., Bengston, V.L., & Peterson, J.A. (1972). An exploration of the
activity theory of aging: Activity types and life satisfaction among in-movers to a
retirement community. Journal of Gerontology, 27, 511-523.
Mill, J.S. (186311967). Utilitarianism. In E.A. Burtt (Ed.), The english philosophers
from Bacon To Mill. Toronto: Random House.
Neugarten, B.L. (1977). Personality and aging. In J.E. Birren & K.W. Schaie
(Eds.), Handbook of the psychology of aging. New York: VanNostrand
Reinhold.
Neugarten, B.L., Havighurst, R.J., & Tobin, S.S. (1968). Personality and patterns
of aging. In B.L. Neugarten (Ed.), Middle age and aging: A reader in social
psychology. Chicago: University of Chicago Press.
Oser, F. (1980). Stages of religious judgment. In J. Fowler & A. Vergote (Eds.),
Toward moral and religious maturity. Morristown, NJ: Silver-Burdett.
Power, C., Power, A.R., & Snarey, J. (1986). Faith and moral development among
the aged: Structural and functional perspectives. Paper presented to the American
Psychological Association, Washington, DC.
Shulik, R. (1979). Faith development, moral development, and old age: An assess-
ment of Fowler's faith development paradigm. Ph.D. dissertation, University of
Chicago.
Snarey, J., Kohlberg, L., & Noam, G. (1983). Ego development in perspective:
Structural stage, functional phase, and cultural age-period models. Developmen-
tal Review, 3, 303-338.
Tillich, P. (1952). The courage to be. New Haven, CT: Yale University Press.
Tracy, D. (1975). Blessed rage for order. New York: Seabury Press.
8
The Relationship Between Ego and
Moral Development: A Theoretical
Review and Empirical Analysis *
LOREN LEE AND JOHN SNAREY
Over the last two decades, a persistent but unresolved question within
structural-developmental psychology has involved the relationship be-
tween ego development (I), as defined by Jane Loevinger, and moral judg-
ment development (M), as defined by Lawrence Kohlberg (e.g., Erickson,
1977a, 1977b; Hauser, 1976; Kohlberg, 1964, 1966, 1984, 1986; Lambert,
1972; Loevinger, 1976, 1986; Snarey, 1986; Snarey, Kohlberg, & Noam,
1983; Sullivan, McCollough, & Stager, 1970). Both theorists present inde-
pendent (though related) stage models of human development and both
have designed measures to assess these stages: the Sentence Completion
Test (SCT) and the Moral Judgment Interview (MJI). Although both
theorists have posited a conceptual correspondence between the two mod-
els, there is little known about the precise empirical relationship between
ego and moral stages.
Previous studies addressing this issue have produced conflicting portraits
of the relationship. Difficulties in defining the connection have been attri-
buted to obstacles ranging from disparate methods of assessing ego and
moral growth (cf. Snarey, 1986; Sullivan et aI., 1970) to inadequacies in
research design, most notably small sample sizes and the failure to antici-
pate the influence of moderating variables (cf. Haan, Stroud, & Holstein,
1973; Hauser, 1976). The present study explores the relationship between
ego and moral development through a comprehensive secondary analysis
of the combined subject data from nine previous studies. It incorporates
three dimensions of analysis previously unaddressed in the research: (a)
using standardized measurements, including like-scoring algorithms, to
* This study was conducted with the collaboration of the following individuals
who generously supplied data from their previous research: V. Lois Erickson,
Andrew Garrod, Stuart Hauser, Alan Jacobson, Lawrence Kohlberg, Marcia Ment-
kowski, Gil Noam, Sally Powers, Richard Shulik, and John Whiteley. Preparation
of this review was supported in part by a Woodrow Wilson Newcombe Fellow-
ship to Loren Lee and National Institute of Mental Health Grant MHl4088 to John
Snarey.
152 Loren Lee and John Snarey
compare ego and moral stages; (b) partitioning stage outcomes by indi-
vidual factors such as age and education; and (c) including longitudinal
subject measures as a means by which to monitor temporal changes in the
ego-moral relationship. By employing these dimensions, this review
addresses the long-standing question: What is the relationship between ego
and moral development?
Theoretical Review
The link between moral development and "ego theory," broadly con-
ceived, goes back as far as the former's inception. In his early writings,
Kohlberg (e.g., 1966, 1972) distinguished moral judgment from "ego
strength." In later years, as Loevinger's model became well-known, Kohl-
berg's references to the ego highlighted the relationship between these two
constructs. Loevinger (1976, 1978) also cited Kohlberg's work as one of
multiple intellectual roots of her theory. Despite these early allusions,
however, the two theories make somewhat different claims regarding the
nature of the constructs, the relationship between the constructs, and de-
velopment across stages. This review points to comparisons between the
two models and to central issues involved in clarifying their relationship
and integration.
NATURE OF THE CONSTRUCTS

Inner Logic
A principal difference between the two theories is that Kohlberg aims to
isolate "structure" alone in his model of moral progression; this means that
stages are determined by styles or levels of cognition apart from the specific
content matter about which the individual is reasoning. In contrast,
Loevinger derives her ego assessments from both the content and structure
of psychological functioning. A second difference in this area is Kohlberg's
concern with a philosophical rationale that serves as an infrastructure to his
stage model, hence his claim that higher stages are better stages (Kohlberg,
1981, 1984). Loevinger has indicated that she does not seek a philosophical
foundation for her work (Loevinger, 1986).
Nature of Stages
Originally, both schemata were thought to represent Piagetian progres-
sions describing: (a) invariant sequences of (b) hierarchical transforma-
tions, which are (c) structural wholes. Subsequently, Kohlberg suggested
that ego development ought to be conceived as a matter of levels rather
than as Piagetian structural stages, "since the unity of ego levels is not
that of logical or moral structures. . . . There are relatively clear criteria
8. Relationship Between Ego and Moral Development 153
of increased adequacy in logical and moral hierarchies, but not in ego

levels" (1984, p. 204). Loevinger, also, has tended to remove her theory
from any strict allegiance to a structural paradigm, indicating that the
processes of personality subsumed by the term ego development, while
"psychologically convincing," are best viewed as "broad and amorphous."
Such a domain, she implies, is best represented by a stochastic model of
probabilities, rather than by a sequential progression of structural wholes
(Loevinger, 1986).
RELATIONSHIP BETWEEN THE CONSTRUCTS

Construct Similarity
It is generally accepted that, at conceptual levels, striking parallels exist
between the ego and moral schemata. The similarity can be noted, for
instance, when comparing Loevinger's Stage 1-3 (conformity to external
rules and preoccupation with appearance and social acceptability) to Kohl-
berg's Stage M-3 (living up to what is expected by people close to you and
the need to be seen by others as a good person). Similarly, characteristics
of Kohlberg's Egocentric Stage M-1 (doing right to avoid punishment)
mirror components of the description for Loevinger's Impulsive Stage 1-2
(character controlled by fear of retaliation).
Despite the similarities at this level of comparison, however, there is also
good reason to view the two schemata as far from identical (cf. Lambert,
1972). Kohlberg's model, for instance, applies most directly to issues of
moral decision making and action; Loevinger's model, in contrast, specifies
content that is broadly related to, among other things, interpersonal and
cognitive styles. Examination of their respective assessment measures and
the kind of data collected (e.g., dilemma solutions versus stem com-
pletions) substantiate further the differential underpinnings of the two
theories.
Construct Overlap
Given that the two constructs are related but not identical, a further ques-
tion involves the nature of that relationship. In 1976, Loevinger described
the ego as subsuming moral development. She considers moral develop-
ment as one of "four facets of a single coherent process" of ego develop-
ment. As she states, "ego development includes some topics previously
discussed under moral development, socialization, character structure, and
even cognitive development. Indeed, the breadth of topics subsumed
under ego development justifies the term, for nothing less than the ego has
so wide a scope" (Loevinger, 1976, p. 4). Kohlberg, in the same year,
stated the issue as a question of which constructs were more "general" and
embodied by the other. He suggested that cognitive development struc-
tures are more general than, and embedded in, both moral and ego struc-
tures; moral structures, in turn, are more general than, and embedded in,
ego structures (1976). Later, Kohlberg (1984) elaborated this view and
emphasized moral judgment as one of the more narrow and tightly refined
areas of personality functioning that is incorporated as one aspect of gener-
al ego functioning. Whereas both investigators have delineated different
facets of that relationship, both lean toward support of an "ego subsumes
moral" position.
The relationship between "general ego" and the moral development
domain, however, is far from settled. The question also involves the fun-
damental independence of components or subdomains of development
within the ego. In Loevinger's view, the ego represents one unified struc-
ture composed of interwoven, inseparable threads, one of which (character
development/impulse control) corresponds to moral development. The ego
is differentiated, with each "thread" performing its role; but the whole of
the ego is structurally unitary, and inseparable for analysis by individual
domain or function (cf. Lambert, 1972). In contrast to this view, Kohlberg
and his colleagues have tended to view the ego as comprised of relatively
circumscribed and self-contained subdomains, each possessed of a distinct
substructure, and capable of empirical separation (Snarey et aI., 1983).
Within this "multiple subdomain" approach, each domain (e.g., cognitive
development, moral jUdgment) is characterized by a relatively distinct sub-
structure that underlies the more holistic superstructure of the unifying
ego. Kohlberg and colleagues also have hypothesized that development in
one sub domain may precede and be necessary but not sufficient for equiva-
lent development in contiguous subdomains.
RELATIONSHIP BETWEEN STAGES

Ego-Moral Correlation
One of several ways to consider the empirical relationship between ego and
moral stages is to ask to what degree the two measures are correlated. To
the extent that there is a moderate to strong correlation between scores on
the MJI and SCT, one may say that they would appear to measure dimen-
sions that coincide across the developmental continuum. Previous empiri-
cal studies show substantial variance in the correlations, ranging from an r
of .24 for a group of college freshmen (Whiteley, 1982) to an r of .80 for a
sample of middle-class adults (Lambert, 1972), with all other studies falling
between these extremes. These results are difficult to interpret, however,
given the small samples and limited age ranges of the subjects included in
the studies.
Ego-Moral Correspondence
Which stages in both systems, however, logically correspond to one
another? Two primary models of correspondence have emerged. The first
TABLE 8.1.
Alternative suggested parallels between stages of moral and ego
development.
Half-stage juxtaposition Whole-stage juxtaposition
Moral Ego Moral Ego
M-l 1-2 M-l 1-2

Obedience Impulsive Obedience Impulsive
M-2 I-Delta M-2 I-Delta
Instrumental Self-protective Instrumental Self-protective
M-2/3 I-Delta/3
M-3 1-3 M-3 1-3
Interpersonal Conformist Interpersonal Conformist
M-3/4 1-3/4
M-4 1-3/4 M-4 1-4
Social accord Conscientious-conformist Social accord Conscientious
M-4/5 1-4/5
M-5 1-4 M-5 1-5
Social contract Conscientious Social contract Autonomous
M-6 1-4/5 M-6 1-6
Principled Individualistic Principled Integrated
Note. Both suggested parallels eliminate reference to 1-1; this ego stage also is not measured
by Loevinger's Sentence Completion Test.
model suggests that one or more stages of one schema be matched with
half-stages of the second schema (see Table 8.1, column 1). Loevinger's
(1976) proposal is the most frequently cited "half-stage" juxtaposition.
Here, whole stages from the two schemata are matched from M-l/1-2
through M-3/1-3; at that point, however, each remaining moral stage is
matched with an ego development half-stage. Beyond the conformity
stage, that is, "each Kohlberg moral stage parallels the Loevinger half-
stage" (Erickson, 1977b, p. 3). The second approach involves a match
between whole stages of the Kohlberg and Loevinger schemata (see
Table 8.1, column 2). The most well-known example of "whole-stage" jux-
taposition is that originally proposed by Kohlberg and his colleagues,
matching M-1 with 1-2, M-2 with 1-0, M-3 with 1-3, M-4 with 1-4, M-5
with 1-5, and M-6 with 1-6. This approach, furthermore, is based on an
analysis of the logical parallels between the stage descriptions of Kohl-
berg's and Loevinger's models (Kohlberg & Kramer, 1969; Snarey et aI.,
1983).
These two alternative models of correspondence are theoretically di-
verse, but there is some practical agreement between them. The suggested
parallels, for instance, are virtually identical for the first three levels. The
models are markedly different only at Stages 5 and 6, which may be of less
practical significance in that these stages are rare empirical phenomena.
DEVELOPMENT ACROSS STAGES

Ego-Moral Sequencing
An issue of pivotal concern involves the question of which parallel stages in
each system precede one another in development. Determining an existing
sequence is complicated by the fact that each correspondence model
suggests a different set of sequencing possibilities. In general, however,
there are four alternatives: (a) Ego stage may precede moral stage in
development-ego primacy; (b) moral stage may precede ego stage-
moral primacy; (c) a more complex, yet patterned, route may exist be-
tween the stages; or (d) passage across stages in the two systems may not
involve any directly consistent path. The form of sequencing is important
because it implies a mechanism of development across stages of the two
systems. In particular, ego and moral primacy have been used as strong
evidence for the existence of a "necessary but not sufficient" causal rela-
tionship between the constructs; this possible decal age relationship is dis-
cussed in a subsequent section.
Investigators may base their evaluation of ego-moral sequencing upon
(a) longitudinal simultaneous measurements of ego and moral stage in
order to view which stages change and in which order, or (b) a cross-
sectional, one-time assessment in which for all or most subjects, ego stage
exceeds the parallel moral stage (implying moral primacy) or moral stage
exceeds the parallel ego stage (implying ego primacy). Most studies to date
have employed the latter approach, often within a narrowly defined age
range, and have achieved conflicting results. Some studies have reported
that a clear majority of the subjects demonstrated a pattern of moral stage
exceeding ego stage, thus supporting moral primacy (e.g., Lambert, 1972;
Snarey, 1986). An equal number of studies, however, report the
opposite-a higher proportion of subjects demonstrated ego primacy
(e.g., Magana, Whiteley, & Nelson, 1980; Sullivan, 1975). Even within the
studies providing the clearest support for a position, there is substantial
room for alternative interpretations because there are always a number of
subjects for whom the opposite is true. The findings are further confused
by an occasional failure to indicate which correspondence model was being
assumed. The general sequencing trends between ego and moral stages are
clearly ambiguous, given the present individual research reports.
Transition Mechanisms
A final issue concerns the existence of causal pathways between the two
constructs and the mechanisms of transition across the two stage systems.
As was noted earlier, Kohlberg and colleagues have hypothesized that
there is a necessary-but-not-sufficient relationship between subdomains
of the ego. In 1972, Kohlberg noted that, in general, attainment of a
Piagetian cognitive stage is a necessary-but-not-sufficient condition for
attainment of a parallel ego stage (Kohlberg & Mayer, 1972). As general

evidence for a necessary-but-not-sufficient relatedness, Kohlberg (1979)
subsequently cited numerous studies indicating that cognitive development
stages (as defined by Piaget) appear to precede social perspective-taking
stages (as defined by Selman) which, in turn, can be shown to precede
moral stages (as defined by Kohlberg). In 1983, Kohlberg and colleagues
reiterated this position and further hypothesized that moral development,
in turn, may be a necessary-but-not-sufficient condition for the develop-
ment of parallel ego stage structures (Snarey et al., 1983). But which stages
are to stand in a "necessary-but-not-sufficient" parallelism? Which features
of one domain of development are necessary for the inception of new struc-
tures in the other domain? Can the "necessary-but-not-sufficient" position
be held simultaneously with a notion of one construct subsuming the
other?
We believe that an additional possibility (and one that does not rule out
the existence of a necessary-but-not-sufficient relatedness) is that second-
ary variables may be involved in moderating the relationship between the
two constructs. In particular, it is possible that individual background vari-
ables (e.g., age, gender, and level of education) may playa role in the
evolving relationship between ego and moral development.
SUMMARY OF PROBLEMS WITH INDIVIDUAL STUDIES

The preceding discussion indicates that connections between the ego and
moral constructs are far from clear. Studies employing the Loevinger SCT
and Kohlberg MJI have failed to yield a consistent picture of the ego-moral
relationship. To a great degree, this has resulted from limitations in sam-
pling. Subject pools have been, of necessity, small (i.e., N < 60), even
though each usually represents a random sample of a defined population.
Three other major limitations also have plagued the efforts to delineate
clear answers; each will now be discussed.
Impact of Variation in Measurement Procedure

Among studies, a number of assessment variations exist: (a) oral versus
written administration of both the MJI and SCT; (b) obsolete or intuitive
scoring methods versus those utilizing the current standardized Kohlberg
and Loevinger manuals; and (c) high interscorer reliability (r = .80+ ) ver-
sus lower scoring reliability (r = .70s). These discrepancies in themselves
are enough to produce interstudy disparities in the ego-moral relationship.
As Hauser notes in his superb review of Loevinger's model and measure,
when measurement procedures vary as greatly as they do, it is very difficult
to interpret "whether the findings reflect method variance or new knowl-
edge about complex lines of development" (1976, pp. 945-946). Thus, we
statistically control for variations in measurement.
In addition to these issues, however, there is a larger measurement dif-

ference involving the scoring algorithms utilized by Loevinger and Kohl-
berg to arrive at the designation of final stage scores. Whereas Kohlberg
employs a modal summation algorithm, Loevinger employs an "ogive"
approach that involves utilizing the subject's distribution of responses with
a different cutting score for each level (Loevinger & Wessler, 1970). Be-
cause each method utilizes different criteria for summarizing the informa-
tion from responses by subjects, each algorithm may selectively ignore or
transform some data in its own way. As early as 1970, investigators had
noted the confusion in comparison caused by, among other factors, "diver-
gent ways in which they [the theorists] calculate overall scores" (Sullivan et
al., 1970, p. 408). While there are several ways to address this measure-
ment problem, our straightforward solution will be to use the same algo-
rithm to calculate the final stage scores for both measures (cf. Lambert,
1972, p. 83; Snarey, 1986).
Influence of Individual Moderating Variables

Most of the individual studies cited in support of either ego primacy or
moral primacy in sequencing have been conducted on homogeneous
populations (e.g., only adolescents or only middle-aged adults, only males
or only females, only high school graduates or only college graduates).
One study does present some evidence that the correlation between ego
and moral constructs may vary across age groups (Sullivan et aI., 1970).
Yet, again, no single investigation has included individuals of various ages
across the life span, monitoring age as a determinant in the ego-moral rela-
tionship. Similarly, gender has been underrecognized as a possible link in
that association but, as Lois Erickson was one of the first to suggest
(1977a), the relationship between ego and moral stage positions may be
different for men and women. Finally, educational achievement, although
occasionally assessed as a background variable, has rarely been entered
into the configuration of variables used to interpret the ego-moral con-
nection. The influence of these variables, in sum, requires serious con-
sideration if the variance in ego-moral stage scores is to be more fully
explained.
Longitudinal Life-Span Changes in the Ego-Moral Relationship

Longitudinal studies provide us with the opportunity to specifically com-
pare the longitudinal change in the relationship between ego and moral
development. The absence of such an analysis leaves the determination of
issues such as stage sequencing up to conjecture following single assess-
ments. Unless one monitors change longitudinally across the life span, it is
impossible to draw reasonably confident conclusions regarding transitions
across stages or the mechanisms by which such transitions occur.
Research Questions
The following three groups of questions provide a foundation for our re-
view and reanalysis of the prior research on the relationship between ego
and moral development.
MEASUREMENT FACTORS
(a) Would the ego-moral relationship change substantially if like algo-
rithms were employed to calculate final Loevinger and Kohlberg scores
(e.g., using the ogive scoring algorithm for both)? (b) Could part of the
relationship be explained by differences in test administration (e.g., oral
versus written forms) or scoring practices (e.g., different versions of scor-
ing manuals and variations in scoring reliability)?
SUBJECT FACTORS
(a) What is the relationship between age, gender, and years of education of
individual subjects and their ego and moral development scores? (b) Could
these personal background variables be linked to variations in the ego-
moral correspondence patterns? (c) Which of these factors accounts for the
most variance in global stage scores and which accounts for the most
variance in the ego-moral relationship?
LONGITUDINAL FACTORS
(a) For individuals, is there a patterned change in the ego-moral rela-
tionship over time, or is the relationship constant? (b) If there are multiple
patterns, are any of the subject or measurement variables noted above
related to longitudinal changes in the patterned relationships between ego
and moral development?
Review Method
To address these research questions, we collected case records from pre-
vious studies and conducted a comprehensive secondary analysis of the
combined data sets (Light & Pillemer, 1984). A primary criterion for selec-
tion of studies was that they could provide "per individual" scores, not
simply group means, on the ego and moral development measures. In addi-
tion, we placed particular emphasis on obtaining subject data from studies
for which it was possible to stipulate three dimensions of analysis: (a)
measurement-related variables (e.g., test administration methods, scoring
methods, and scoring reliability), (b) subject's characteristics (age, gender,
and years of education), and (c) ego and moral data at more than one point
in time (longitudinal stage scores).
SAMPLES
We obtained data from nine prior or ongoing investigations that fulfilled all
or most of these criteria. The exceptions were as follows: demographic
data on the subjects were unavailable from one study (Lambert, 1972) and
the original data summary forms were unavailable from three studies
(Hauser et al., 1984; Lambert, 1972; Shulik, 1979). Table 8.2 summarizes
the individual subject characteristics, test administration procedure, and
other information for all nine studies.
The final composite sample for the nine studies included 567 subjects. Of
these subjects, 130 had two or more longitudinally gathered sets of scores,
giving a total of 787 concurrent measurements of ego and moral develop-
ment that could be included in this investigation. These cases represented a
diversity of individual characteristics. The age of the subjects ranged from
11 to 82 years, with a mean age of 24.6 (SD = 13.4); 65% of the interviews
were from females and 35% were from males. Case records included those
with a high school education or less (41 % ), those possessing some level of
college education (53%), and a small proportion who held a postgraduate
degree (6%).
Creation of a Longitudinal Data Set

Five of the samples included in this review were from longitudinal studies.
To find out whether, and how, the relationship between ego and moral
stages changes over time, we constructed two longitudinal subsamples,
consisting of those subjects in the original subject pool for whom a re-
peated assessment of ego and moral development had been conducted
at least 1 full year after the Time One assessment, and for whom the
necessary background data were available. The first subsample included
104 subjects who had been retested 1 to 4 years after Time One. The
average longitudinal time span was 2.2 years. These subjects included 23
males and 81 females, 68 of whom were adolescents (ages 15 to 18) and
36 of whom were young adults (ages 19 to 30) at Time One. The second
subs ample included 90 subjects who had been retested 5 to 8 years after
Time One. The average longitudinal time span was 5.4 years. These sub-
jects included 23 males and 67 females, 65 of whom were adolescents and
25 of whom were young adults at Time One.
MEASURES
In addition to doing contingency table analysis of raw stage scores, we
calculated two sets of indices to examine the ego-moral relationship: (a)
TABLE 8.2.
Composite sample: contributing studies, sample characteristics, and
methodology.
Procedurea
Studies Design and sample MJI SCT Subjects/testsb
Erickson, 1977a, Longitudinal (4-yr) study of female B A 13/36

1977b, 1977c high school students, ages 15-24
Garrod, 1982 Longitudinal (l-yr) study of male and A A 35nO
female high school students, ages
17-18
Hauser et al., Cross-sectional study of male and A A 124/124
1984; Noam female high school students and
et al., 1984 teenagers at an inpatient mental
health center, ages 12-16
Colby et al., Cross-sectional and longitudinal A A 45/73
1983; (20-yr) study of males; SCT
Kohlberg, administered at 3 test waves,
1958,1982 ages 19-36
Lambert, 1972 Cross-sectional study of students, C A 119/119
faculty, and mental health
professionals, ages 11-60
Mentkowski & Longitudinal (5-yr) study of primarily B A 1111224
Strait, 1983; working class students at a women's
Much & college, ages 17-45
Mentkowski,
1983
Shulik, 1979 Cross-sectional study of elderly men A D 18/18
and women, ages 60-82
Snarey, 1986 Cross-sectional study of senior A A 56/56
kibbutz founders, ages 51-57
Whiteley, 1982 Longitudinal (5-yr) study of university A A 46/67
students, ages 18-22
N 567n87
a The following grading system summarizes the test administration and scoring procedures
used by each study: (A) standard test administration (oral MJI, written SCT) and standard
scoring systems; (B) nonstandard test administration (written MJI, oral SCT) and standard
scoring; (C) standard test administration and nonstandard scoring (obsolete manual or idio-
syncratic method); (D) nonstandard test administration and nonstandard scoring system.
b Subjects/tests refers to the number of subjects and the total number of interviews, including
repeated longitudinal assessments. Unless otherwise specified (e.g., in the longitudinal
analysis), the unit of analysis is each observation or interview time (N = 787; cf. Hays, 1973,
p. 735). Adjoining analyses were also conducted on just Time One interviews (N =567) for all
statistics reported in this review; these results were virtually identical to the findings accrued
from the entire sample. For analyses of the longitudinal data, the unit of analysis was always
the individual subject.
The I-M Continuous Index provides a continuous measure of the separa-

tion in ego and moral scores. It is the result of subtracting a subject's moral
stage score from his or her ego stage score (I minus M), using the following
lO-point scales for both models: 1 = M-1 and 1-2; 2 = M-l/2 and I-2/Delta;
3 = M-2 and I-Delta; 4 = M-2/3 and I-Delta/3; 5 = M-3 and 1-3; 6 = M-3/4
and 1-3/4; 7 = M-4 and 1-4; 8 = M-4/5 and 1-4/5; 9 = M-5 and 1-5; and
10 = M-6 and 1-6. Thus, a single unit in the I-M Continuum indicates a
half-stage difference (e.g., a score of 2.0 means that the ego score exceeds
the moral score by one full stage). (b) The I:M Categorical Ranking catego-
rizes the relationship more succinctly by indicating, for a specific subject at
a particular testing time, whether the ego stage exceeds the moral stage
(I> M category), is equivalent to the moral stage (I = M category), or is
less than the moral stage (M > I category).
Construction of Like Algorithms

One purpose of the present analysis was to permit comparisons of ego and
moral stages when both had been calculated using a common algorithm. To
accomplish this, either individual Sentence Completion Test (SCT) re-
sponses would need to be summed using a modal calculation (allowing an
"ego-modal" by "moral-modal" comparison) or individual Moral Judg-
ment Interview (MJI) responses would need to be summed using an ogive
calculation (permitting an "ego-ogive" by "moral-ogive" comparison).
The standard rules for deriving a modal moral stage score (Colby et aI.,
1978/1987) were first "translated" and applied to the scored sentence stems
for each subject's SCT in order to obtain a modal ego stage score. Briefly,
this involved: (a) calculating the percentage of ego stems scored at each
half-stage, and (b) applying Kohlberg's "criterion cut-offs" for determina-
tion of the modal stage. For instance, in the case where 25% or more of the
scored units (in this case, stems) are assigned to one stage, and only one
TABLE 8.3. Cross-tabulation of subjects' moral and ego stage scores.

Ego stages
Moral stages 2 D D/3 3 3/4 4 4/5 5 6 N
1 1 1
112 3 8 4 1 2 18
2 9 16 8 6 7 1 47
213 10 27 23 15 40 4 1 120
3 3 14 15 25 68 42 6 173
3/4 4 11 30 101 84 26 4 260
4 2 6 33 32 13 1 87
4/5 2 13 16 13 10 54
5 1 2 10 13 1 27
N 25 69 64 85 263 183 69 28 1 787
Note. r= .65.
stage possesses that number of stems, Kohlberg's modal scoring rules

indicate that that stage be designated as the stage score. The rules become
increasingly sophisticated to handle more complex cases (e.g., when
equivalent fractions of stems are scored at several different stages). In each
case, however, it was possible to develop a "transferable" rule from Kohl-
berg's modal summation procedures, which permitted the calculation of a
final modal score for each SCT protocol for which there were available
original sentence stem scores.
By making similar assumptions, it was possible to construct an ogive
algorithm for the summation of final moral stage scores. For instance, the
"cut-offs" employed in ego stage scoring to obtain a final global stage score
were translated into percentages and employed as cut-offs in deriving
moral stages (Loevinger & Wessler, 1970, p. 129).
Results
Table 8.3 presents a cross-tabulation of ego and moral stages for the com-
bined samples, using the standard scoring algorithms. There is a strikingly
high correlation between ego and moral stage scores (r= .65, p< .0001).
Stage 3/4 was the modal stage for both ego development (n = 263) and
moral development (n = 260); 101 subjects scored at Stage 3/4 on both
measures. Using the half-stage juxtaposition (Table 8.3), only 13%
(n = 99) of the subjects were at parallel stages in moral and ego develop-
ment and, using the whole-stage juxtaposition model, 28% (n = 223) of
the subjects were scored at the same parallel stages of moral and ego
development. Considering only those stages represented in our data for
which the half-stage and whole-stage juxtaposition models make different
predictions (Moral Stages 4 and 5), we find that the whole-stage model
correctly predicts the placement of 133 cases in contrast to the half-stage
model, which correctly predicts only 35 cases. The whole-stage model, that
is, more closely reflects the empirical data. With both models, however,
there were a greater number of subjects who scored higher on either ego or
moral development-far more than could be accounted for by measure-
ment error-which suggests that the relationship between the two schema-
ta is not precise enough to indicate that a particular stage in one model is
nearly always found in correspondence with a particular stage in the second
model.
MEASUREMENT-RELATED PATTERNS
Alternate Algorithms
The use of Loevinger's ogive algorithm to derive both moral and ego stage
scores (Ogive Contrast) results in a slight elevation of moral stage scores in
that the mode shifts one half-stage higher, from M-Stage 3/4 (M-Modal) to
M-Stage 4 (M-Ogive). Similarly, using Kohlberg's model algorithm to de-

rive both ego and moral stage scores (Modal Contrast) results in a slight
lowering of ego stage scores in that the mode shifts one-half stage lower,
from 1-3/4 (I-Ogive) to 1-3 (I-Modal). Both shifts result in a slight increase
in the number of cases (8% and 12%) in which a subject's moral stage is
higher than his or her ego stage. Considering the correlation coefficients
between ego and moral development for the 525 cases for which it was
possible to calculate scores by all three algorithms, we find that r = .42 for
the Standard Contrast, .43 for the Ogive Contrast, and .29 for the Modal
Contrast. That is, the Standard and Ogive Contrasts have virtually identical
correlation coefficients, whereas the strength of the association is markedly
lower under the Modal Contrast.
All of the analyses that we will subsequently report were done originally
with ego and moral scores compared under all three contrasts. The results
using alternative algorithms were usually virtually identical to the results
using the conventional standard algorithms, though there were occasions
when a trend evident under the Standard Contrast became stronger or
clearer under the Ogive Constrast. In no instance, however, does the use
of the Ogive or Modal Contrast contradict general results garnered via the
Standard Contrast. Thus, we will report subsequently only comparisons of
ego and moral stage scores calculated by the standard methods (i.e., 1-
Ogive versus M-Modal, Standard Contrast) unless the patterns are
markedly stronger under the use of the Ogive Contrast.
In sum, for all practical purposes, Loevinger (1986) may be correct when
she suggests that there are good psychometric reasons why the two differ-
ent measures should use different algorithms for calculating global stage
scores. She points out that her Sentence Completion Test elicits typical or
modal unprobed responses, while Kohlberg's Moral Judgment Interview
elicits a subject's maximum competency through the process of probing. In
contrast, a "maximizing" algorithm (the ogive) is used to score a SCT and a
"typical" algorithm (the mode) is used to score a MJI. Given that the dif-
ferent algorithms offset the different types of data gathered, the resulting
scores are more comparable than if the different measures had been scored
with the same algorithm. Whether or not one finds Loevinger's argument
compelling, her conclusion is correct in that the present data provide no
compelling reason not to use the standard scoring algorithms.
Test Administration
A major difference between investigators employing the Kohlberg and

Loevinger measures involves the choice of whether to administer the mea-
sures and collect subject responses in oral or written form. To test whether
a difference in administration, oral or written, might affect the outcome of
the ego-moral relationship, we grouped all case records employing stan-
dard test administration procedures (i.e., oral MJI and written SCT) and
TABLE 8.4. Ego-moral categorical index by measurement and subject variables.
Ego-moral categorical indexa
Variables I>M I=M M>I N
1. Test administration
Standard (oral MJI and written SCT) 35.6 28.8 35.6 320
Nonstandard 51.2 28.7 20.1 467
2. Scoring method
Standard (1978 MJI and 1970 SCT manuals) 46.9 27.7 25.4 650
Nonstandard 35.0 33.6 31.4 137
3. Interrater scoring reliability
High (r> .79 on both measures) 46.9 28.1 25.0 580
Lower 39.1 30.4 30.4 207
4. Age in years
12-18 50.5 21.9 27.6 311
19-29 52.9 30.8 16.3 221
30-49 36.1 41.0 22.9 61
50-82 21.3 34.7 44.0 75
5. Gender
Males 32.3 29.2 38.5 226
Females 54.6 26.8 18.6 425
6. Education in years
12 or less 52.5 22.3 25.2 242
13-16 50.3 31.9 17.7 310
17-22 22.9 28.6 48.6 35
Overall 44.9 28.7 26.4 787
aPercentage of records where the ego stage score is higher than, equal to, or lower than the
moral stage score. Percentage figures refer to rows.
all of those that deviated from standard administration procedures (i.e.,

either a written MJI, oral SCT, or both). As Table 8.4 (Item 1) indicates,
there is a significant difference between these groups with regard to the I:M
categories, X2 (2, N = 787) = 27.49, p < .0001. In the group that follows
standard administration practices, the cases tend to distribute more evenly
across I:M categories; in the group that failed to employ the standard test
administration procedures, a clear majority of the cases favor I> M.
To further isolate which practice, the MJI or SCT administration, may
be primarily responsible for these differences, cases were broken down into
an initial contrast, comparing those that had employed the standard oral
MJI to those that had used a nonstandard written MJI. Similarly, a second
contrast consisted of those that had used a standard written SCT versus
those that had used a nonstandard oral SCT. The relationship between ego
and moral stage scores is significantly different for the standard versus non-
standard MJI test administration but not for the standard versus nonstan-
dard SCT test administration, X2 (2, N = 787) = 53.11, p < .0001; X2 (2,
N = 787) = 0.93, p = ns. That is, only the method of MJI administration
significantly affects the observed relationship between ego and moral

constructs-those records employing a nonstandard written form of the
moral interview have a higher probability of arriving at a I> M rela-
tionship (i.e., somewhat lower moral stage scores), apparently because
written responses to Kohlberg's interview inadequately assesses a subject's
level of moral judgment, resulting in a lower moral stage score. In sum,
studies should consistently employ the standard test administration proce-
dures; if this were done, the proportion of M > I cases would be higher.
Scoring Method
Both the ego and the moral measures, over the years, have been scored
using successively refined approaches. The most recent of these (Colby et
aI., 197811987; Loevinger, Wessler, & Redmore, 1970) serve as standard
scoring manuals for the assignment of stage scores to individual subject
responses. To determine whether any differences in the ego-moral rela-
tionship might result from differences in the scoring methods used, cases
were divided between those utilizing the most recent Kohlberg and Loevin-
ger scoring manuals versus those employing an earlier scoring manual or an
intuitive "clinical" approach for one or both measures. As Table 8.4 (Item
2) indicates, there is a small but significant difference between these groups
with regard to the resulting relationship between ego and moral develop-
ment, X2 (2, N = 787) = 6.47, P < .05, V = .06. When the standardized
scoring methods have been used, there is a slight increase in the number of
I> M cases. To further isolate whether variation in scoring the MJI or the
SCT is primarily responsible for this difference, standard versus nonstan-
dard scoring methods were considered separately for the MJI and for the
SCT. In both cases, the above association failed to be replicated. In sum,
only when both moral and ego development are scored by nonstandard
methods is the relationship between ego and moral stage scores significant-
ly changed.
Scoring Reliability
Interrater scoring reliability was investigated as a possible factor affecting
the ego-moral outcomes. We considered two groups: those cases for which
high interrater reliability was achieved (defined as a reliability coefficient of
.80 or higher on both measures), and a group of those remaining cases for
which lower reliability was achieved (.70s). In a chi-square comparison
employing the I:M category variable, differences between these two groups
are not significant (see Table 8.4, Item 3). Similarly, when ego and moral
scoring reliability is considered separately, differences in the relationship
between ego and moral stage scores are not significantly associated with
scoring reliability. In sum, interrater scoring reliabilities in the .70s appear
to be as adequate as reliabilities in the .80s or higher.
INDIVIDUAL-RELATED PATTERNS
Age
Subject age demonstrates a moderate correlation with ego stage (r = .31,
p < .0001), and a high correlation with moral stage (r = .54, p < .(001).
There is a strong and significant age effect with regard to the ego-moral
relationship (Table 8.4, Item 4). When case records are partitioned by age
into adolescence (12 to 18 years), young adulthood (19 to 29 years), middle
adulthood (30 to 49 years), and later adulthood (50 to 82 years), these
groups emerge as significantly different in their I:M categorical ranking, X2
(6, N = 668) = 42.53, p < .0001. The majority of adolescent and young-
adult cases demonstrate an I> M relationship; those in middle adulthood
tend more frequently to have 1= M; in later adulthood, cases concentrate
primarily in the M > I bracket.
The age effect continues to be supported when comparing these groups
on the I-M continuous index [One-Way ANOYA, F(3,664) = 8.48, p <
.0001], and Scheffe's post hoc test indicates that the group means are signi-
ficantly different during adolescence and early adulthood versus middle
adulthood versus later adulthood, p < .05. These differences are even
more apparent in the Ogive and Modal Contrasts [F(3,526) = 36.50, p <
.0001 (Ogive); F(3,525) = 51.67, p < .0001 (Modal)], where post hoc tests
indicate that the means for all four age groups are significantly different,
p < .05 (Scheffe test). In addition, for both contrasts, a linear trend analy-
sis is significant, F(1,522) = 100.67, p < .0001 (Ogive); F(1,522) = 148.30,
p < .0001 (Modal). This suggests that, with other factors held constant,
changes in the relationship between ego and moral stage scores may be
predicted, in part, as a linear function of age (see Figure 8.1, Ogive Con-
trast).
Gender
With gender (femaleness) coded as a variable, a low but significant correla-
tion occurs with ego stages (r = .22, p < .0001), indicating that women are
somewhat more likely to have higher ego scores than men. In contrast,
there is no significant correlation between gender and moral stage scores.
As with age, there is a strong gender difference with regard to the ego-
moral relationship (see Table 8.4, Item 5). This is apparent in a chi-square
analysis of the I:M categories by gender, X2 (2, N = 651) = 38.88,
p < .0001. Illustrated as a difference in means (ANOYA: I-M continuum),
this comparison is also highly significant, F(1,649) = 36.99, p < .0001. In
sum, men in the sample are more likely to display a mean index that is
negative (indicating a higher moral stage), whereas women are more likely
to display a mean index that is positive (indicating a higher ego stage).
These results lend strong support to the idea that there are gender differ-
M > 1 1~ stage
M>I 1 stage -.1.07
M>I ~stage
1= M stage
I>M ~stage
1 >M 1 stage
I> M 1~ stage 'I1//....L..1_ _ _ _...1.1_ _ _ _->1_ _ _ _""'---_
12-18 19-29 30-49 50-82

Age Groups
FIGURE 8.1. Changes in the relationship between ego and moral stage scores as a
linear function of age.
ences in ego development and in the ego-moral relationship (cf. Erickson,

1977a; Magana et ai., 1980), but they do not support the attribution of
gender differences in moral development alone (cf. Gilligan, 1982).
Education
There is a moderately strong correlation between years of education and
ego stage scores (r = .33, p < .0001) and a strong correlation between
education and moral stage scores (r = .50, p < .0001).
When case records are partitioned into educational brackets (12 or less,
13 to 16, and 17 to 22 years), these groups demonstrate significant differ-
ences with regard to the I:M categorical ranking, X2 (4, N = 587) = 24.40,
p < .0001. The majority of cases in the first two groups (high school or less,
and college educated) indicate an I> M relationship; however, those cases
with postgraduate education (17 to 22 years) demonstrate aM> I con-
figuration much more frequently (see Table 8.4, Item 6).
The education effect shows up, as would be expected, as a significant
difference in the I-M continuous index across these three groups,
F(3,664) = 8.84, p < .0001 (One-Way ANOVA). Individuals with more
education exhibit a more negative I-M continuous index than those with
fewer years of formal schooling; a linear trend analysis is also significant at
p < .0001. Hence, as in the case of age, years of education appears directly
and linearly related to the relationship between ego and moral stages.
Measurement and Subject Variables

Multiple regression analyses were used to separate the contribution of
individual subject variables and measurement variables to explaining the
variance in ego development, moral development, and the ego-moral rela-
tionship. Two sets of independent or predictor variables were used: (a)
age, gender, and years of education; and (b) measurement variables, in-
cluding moral and ego test administration, scoring method, and reliability.
Dependent variables included ego stage scores, moral stage scores, and the
relationship between ego and moral constructs as indicated by the I-M con-
tinuous index.
When ego stage is entered as a dependent variable in a stepwise regres-
sion procedure, age, gender, and education together account for 15.8%
of the stage variance; the subsequent inclusion of ego test administration
procedure increases the explained variance by 1% to 16.6% [R = .408,
F(2,567) = 28.26, p < .0001]'
When moral stage is the dependent variable in a similar stepwise regres-
sion procedure, age and education together account for 37.5% of the stage
variance; the contribution of gender is not significant, but the subsequent
inclusion of moral interview administration-procedure increases the ex-
plained variance by 3.1% to 40.6%, and the contribution is significant
[R = .637, F(2,567) = 96.65, P < .0001].
When the I-M continuous index is entered as a dependent variable in a
stepwise regression procedure, age, then gender, then years of education
account for 6.2%, 4.6%, and 1.6% of the variance successively, for a total
of 12.4% [R = .352, F(3,566) = 26.72, P < .0001]. Measurement variables
do not make a significant contribution to the explained variance.
To further clarify the contribution of test administration procedures to
explaining variance in the I-M continuous index scores, we employed a
hierarchical regression procedure, entering first moral and ego test ad-
ministration variables, and, second, the individual variables (gender, age,
and education). The moral administration variable explains 2.6% and the
ego administration variable explains less than .01% of the variance. Gen-
der, age, and education were entered successively after entering the
measurement variables, and account for an additional 3.3%, 6.5%, and
0.1 % of the variance, respectively; all individual p values are significant at
the .001 level [R = .354, F(5,564) = 16.18,p < .0001]. In sum, age and gen-
der are the major variables accounting for variance in the I-M continuum
for all contrasts, even after controlling for measurement variables.
LONGITUDINAL PATIERNS
General Changes
Table 8.5 summarizes the longitudinal change in the I:M categorical rank-
ing. For the 1- to 4-year follow-up, nearly half of the 104 subjects (44%)
TABLE 8.5. Longitudinal change in the ego-moral categorical relationship.
.....
-..I
0
Ego-moral ranking from Time 1 to Time 2

t""'
Constant Changed 0
..
G
TowardM> I Toward I > M =

t""'
G
G
Sample I>M I= M M>I I>M I=M I>M M>I I= M M>I I>'
characteristics to to to to to to to to to Co
=
at Time 1 I>M I= M M>I n% M>I M>I M=I I>M I>M I=M n%
.....
0
n%
::r
1-4-Year longitudinal period =
CIl
I>'
All subjects (N = 104) 42 11 5 56% 7 3 21 30% 3 9 3 14%
..=
G
'<
By age groups
Adolescents (n = 68) 35 7 4 68% 3 1 12 23% 2 4 0 9%
Young adults (n = 36) 7 4 33% 4 2 9 42% 1 5 3 25%
By gender groups
Males (n = 23) 1 3 2 26% 4 3 9 70% 0 1 0 4%
Females (n = 81) 41 8 3 64% 3 0 12 19% 3 8 3 17%
5-8-Year longitudinal period
All subjects (N = 90) 25 4 14 48% 10 8 16 38% 1 7 5 14%

By age groups
Adolescents (n = 65) 24 2 8 52% 7 2 11 31% 1 6 4 17%
Young adults (n = 25) 1 2 6 36% 3 6 5 56% 0 1 8%
By gender groups
Males (n = 23) 2 0 6 35% 4 4 5 56% 0 2 0 9%
Females (n = 67) 23 4 8 52% 6 4 11 31% 1 5 5 16%
demonstrate a change in the relationship between ego and moral develop-

ment (Le., they move from one I:M category to another); 30% changed
toward M > I, compared to only 14% who changed toward I > M. At the 5-
to 8-year follow-up, again nearly half of the 90 subjects (47%) demonstrate
a change in the relationship between ego and moral development; 38%
changed toward M > I, compared to only 14% who changed toward I > M.
At both testing times, in sum, the net direction of movement is from I > M
toward M> I categories. Furthermore, it must be remembered that the
majority (87% and 91 %) of the subjects in our two longitudinal samples
were adolescents or young adults (ages 12 to 30) at Time One; if the age
trends from the older subjects in the larger cross-sectional sample are cor-
rect, this movement toward M> I should continue through middle and late
adulthood. As we will now discuss, a portion of this longitudinal variation
can be explained as a function of the subjects' age and gender.
Age
For the 1- to 4-year follow-up, significantly fewer adolescents in the sample
(32%) change from their original Time One I:M categorical ranking, com-
pared to the larger number of young adults (67%) who move into another
I:M categorical ranking [x2 (1, n = 104) = 11.24, P < .001, phi = .33].
Among those who change in the two age groups, the direction of change
was more often toward M> I among both the adolescents (73%) and the
young adults (63%) [X2 (1, n = 46) = 0.55, P = ns]. For the 5- to 8-year
follow-up, a larger number of adolescents in the sample change (48%), and
their rate of change is now not significantly different from that among the
adults (64%) [X2 (1, n = 90) = 1.92, P = ns]. Among those subjects who
changed, the direction of the change is again more often toward M> I for
both the adolescents (65%) and young adults (88%) [X2 (1, n = 47) = 2.79,
P = ns]. In sum, change in I:M categorical ranking is somewhat more fre-
quent among the older subjects, but the direction of change toward M> I
is consistent across age groups.
Gender
For the 1- to 4-year follow-up, significantly fewer females (34%) change
from their original Time One I:M categorical ranking compared to the
larger number of males (74%) who move into another I:M categorical
ranking [X2 (1, n = 104) = 10.54, P < .005, phi = .32]. For instance, approx-
imately twice as many women as men in this sample retain their initial I:M
ranking. Among those who changed, the direction was more often toward
M> I for both sexes, but a significantly larger number of the males (94%)
compared to females (52%) exhibited this pattern [X2 (1, n = 46) = 8.76,
P < .005, phi = .29]. For instance, at least three times as many men go
through "reversals," all from I> M to M> I. For the 5- to 8-year follow-
up, again a larger number of males changed categories (65%), but their
rate of change is not significantly different from that of females (40%) [X2
(1, n = 90) = 2.09, p = ns]. Among those subjects who changed, the direc-
tion of the change was again more often toward M > I for both males
(87%) and females (66%) [X2 (1, n = 47) = 2.26,p = ns]. In sum, change in
I:M categorical ranking is more frequent among males, but the net direc-
tion of change is toward M> I for both males and females during both
longitudinal periods.
Age, Gender, and Education

Which variables are better predictors? A proportional-reduction-in-error
(PRE) measure is used to summarize the ability of background variables to
correctly predict an individual's change in the ego-moral configuration. For
the 1- to 4-year testing time, the PRE was .096 for the association between
age as the independent variable and I:M ranking as the dependent vari-
able. Knowing if a subject was a member of the adolescent or adult cohort,
that is, improved our ability to predict the relationship between ego and
moral development by almost 10%. Adding in gender further improved
the PRE by .065 to 16%, and adding in education improved the predictive
value by .022, increasing the PRE to 18%. Beyond the I:M changes
explained by age, gender, and education, no additional variation was
accounted for by measurement variables. For the 4- to 8-year longitudinal
testing time, the PRE was .107 or 11% for age; adding in gender further
improved prediction by .031, increasing the PRE to 14%. Beyond the I:M
changes explained by age and gender, no additional variation was
accounted for by education or measurement variables. In sum, knowing
age, gender, and education improved our ability to predict subsequent ego-
moral relationships by 14% to 18% beyond what would be expected by
chance.
Discussion
We begin with a general discussion of the findings as they apply to the issue
noted in the literature review as central in the controversy regarding the
ego-moral relationship. We then discuss the evolving longitudinal rela-
tionship between ego and moral development across the life span, drawing
on Erik Erikson's model as an integrative perspective for understanding
the relationship between ego and moral development.
CENTRAL ISSUES IN THE CONNECTION BETWEEN EGO

AND MORAL CONSTRUCTS
Ego-Moral Correlation
One problematic issue in achieving a consistent reading on the degree of re-
latedness between ego and moral stage schemata has been that no single
study provides a sample encompassing a broad range of age groups. The

present investigation addresses this issue by using a composite subject pool
spanning early adolescence to later adulthood. These data provide evi-
dence that ego development and moral development are empirically re-
lated to one another across the life span. This can be seen informally by
examining the ego-by-moral stage distribution (cf. Table 8.3). Were the
two sets of stages to diverge in terms of phenomena measured, there would
be little observable relationship between the two at some point in the stage
progressions. However, the cases extend in a relatively broad diagonal
from lower ego and moral stage scores to higher ego and moral stage scores
(r = .65, P < .(001).
Ego-Moral Correspondence
What emerges less clearly from the data is evidence for a consistent one-to-
one correspondence between stages in the two systems. For the half-stage
juxtaposition model (cf. Table 8.3, column 1), approximately 13% of the
cases are at directly parallel ego and moral stages. Using the whole-stage
juxtaposition model (cf. Table 8.3, column 2), approximately 28% of the
cases are at identical ego and moral stages. The whole-stage model is clear-
ly more consistent with the empirical data but, since measurement error
cannot account for such a high proportion of cases not fitting either model,
it seems reasonable to conclude that the frequent line-up between certain
ego and moral stages falls short of an empirical stage-per-stage identity.
The distribution of ego-by-moral scores, although it shows clear stage par-
allels between the two constructs, is simply not precise enough to "prove"
a particular stage-by-stage isomorphism.
Ego-Moral Sequencing
As earlier noted, sequencing refers to the issue of which stages in the
Loevinger and Kohlberg schemata precede one another in development.
Ego primacy and moral primacy assume that stages of one construct consis-
tently precede parallel stages of the second construct during development.
Which stages precede which is in part a function of the correspondence
model chosen. Under a whole-stage juxtaposition model, for instance, the
demonstration of moral primacy involves showing that a subject reaches
M-3 prior to 1-3, and M-4 prior to 1-4, since these are "matched" points in
development. The most conclusive proof of such a sequencing, short of
regular and frequent monitoring of moral and ego stage change, is to de-
monstrate that, at any single point in time, all or most subjects have a
moral stage exceeding the parallel ego stage (or vice versa, for ego prima-
cy). This would then suggest that one set of stages is the precursor for the
other.
The present data appear to deny the possibility of a simple universal
sequence in terms of either moral or ego primacy. Comparing ego and
moral stages under the whole-stage juxtaposition model, one notes the fol-
lowing: (a) The number of cases are well divided between those for which
ego stage exceeds moral stage (27%) and those for which moral stage ex-
ceeds ego stage (45%). It cannot be concluded, that is, that all or most of
the subjects are higher on one set of stages than the other. (b) Longitudinal
data from this study indicate that nearly one half of the subjects change
their I:M pattern over a short period (1-4 years). This short-term change is
bidirectional, although the number of subjects moving toward the M> I
pole is greater. Furthermore, within that short span of time, 10% of the
subjects experience a complete reversal (e.g., a shift from M> I to I> M).
Over a longer time span (5-8 years), the data indicate that a larger major-
ity of the individuals who changed have moved from a I > M position to a
M> I position, but 13% still show the opposite pattern. Hence, it is dif-
ficult to contend that one set of stages, at all times and for all persons,
precedes the other.
Transition Mechanisms
As earlier noted, the two primary approaches to explaining the differential
development across ego and moral stage continua include the influence of
moderating variables (age, gender, education) and a possible "necessary-
but-not-sufficient" relationship between the two constructs. Regarding the
moderating variables, it is now evident that age and gender account for
sizable portions of the variance in the ego-moral relationship. These fac-
tors, that is, help to explain the variance as a function of the differential
performance on ego and moral measures (e.g., between men and women),
although they do not establish per se that ego and moral development
are causally related. With regard to the "necessary-but-not-sufficient"
theorem, it is now clear that the data do not show a universal sequence in
ego and moral correspondence across the entire life span. As we will dis-
cuss in the next integrative section, it appears that for some periods in the
life cycle, moral stage progression is the first order of business, and that for
other periods, ego development is preeminent.
THE EVOLVING EGO-MORAL RELATIONSHIP

ACROSS THE LIFE SPAN
While we regard the longitudinal analyses from the present investigation
as exploratory, the findings represent a substantive first attempt to track
change in the ego-moral relationship over time. Combining these results
with the findings from the larger cross-sectional subject pool makes it
possible to present a tentative developmental pathway that a typical
(modal) person will take with regard to the ego-moral relationship.
During adolescence, the typical youth is likely to score at an ego stage
that is higher than her or his moral stage (hence, a high positive I-M
index). This is true for both sexes, but more pronounced for females.
Furthermore, though adolescents may progress to new stages in ego and
moral development, the typical male or female adolescent is unlikely to
change their initial I:M status (i.e., he or she will not experience a categor-
ical switch during this period).
In the early adulthood years (ages 19 to 29), the modal individual is
likely to change I:M categories at least once; a complete "reversal" would
not be uncommon (e.g., a switch from I>M to M>I). While this is true
for both sexes, it is particularly pronounced for males, who are twice as
likely to change I:M status. Collectively, the net direction of all of those
changing I:M categories will be toward the M> I pole. The modal indi-
vidual, however, will still remain at I> M or 1= M; this is especially true
for women. If a young adult has an advanced postgraduate education,
however, there is a clear tendency, regardless of gender, to move toward
the M> I side of the continuum.
In the middle-adulthood years (ages 30 to 49), chances are good that the
modal man or woman will be matched quite well on ego and moral stage
(i.e., the I-M index will be close to zero); there is also a decreasing likeli-
hood that the modal man or woman will change I:M status. The middle-
adult years, that is, are characterized by greater balance and stability in our
sample.
In the mature years of adulthood (age 50+), both the modal man and
woman will most likely come to rest at the M > I pole of the ego-moral
continuum. This is particularly the case for men. For modal individuals of
both sexes, the changes of having an M > I status, and a larger negative
I-M continuous index, also increases with higher levels of educational
attainment. The mature years, in sum, are characterized by new growth in
the moral domain.
The prior models of the ego-moral relationship do not predict these pat-
terns. Erikson's model of the life cycle, however, appears to be consistent
with our findings (Erikson, 1950, 1982). This notion is not fully without
precedent, of course. Loevinger has stated that Erikson's chronicle of
psychosocial development is the only psychoanalytic model of ego develop-
ment that is compatible with her own (1976, p. 4), and Kohlberg has also
suggested parallels between Erikson's model and his own (Kohlberg, 1984;
Kohlberg & Gilligan, 1971; Kohlberg & Kramer, 1969). We will now
summarize the parallels between Erikson's work and the findings of our
review.
First, the adolescent's preoccupation with identity achievement is consis-
tent with achieving higher levels of ego development than moral develop-
ment. Ego identity formation, while a constant activity throughout life, is
an overwhelming preoccupation during adolescence.
Making greater strides in moral than in ego development during early
adulthood parallels the characteristics that Erikson assigns to this period-
the virtue that emerges after the achievement of identity is fidelity and the
task of young adulthood is the achievement of intimacy, which gives rise to

the capacity for mature love. Adulthood fidelity and love, that is, both
represent successive capacities for social-moral commitments that are free-
ly made but ethically binding.
The subsequent stability in ego and moral development and the ego-
moral relationship during middle adulthood also parallels the next phase in
Erikson's model-generativity. Here the individual's focus shifts more
strikingly from personal development to the development, care, and con-
tinuity of the next generation. It is an inherently conservative phase of life
in the sense that the task is to pass on to other younger adults those values,
ideas, and things that one has come to value.
Finally, the general higher level of moral development in the mature
years may also reflect the moral virtue that arises from the achievement of
generativity-care. The characteristics of the higher stages in moral de-
velopment also appear to us to be strikingly similar to the characteristics
that Erikson assigns to mature aging-integrity and wisdom. In sum, Erik-
son's model functions well as an integrative framework for accounting for
the empirical relationship between ego and moral development across the
life span.
REFERENCES
Colby, A., Kohlberg, L., Candee, D., Gibbs, J., Speicher-Dubin, B., & Hewer, A.
(1987). The measurement of moral judgment: Standard issue scoring manual.
New York: Cambridge University Press. (Prepublication manuscript originally
distributed by the Harvard Center for Moral Education in 1978.)
Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study
of moral judgment. Monographs of the Society for Research in Child Develop-
ment, 48(1-2), 1-124.
Erickson, V.L. (1977a). Beyond Cinderella: Ego maturity and attitudes towards
the rights and roles of women. Counseling Psychologist, 7, 83-88.
Erickson, V.L. (1977b). The domains of ego and moral development. Moral
Education Forum, 4(2), 1-4.
Erickson, V.L. (1977c). Deliberate psychological education for women: curriculum
follow-up study. Counseling Psychologist, 6,25-29.
Erikson, E.H. (1950). Childhood and society. New York: Norton.
Erikson, E.H. (1982). The life cycle completed. New York: Norton.
Garrod, A. (1982). A developmental approach to the teaching of literature: A con-
text for moral and ego growth in adolescence. Doctoral dissertation, Harvard
University, Cambridge, MA.
Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University
Press.
Haan, N., Stroud, J., & Holstein, C. (1973). Moral and ego stages in relationship
to ego processes: A study of "hippies." Journal of Personality, 41, 596-612.
Hauser, S. (1976). Loevinger's model and measure of ego development: A critical
review. Psychological Bulletin, 33, 928-955.
Hauser, S., Powers, S., Noam, G., Jacobson, A., Weiss, B., & Follansbee, D.
(1984). Familial contexts of adolescent development. Child Development, 55(1),

195-213.
Hays, W. (1973). Statistics for the social sciences. New York: Holt, Rinehart &
Winston.
Kohlberg, L. (1958). The development of modes of thinking and choice in the years
10 to 16. Doctoral dissertation, University of Chicago, Chicago, IL.
Kohlberg, L. (1964). Development of moral character and moral ideology. In M.
Hoffman & L. Hoffman (Eds.), Review of child development research (pp. 383-
481). New York: Russell Sage.
Kohlberg, L. (1966). A cognitive-developmental analysis of children's sex-role
concepts and attitudes. In E. Maccoby (Ed.), The development of sex differences
(pp. 82-173). Stanford, CA: Stanford University Press.
Kohlberg, L. (1972). The concepts of developmental psychology as the central
guide to education: Examples from cognitive, moral, and psychological educa-
tion. In M. Reynolds (Ed.), Psychology and the process of schooling in the next
decade (pp. 1-55). Minneapolis: University of Minnesota Press.
Kohlberg, L. (1976). Moral stages and moralization. In T. Lickona (Ed.), Moral
development and behavior. New York: Holt, Rinehart & Winston.
Kohlberg, L. (1979). The meaning and measurement of moral development. Heinz
Werner Memorial Lecture. Worcester, MA: Clark University Press.
Kohlberg, L. (1981). Essays on moral development: l. The philosophy of moral
development. San Francisco: Harper & Row.
Kohlberg, L. (1982). The empirical relationship between moral and ego develop-
ment: A preliminary report. Presentation made at Harvard University, Graduate
School of Education, Cambridge, MA.
Kohlberg, L. (1984). Essays on moral development: II. The psychology of moral
development. San Francisco: Harper & Row.
Kohlberg, L. (1986). An update of the moral stage theory. In S. Modgill & C.
Mogdill (Eds.), Kohlberg: Consensus and controversy (pp. 485-545). Brighton,
England: Falmer.
Kohlberg, L., Colby, A., Lieberman, M., & Speicher-Dubin, B. (1973). Standard
issues scoring manual. Cambridge, MA: Center For Moral Education, Harvard
University.
Kohlberg, L., & Gilligan, C. (1971). Adolescent as philosopher: The discovery of
the self in a post-conventional world. Daedalus, 100(4), 1051-1086.
Kohlberg, L., & Kramer, R. (1969). Continuities and discontinuities in childhood
and adult moral development. Human Development, 12, 93-120.
Kohlberg, L., & Mayer, R. (1972). Development as the aim of education. Harvard
Educational Review, 42(2), 449-496.
Lambert, H. (1972). A comparison of Jane Loevinger's theory of ego development
and Lawrence Kohlberg's theory of moral development. Doctoral dissertation,
University of Chicago, Chicago, IL.
Light, R., & Pillemer, D. (1984). Summing up: The science of reviewing research.
Loevinger, J. (1978). Scientific ways in the study of ego development. Worcester,
MA: Clark University Press.
Loevinger, J. (1979). Construct validity of the sentence completion test of ego
development. Applied Psychological Measurement, 3(3),281-311.
Loevinger, J. (1986). On Kohlberg's contribution to ego development. In S. Mog-

dill & c. Mogdill (Eds.), Kohlberg: Consensus and controversy (pp. 183-193.
Brighton, England: Falmer.
Loevinger, J., & Wessler, R. (1970). Measuring ego development: I. Construction
and use of a Sentence Completion Test. San Francisco: Jossey-Bass.
Loevinger, J., Wessler, R., & Redmore, C. (1970). Measuring ego development: II.
Scoring manual. San Francisco: Jossey-Bass.
Magana, H., Whiteley, J., & Nelson, K. (1980). Sequencing of experiences in
psychological interventions: Relationships among locus of control, moral reason-
ing, and ego development. In V. L. Erickson & J. Whiteley (Eds), Developmen-
tal counseling and teaching (pp. 298-328). Monterey, CA: Brooks/Cole.
Mentkowski, M., & Strait, M. (1983). A longitudinal study of student change in
cognitive development and generic abilities in an outcome-centered liberal arts
curriculum: Final report to the National Institute of Education. Milwaukee, WI:
Alvemo College.
Much, N., & Mentkowski, M. (1«:983). Student perspectives on liberal learning at
Alverno College: Justifying learning as relevant to performance in personal and
professional roles: Final report to the National Institute of Education. Milwaukee,
WI: Alvemo College.
Noam, G., Hauser, S., Santostefano, S., Garrison, W., Jacobson, A., Powers, S.,
& Mean, M. (1984). Ego development and psychopathology: A study of hospi-
talized adolescents. Child Development, 55(1), 184-194.
Shulik, R. (1979). Faith development, moral development, and old age: An assess-
ment of Fowler's faith development paradigm. Doctoral dissertation, University
of Chicgo, Chicago, IL.
Snarey, J. (1986). The relationship of social-moral development with cognitive and
ego development: A cross-cultural study. Behavior Science Research, 21, 132-
146.
Snarey, J., Kohlberg, L., & Noam, G. (1983). Ego development in perspective:
Structural stage, functional phase, and cultural age-period models. Developmen-
tal Review, 3(3), 303-338.
Sullivan, E., McCollough, G., & Stager, M. (1970). A developmental study of the
relationship between conceptual, ego, and moral development. Child Develop-
ment, 41,399-411.
Sullivan, J. (1975). A curriculum for stimulating moral reasoning and ego develop-
ment in adolescents. Doctoral dissertation, Boston University, Boston, MA.
Whiteley, J. (1982). Character development in college: The freshman year. New
York: Character Research Press.
9
Ego and Ideology: A Critical Review
of Loevinger's Theory
JOHN M. BROUGHTON AND MARTA K. ZAHAYKEVICH
The burning ambition to be considered "scientific" or, rather, the dread of being
judged "unscientific" may confine a psychologist's field of vision to those phe-
nomena which are wholly objective, relatively simple, and mechanically measur-
able, and thus black out most of the activities of human beings.
-Henry Murray, 1951, p. 443
Ego psychology represents a dominant paradigm in the study of personal-
ity. However, it is a confusing tradition, embracing a number of different
schools of thought. One could point to at least two distinct European
varieties, linked with the names of Jung and Anna Freud, and as many as
three North American branches, associated with the following figures:
Hartmann, Rapaport, and Nunberg; Allport, Murray, Ausubel, and
White; and Sullivan and Erikson.
The American schools, in contrast to the European, are positivistic in
their orientation to psychodynamic studies as part of natural rather then
interpretive science. They fall squarely in the mainstream functionalist
tradition of personality psychology, as defined in the 1950s by research-
oriented theorists such as McClelland (1951). The functionalist branch of
positivism is built on the root metaphor of the adaptive organization of
purposive action (Pepper, 1966) and is spelled out in terms of organismic
constructs of structure and function, and differentiation and integration.!
Within the social sciences, this tradition is the psychological equivalent of
the sociology of Talcott Parsons and Robert Merton (see Martindale,
1960).2
1 For an overview of the neo-Darwinist paradigm of functionalism in the twentieth

century, see Hobhouse (1913), Gutman (1964), and Beilin (1987). For a summary
of the historical emergence of ego psychology within American personality theory,
see Green and Rieber (1980).
2This parallel is reflected in the landmark collaboration, Toward a Theory of Social
Action (Parsons & Shils, 1951) and in the recent collection edited by Smelser and
Erikson (1980).
180 John M. Broughton and Marta K. Zahaykevich
Ego psychology is concerned to offset the morbid clinical preoccupation

with pathology. It attempts to do so by stressing the developmental poten-
tial residing in the ego, and the red-blooded capacity of the individual
actively to realize its own competence and autonomy through the mastery
of reality via the initiative of its relatively conflict-free sphere of intelligent
mental organization.
As will become apparent subsequently, the exaggeration of the dimen-
sion of normal, individual growth in human thought, with its accompanying
claim to make psychoanalysis into a natural science, is not only dynamical-
ly conflicted but also represents an ideological distortion and a virtual
dismissal of Freud's psychological and social insights. Nevertheless, within
the established profession of American psychology, which is resoundingly
functionalist in its commitments, and in those sectors of the European
community that borrow from the American paradigm, the synthesis of the
dynamic psychology of personality with the emerging cognitive psychol-
ogy of development is increasingly taken as a cause for celebration (for
example, Dobert, Habermas, & Nunner-Winkler, 1987; Fast, 1985; Kohl-
berg, 1973). One must at least respect the prudent consistency of such
a reaction: functionalists rejoice at the adaptiveness of an intellectual
synthesis that has a high survival potential in a technological era when the
cognitive is in its ascendancy.
If one were to settle upon a single theorist to represent the new synthesis
of dynamic and cognitive psychologies, a logical choice would be Jane
Loevinger. Her popularity and influence are without question (for exam-
ple, Hauser, 1976; McAdams, 1985; Swensen, 1980; Wright, 1975). Within
the functionalist paradigm, and allied chiefly with the HartmanlNunberg
school, Loevinger has accomplished a differentiated synthesis of clinical
and academic approaches. In the ego psychological tradition, she has em-
ployed her statistical expertise to ground her thinking empirically in a
dense web of psychometric maneuvers. Last but not least, in a tradition
that has always had a rather masculine cast to it, it is refreshing to con-
sider a theory constructed by women, about women. 3 The outcome is an
impressive edifice that has already left its mark on the skyline of modern
psychology.
There is a rich and eclectic complexity to the integrative framework that
Loevinger has provided. However, psychology has not seen fit to show this
framework the respect of a critical review. In contrast, the work of Freud
and his colleagues has been dignified by being subjected to considerable
scholarly scrutiny. Since the work of ego psychologists is discontinuous
3The original data base was obtained from nearly 1,500 women, age 11 and up
(Loevinger & Wessler, 1970). More recently, forms for men have been constructed
on the basis of data gathered from a male population (Redmore, Loevinger, &
Tamashiro, 1978; Redmore, Wright, & Rashbaum, 1974).
9. Ego and Ideology 181
with Freud's reuvre, it would seem fitting that the products of the ego
psychological tradition be judged on their own grounds. If ego psycholo-
gists are not to seem content to ride to fame and fortune on Freud's coat
tails, then their theoretical constructions must be evaluated on their own
merits.
There are several steps entailed in the task of illuminating the com-
plexity of Loevinger's achievement. Since the details of her framework and
of her sentence completion task are well known, and have been laid out by
herself (Loevinger, 1966, 1976a; Loevinger & Wessler, 1970/1978) and
others (Broughton & Zahaykevich, 1977, 1980; Hauser, 1976), the analysis
of her work can move ahead to a critical reinterpretation. The following
review starts out with an identification of her basic concepts, setting them
in the context of relevant intellectual-historical traditions. Next, her
methodology and research results are examined. Finally, general meta-
theoretical conclusions are drawn regarding her tacit assumptions about
mind, life, and society.
Our use of Loevinger's theory on this occasion is metonymic. It is
intended as a case study approach to the critical assessment of certain
trends endemic to ego psychology in general. In so doing, the hope is ex-
pressed that it may be possible to move beyond the bounds of this one
specific theory to illuminate the notion of "ego development" in general,
and perhaps even the whole task of constructing explanations of the way in
which personality develops.
Ego Development
THE CONCEPT
What is ego development? As Looft (1973) and Holt (1974) have pointed
out, Loevinger does not give a clear answer. She does not want to: "The
subject of ego development cannot be encompassed by a formal definition"
(Loevinger, 1976a, p. 54; cf. Loevinger, 1966, p. 205). In this, she repli-
cates the retreat from defining that is endemic to ego psychology (Leites,
1971, pp. 199, 249, 257). Her metatheoretical predilections lead her to
prefer an ostensive definition of ego development via an enumeration of
its functions. She argues that these functions are difficult to enumerate
(cf. Leites, 1971, pp. 257-258) and are best understood via the operations
involved in their scientific objectification.
Drawing upon an antithesis traditional to positivism, she stresses the
empirical nature of ego development as "psychological" not "ontological"
(Loevinger, 1976a, p. 63). She stands in the honorable American tradition
of William James-opposed to the metaphysical disposition of psycholo-
gy's prescientific forebears and proud of her refusal to reify the self. Never-
theless, Loevinger is a realist-not a nominalist, or even a probabilist-in
.....
00
N
TABLE 9.1. Some milestones of ego development.

....
0
Impulse control, Sample responses ::r
character Interpersonal Conscious Cognitive to "when people =
~
Stage Code development style cognitive style style are helpless. . .
e;,
...0
Presocial Autistic c:
Symbiotic 1-1 Symbiotic Self vs. nonself ~
...
Impulsive 1-2 Impulsive, fear of Receiving, depen- Bodily feelings, espe- Stereotyping, concep- -they feel bad." 0
retaliation dent, exploitative cially sexual and tual confusion =
I»
aggressive Co
=
Self-protective I'!,. Fear of being caught, Wary, manipulative, Self-protection, trou- -they are looking for ~
externalizing exploitative ble, wishes, things, sympathy." ~
I»
blame, advantage, control
opportunistic ~
Conformist 1-3 Conformity to exter- Belonging, superficial Appearance, social Conceptual simplicity, - I try to help them." N
I»
nal rules, shame, niceness acceptability, banal stereotypes, cliches ::r
guilt for breaking feelings, behavior ~
~
rules
Conscientious 1-3/4 Differentiation of Aware of self in rela- Adjustment, prob- Multiplicity - I want to help
~.
::r
conformist norms, goals tion to group, lems, reasons, them."
helping opportunities
(vague)
Conscientious 1-4 Self-evaluated Intensive, respon- Differentiated feel- Conceptual com- - I feel a need to
standards, self- sible, mutual, ings, motives plexity, idea of help."
criticism, guilt for concern for for behavior, patterning
consequences, long- communication self-respect,
term goals and achievements,
ideals traits, expression
Individualistic 1-4/5 Add: Respect for Add: Dependence as Add: Development, Add: Distinction of -they should make
individuality an emotional social problems, process and out- the most of their
problem differentiation of in- come own abilities."
ner life from outer
Autonomous 1-5 Add: Coping with Add: Respect for Vividly conveyed feel- Increased conceptual -they need help to-
conflicting inner autonomy, inter- ings, integration of complexity, com- ward self-help."
needs, toleration dependence physiological and plex patterns, tol-
psychological eration for ambigui-
causation of ty, broad scope, ob-
behavior, role jectivity
conception, self-
fulfillment, self in
social context
Integrated 1-6 Add: Reconciling Add: Cherishing of Add: Identity -it may be because
inner conflicts, individuality of a lack of self-
renunciation of confidence or
unattainable the will to help
themselves or
knowledge. "
Note. "Add" means in addition to the description applying to the previous level. (Adapted from Loevinger, 1976a, p. 24, with permission of the publishers.)
:0
trl
~
§
0..
......
0..
(1)
o
0-
~
>-'
00
~
regard to her developmental levels, asserting that the stage-like progress

actually exists in the minds of evolving individuals.
It is difficult to convey the global transformation of the person that is
supposed to occur in the full course of ego development, since Loevinger
steadfastly refuses to define a clear end point or telos toward which pro-
gress is directed, and her account of any given level tends to spread out
across four related domains rather than integrating these into a single de-
scription (Table 9.1). Although she eschews the idea of polar variables in
favor of what she calls "milestones," she is not above summarizing ego
development in simplified terms. In her earlier work, she was inclined to
admit her indebtedness to the theory of the "authoritarian personality"
(Adorno, Frenkel-Brunswik, Levinson & Sanford, 1950): "The progress is
from resistance to authority, to identification with authority, to emancipa-
tion from arbitrary authority" (Loevinger, 1962, p. 116). In these forma-
tive years, she was clear about her commitment to the Freudian notion
that, paradoxically, the sole path to freedom must be pursued via an iden-
tification with whatever authority prevails. 4 However, she was quick to
emphasize other dimensions in the developmental trajectory:
(F)rom uncontrolled impulsivity to rigid stereotyped impulse control, to flexible
controls; from no conceptualization of self as a psychological person, to a con-
ventional and usually socially approved stereotyped self-conception, to a more
realistic and differentiated self-concept. (Loevinger, 1962, p. 116)
It is approximately true to summarize her stage-by-stage account by

saying that, with development, egos become more active, self-aware, toler-
ant, objective, flexible, and autonomous. Otherwise said, more mature
subjects tend to give less impulsive, hedonistic, authoritarian, and self-
centered answers. But it should be remembered that Loevinger's distinct
contribution lies in her avoidance of a trait psychology, and her submitting
in its place an account of the self in terms of its passage through qualitative-
ly different levels of organization, each of which gives a new function to old
content (Broughton & Zahaykevich, 1977; Candee, 1974).
THEORY AND MEASURE

It cannot be denied that Loevinger has succeeded in the careful construc-
tion and standardization of her instrument, meeting stringent criteria for
reliability and internal consistency, thereby provoking quantities of re-
search across a broad front (Hauser, 1976; Swensen, 1980). It is this char-
acteristic that attracts psychological theorists like McAdams (1985) and
social theorists like Habermas (1975a: Dobert et aI., 1987) to Loevinger's
4This assumption of Freud's has been reexamined and taken to task by feminist
critics of patriarchal psychology, such as Susan Buck-Morss (1987) and Jessica
Benjamin (1987).
work. Nevertheless, one cannot help wondering, along with Loevinger

(1979) herself, at the rather routine employment of the instrument without
concern for its validity.
The first sign of this disregard for validity, particularly construct validity,
is the way in which Loevinger has violated the normal continuity between
instrument and analytic scheme. The items in the instrument are of a stan-
dard projective type and are similar to those used by earlier investigators
(for example, Borgatta, 1961; Cameron, 1938; Hess & Handel, 1959;
Lorge & Thorndike, 1941; Payne, 1928; Sanford, 1943; Symonds, 1947;
Tendler, 1930). However, Loevinger not only divorces sentence comple-
tion testing from its traditional clinical purpose, but also denies the connec-
tion to primary process. At first glance, her emphasis on the organizational
complexity of subjects' responses at the expense of their content may
appear to follow the tradition of Rapaport and coworkers (1968), who
stressed formal structuralized ego characteristics at the expense of dynamic
meanings. However, Rapaport and coworkers emphasized the importance
of separately assessing types of defensive structure, adaptational style, and
intellectual functioning in order that they might then be woven together
into a rich diagnostic picture of the personality. In contrast, Loevinger is
inclined to collapse these broad areas of personality structure into a single
focus on organizational complexity.
In so doing, Loevinger might be congratulated for the creativity of her
imagination. However, she risks two kinds of impairment of validity. On
the one hand, the isolation of measurement from its original theoretical
ground, without providing a substitute for the latter, compromises con-
struct validity. On the other, taking subjects' responses so literally renders
her instrument vulnerable to "desirability" response bias-the systematic
misrecognition of her subjects' creativity:
In the sentence completion method ... the purpose of the test is less disguised than
in other projective techniques, and the sophisticated subject may tell the examiner
only what he wants him to know. (Rotter, 1951, pp. 295-296)5
Meaning, in general, suffers at Loevinger's hands. Each level in her
scheme binds together in a complex way a whole variety of responses, each
of which, alone, is ambiguous and indicates a particular level of develop-
ment in only a probabilistic manner (Loevinger & Blasi, 1976).6 It is tempt-
ing to place Loevinger in the Piagetian tradition-indeed, she appears to
have succumbed to this very temptation herself.? However, locating her
thus would miss the fact that, due to the particular "scientific" methodol-
5It is not the intention here to impugn the sentence completion method. Far from
it, I have employed the method myself in recent research (Broughton, 1988).
60n Loevinger's probabilism, drawn from Egon Brunswik (e.g., 1947), see Loevin-
ger (1984) and Broughton and Zahaykevich (1980).
71n a puzzling table, Loevinger (1976a, p. 414) names only two progenitors in the
genealogy of her own position: Sullivan and Piaget.
ogy dictated by her epistemology, responses to the sentence stems cannot

be followed up by an interviewer in order to elucidate their meaning. In-
stead, she uses psychometric methods like those developed for "objective
tests" (ct. Green, Ford, & Flamer, 1971).8 Empirical clustering, rather
than internal logical coherence, tends to predominate as the criterion in
the construction of levels in the scoring manual. Her business is primarily
measure construction not theory construction. 9
The use of empirical co-occurrence might appear to weaken the con-
struct validity of the test. In addition, it would seem to undermine a belief
that there is any logical necessity to the sequence of the levels (Kohlberg,
1980; Snarey, Kohlberg & Noam, 1983), a point magnified in significance
by Hauser's (1976) observation that there has been little empirical testing
of Loevinger's sequence hypothesis. This lacuna would seem a natural con-
sequence of the failure clearly to distinguish the theory of ego development
from the measurement instrument, and to validate the theory independent-
ly (Broughton & Zahaykevich, 1977; ct. Broughton, 1978).
Ideal Types and Kinds of Consistency

Holt (1974) has argued that Loevinger's levels constitute functional "ideal
types" rather than structural stages, a position that Loevinger (1983) her-
self has come to adopt. Nevertheless, she appears not to have taken the
distinction between types and stages, explicated by Kohlberg (1985, 1973;
Kohlberg & Armon, 1984), as a matter for serious deliberation. The lack
of a clear distinction between content and form in her levels is correlated
with the lack of a formal core or "deep structure" to the levels, the lack of
conceptual consistency within levels, and the absence of any logical nature
to the sequence of levels. Despite her subscription to the general func-
tionalist conception of personality as something that ought to tend toward
an equilibrated structure, the lack of form/content differentiation leads to
her failure to provide an explanation of the process by which movement
from one level to another occurs (such as the Piagetian "equilibration"
mechanism in Piaget's stage theory). Loevinger (1979) herself admits that
"the construct of ego development lacks definite evidence for sequential-
ity" (p. 308). All of these shortcomings militate against the employment of
the term stage regarding ego development theory, although Loevinger her-
self lapses into such usage (as in Table 9.1).
It is certainly to Loevinger's credit that she and her colleagues were able
8The dangers of decontextualization in objective tests of personality are noted by

Zavalloni (1986) and spelled out in some detail by Gergen, Hepburn, and Comer
(1986).
9The details of Loevinger's method for the construction of levels in ego develop·
ment are provided in Broughton and Zahaykevich (1980).
to shake various associative networks of relations out of the mass of empir-

ical material gathered. Nevertheless, it seems that the partial fusing of
form and content in the ideal types makes it difficult to assess the extent of
class and cultural bias in the instrument. Sentence completions featuring
swear words, for example, tend to be rated low on the scale, as though such
words were necessarily univocal in meaning. Yet Labov (1974) and Reis-
man (1974), among others, have shown that the apparently crude language
of lower-class and minority individuals may conceal as much philosophical
sophistication as the discourse of the most genteel of English-speaking peo-
ple. A similar criticism applies to the developmental theory of Gilligan
(1982; cf. Belenky, Clinchy, Goldberger, & Tarule, 1986), which tends to
lead to the rating of the voice of rural and working-class women as relative-
ly immature (Broughton, 1983a).
Under an ideal-typical schema, the four "categories" making up ego
development would not be expected to be real epistemic categories in the
Kantian sense. They would be more descriptive and would perhaps overlap
with each other. This is precisely Habermas's (1975a) conclusion with
regard to Loevinger's theory. Loevinger herself argues that such intercon-
nectedness follows from the definition of the ego as a synthetic principle of
consistency. For much the same reasons, the four major categories (Table
9.1) subsume conceptually heterogeneous content, crossing not only the
boundaries between thought, affect, motivation, action, and communica-
tion, but also those between subjective and objective, normal and de-
fensive, conscious and unconscious, and nonreflective, reflective, and
theoretical levels of consciousness. To what extent can this be lauded as
realistic overlap, and to what extent does it represent a lack of conceptual
precision?
Loevinger presses her argument by saying that such merging reflects
accurately the continuity in our everyday experience of our own personal-
ity. Admittedly, our common sense may lead us to believe that these
various domains and registers of consciousness cannot be isolated from
each other in our concrete phenomenal experience. Nevertheless, this does
not imply that the several concepts underlying such distinctions have no
specific meaning, or that they cannot be segregated at a conceptual level
and defined abstractly in relation to each other .10 The practical conse-
quence of Loevinger's stance, for the lay reader as well as the researcher,
is that we cannot differentiate "consistency" that is due to muddling cate-
gories from that due to family resemblance, to necessary relations of
logical inclusion or logical identity, or to empirical relations indicated by
patterns of research findings.
lOIn fact, the empirical evidence in the cognitive-developmental literature (e.g.,

Turiel, 1983) points rather clearly to the distinctness of different structural-
functional domains in mental organization.
The Ego as "Synthetic Function"

So far we have complied with Loevinger's ostensive definition of ego
development, attempting to understand it in terms of the way in which it
has been scientifically objectified. However, we should not be misled by
Loevinger's view of her own work. To quote Marx (1859/1976), "One does
not judge an individual by what he thinks about himself" (p. 21). Positiv-
ists, not least those within functionalism, like to disown their theoretical
assumptions. These assumptions are nonetheless real and can be discerned
without great difficulty.
In Loevinger's case, the core of her conceptualization lies in a notion
borrowed from Nunberg: the "synthetic function." Nunberg (1955) had
described the ego as exhibiting a "tendency to join other objects, with the
aim of creating a new unit" (p. 149), a quality parallel to the binding force
of Eros in the workings of the id (although free of sexual meaning). The
concept of synthetic function has been revised by Loevinger so that it is the
parts of the self, not "other objects" that are joined. The ego functions to
bind together not only the categories of development, but also the different
levels and types of mental functioning.
In shifting the emphasis to the inwardly turned concern of the ego,
Loevinger shows that "taste for the internal" that characterizes modern
ego psychology (Leites, 1971, p. 77). By her account, all meaning is reduc-
ible to domination of the object world through integrative psychological
organization (Swensen, 1980, p. 331). As Loevinger (1976a, p. 5) points
out, distinguishing her view from Nunberg's, the ego is synthesis, rather
than being something that has this quality (cf. Holt, 1974, p. 314). The
"essence of the ego [is] ... striving to master, to integrate, to make sense
of experience" (Loevinger, 1969, p. 85).11 To borrow a phrase, it is "the
executive of the personality" and so is prior to and underlies qualities such
as "identity" (McAdams, 1985, p. 120).
As in all positivisms (Habermas, 1971), reflection on experience is eli-
minated from this conceptualization of the self. As in all functionalisms
(Giddens, 1976; Taylor, 1971), interpretation is reduced to mastery
through the imposition of hierarchical structure. Loevinger's work is thus
allied with that of Peterfreund (1971) and others who promote the merging
of psychoanalysis with systems theory, according to which distinctions like
reflective/unreflective, signifier/signified, or conscious/unconscious are all
merged. 12
11 Over seventy years ago, Baldwin (1915) characterized such positions within ideal-
ism as "rational mysticism" since, according to this view, only the self is holistic.
12The problematic features of systems theory are discussed by Gouldner (1970) and
Habermas (1975b), among others, and with regard to developmental psychology by
Broughton (1981). The disappearance, in Loevinger's account, of the distinction
between ego and superego has been remarked on by Langer (1969).
In a Newtonian world view such as that appropriated by the theory of the

synthetic function, however, the laws governing the disposition of forces
must be obeyed: 'For every action, an equal and opposite reaction.' The
centripetal implicates the centrifugal. What cybernetic approaches ignore
is the fact that an ego that is a "function of all functions,"13 in its eagerness
to unify, must reject those aspects of experience that its organizational
imperative cannot encompass. Despite her claim to identify the functions
of the ego to be included in ego development by virtue of empirical proce-
dures, Loevinger actually invokes an a priori organismic criterion that only
those functions that develop as a unity should be considered.
Unification proclaims itself as generous and democratic, but actually
polarizes self and other in an abrupt and exclusive way. "Organization"
sounds natural and neutral. In practice, it has to transform its materials to
make them organizable and, if they resist, it turns to coersive incorporation
or exclusion (Adorno, 1968; cf. Greenwald, 1984). Eventually, we shall
need to consider whether or not this intolerant synthetic force deserves to
be dignified with the appellation of "ego." As in factor analysis, naming
the foci of empirical clustering is the most precarious step in the process.
Categories of Ego Development

Because of Loevinger's understandable reluctance to discuss metaphysics,
a better picture of ego development is gained by looking beyond its defini-
tion, and beyond the four major category labels, to the content of the
various stages (Table 9.1). Under the moral qualities of impulse control
and character development, we find two kinds of growth. At the lower end
of the scale, self-preservative tendencies are progressively socialized. This
process leads to an orientation to collective norms in the mid-range of the
scale, and thereafter norms are progressively internalized rules of norma-
tive conscience-the basis of "character." Even though there are stages
beyond this, there is no further overall transformation envisaged at the
upper end of the scale. There is only addition of features such as "toler-
ance," "respect for individuals," and "coping with" or "reconciling" inner
conflicts. These features are presumably meant to represent conscience
turned positive-disinterested and idealistic.
Moving to the second category, interpersonal style, we find shades of
Mead in that development at the lower end appears to be egoism becoming
altruism with correlative self-awareness. At the upper end, the valuing
13This notion is Hartmann's (1939/1958). Hartmann described the synthetic func-

tion as synthesizing three kinds of equilibration relations: the relation among
drives; the relation among ego, id, and superego; and the relation between ego and
reality.
of individuality appears, tempered with an awareness of the role that

"dependence" plays.
Under the third category, cogitive style, following Harvey, Hunt, and
Schroder (1961) and others, "simple" stereotyping is the primitive form,
which gives way to "complexity" in the middle range, which then expands
at the highest level to encompass a tolerance for cognitive "ambiguity."
This latter notion, reminiscent of the therapeutic goal of "integrating
ambivalence," appears to be considered as a form of objectivity (Table
9.1).
Finally, under conscious preoccupations, concern with things, desires,
and bodily feelings progresses to a focus on adjustment, then to reflexive
self-description, and ultimately to a preoccupation with understanding per-
sonality in general in a quasi-theoretical fashion. Loevinger wants to
amalgamate perception of self with perception of other selves at each ego
level. Each level of world view therefore manifests two kinds of "conscious
preoccupation." The first corresponds to what symbolic interactionists call
"self-concept." The second corresponds to the contents of "social cogni-
tion," those internal psychological characteristics of others (e.g., inten-
tions, motives, and feelings) that help to explain and predict their be-
havior. This social-cognitive approach resembles the recent neo-Piagetian
trend in theory and research about the development of social knowledge
(see, for example, Damon, 1978, 1983; Flavell & Ross, 1981).
In general, we can note a certain arbitrariness in the sequencing of
levels, the confusing overlapping of categories, a supplanting of qualitative
changes with quantitative ones at the upper levels, and a lack of clear con-
tinuity of theme within a domain across the different levels. The empiricist
may well take heart at such observations, since they might serve to reassure
us that reality itself is at work, resisting the imposition of any overly sim-
plified framework. However, there is another possibility: we are beginning
to detect tensions between experience and ideology that lead theorists like
Loevinger to hang on to a specific conceptualization despite its failure to
accommodate the data in a satisfactory manner .14
Ultimate Virtues of the Developed Ego

Loevinger assumes that the basic terms used to outline the levels of ego
development are commonsensical and self-evident descriptors, requiring
no reflection or justification. Since the majority of the population suffers
14 Loevinger is not the only culprit in this regard. Such conservativism is endemic to
developmental approaches (as documented and discussed in a recent literature re-
view, Broughton, 1983b), and perhaps to all theory building (Kuhn, 1962; Lakatos,
1978).
from the "natural attitude,"15 being considerably below the higher ego
levels, it is not so surprising that most consumers accept Loevinger's in-
genuous stance without resistance. However, on inspection, the "terms"
turn out to be "constructs" carrying precisely the inferential baggage that
she decries.
Readers whose suspicion may have been aroused by our preliminary re-
connaissance may wonder where best to initiate their 'site visit' to the
theory of ego development. A reasonable starting point is the examination
of the virtuous characteristics that are assumed to define the end point of
development. These serve to identify the most perfected form that the ego
possibly could take, and which is in the nature of all egos to strive
toward. 16 The tetos or goal of growth in a developmental theory not only
defines maturity, but serves to order all previous levels and explain why
and how they are legitimately construed as intermediate steps on the way
to the final stage (Broughton, 1979a). It is the coherence and justifiability
of the definition of this apex that underwrites any attempt to apply the
theory in acts of education, therapy, or social change (Kohlberg & Mayer,
1972). In a theory of psychological development, the final stage must
contain, either explicitly or implicitly, a set of assumptions about human
nature (Kaplan, 1967). Despite Loevinger's reluctance to admit it, this
generalization applies to her theory as well.
Although Loevinger never defines personal maturity in any explicit way,
it is possible to identify certain 'silent stipulations.' She seems to be pro-
posing a general notion that is not far from Jung's (1921) original concept
of life as a process of "individuation." Although Loevinger's is not a trait
psychology, and the variables mentioned possess "milestone" rather than
bipolar characteristics, there appear to be at least five major qualities that
are implicit in her characterization of the fully individuated ego: individual-
ity, self-awareness, complexity, wholeness, and autonomy,17 These are
her psychological and moral "ends," to which ego development is the
"means."
INDIVIDUALITY
In the course of development summarized in Table 9.1, Loevinger uses, as
a central criterion for maturity, "individuality," by which she appears to
15This term, coined by Husserl (1929/1977), refers to the "everyday, unreflective

attitude of naive belief in the existence of the world" (Spiegelberg, 1969, p. 715).
16Unlike some developmental theorists (Fowler, 1981; Kohlberg, 1978), whose
highest stages are pure idealizations that find virtually no human instantiations
(Broughton, 1986a, 1986b), Loevinger claims to find exemplars of her highest ego
level quite frequently.
17Not by coicidence, these are the psychological axes of "liberalism" (Lukes, 1973;
Unger, 1975), which I have commented on in detail elsewhere (Broughton, 1985).
mean "individual difference" or "uniqueness. "18 However there are cer-

tain problems with this way of describing the differentiation of the self.
First, uniqueness is too general a criterion: everything, including each in-
animate object, is unique in some way (Hegel, 1816/1975). Second, any
peculiarity must be defined in relation to contrasting others, and both the
dimension of comparison and the criterion for what should count as an
"other" implies certain prior axes of similarity between the beings so con-
trasted (Husserl, 1929/1977, p. 7). Third, the appeal to individual differ-
ences calls for a relativistic tolerance that is in tension with Loevinger's
own hierarchical arrangement of levels; some unique beings would appear
to be more equal than others. Fourth, her suggestion that at higher levels
we come to respect the uniqueness of others would appear to exclude from
respect someone who showed no differences (i.e., the total conformist).
SELF-AwARENESS
Ultimate selfhood is also characterized by Loevinger in terms of "self-
awareness." Unfortunately, the kind of objectifying reflexivity she pro-
poses is that of the "self-concept"-James's (1890) or Mead's (1932)
"me." It is not possible to equate this empirical ego with the "I," as Mc-
Adams (1985) has attempted to do, since it involves no subjective self (Kuhn
& McPartland, 1965; Moore, 1933). Ironically, in a chapter included by
Loevinger in her book, Blasi (1976) points out that an ego conceptualized
in such terms fails to include reflective consciousness, and so is no self at
all, a fact that, quite recently, she has come to acknowledge (Loevinger,
1986).
Admittedly, Loevinger does offer an alternative vision of self-awareness
as self-monitoring or self-correcting. However, such cybernetic "feed-
back" notions do not require reflectivity either, or even consciousness,
since self-correction is characteristic of even the most primitive servo-
mechanisms. Neither cybernetic self-guidance nor symbolic interactionist
self-observation can escape infinite regress (Broughton, 1980a; Broughton
& Riegel, 1977). Moreover, they are profoundly amoral conceptualiza-
tions, incapable of grounding the activities of judgment or responsibility
central to all human mentation (Blasi, 1975, 1978, 1983; Lonergan, 1968).
COMPLEXITY
As McAdams (1985) points out, the levels of ego development represent a
hierarchy of complexity. Loevinger has followed the trend set by the func-
18She also uses the phrase "respect for individuality," not always distinguishing it
from "individuality." By respect for individuality, she appears to mean respect for
individual differences and human uniqueness.
tionalist cognitive psychology of the 1950s and 1960s that took complexity
in cognitive style as a probabilistic sign of maturity. She is clearly correct to
emphasize the role of emotional relatedness in personality development.
But Loevinger's concern with her subjects is for the complexity of their
typical responses, not the complexity of their overall awareness or their
status as complex persons. Even if Loevinger's approach were able to allow
for the identification of complex people, it would not necessarily follow
that these were "developed" people, let alone normal or healthy people.
For example, neurosis adds greatly to our complexity, but it is not one of
the best indices of personal maturity.
WHOLENESS
A fourth criterion of individuation for Loevinger is "wholeness"-internal
consistency. The problems of holistic synthesis were foreshadowed above
with regard to the overlap between the "categories" or domains of ego
development. How exactly the synthetic function is to contain itself so that
the requirement of wholeness does not conflict with the demand for
complexity, Loevinger's theory does not make clear. 19 At the limit, a syn-
thetic tendency turns into a dedifferentiating tendency. As the work of
Chasseguet-Smirgel (1984) has shown, there is a sadistic component to nor-
mal aggression, pressing toward homogenization-one that is easily con-
founded with the quest for wholeness. Perhaps it is no coincidence that, as
part of her reaction against the sexual focus of Freudian theory, Loevinger
(1966) calls for more attention to the aggressive side of human nature.
An additional problem is that where wholeness becomes an end to be
maximized, there is no moral quality to the synthetic function that would
allow us to trust it to distinguish between consolidating the consistency of
personality and consolidating the fortification of its defensive structure.
Nunberg (1931, p. 136) himself acknowledges this difficulty. Although
Loevinger (1976a, p. 69; 1983, p. 347) claims that her conceptualization of
the ego is close to Sullivan's, she has missed his central insight-that the
"self-system" is a protective structure of anti-anxiety operations. Seamless
menders can tailor battle dress just as well as civilian wear.
AUTONOMY
The last and perhaps most central of Loevinger's criteria is the degree of
autonomous status of the individual. What this individualist approach to
freedom ignores is the role of the individual as subsystem in the larger
social whole. As in Erikson's notion of "identity" (Kovel, 1974), becoming
19Faced with this problem, Hartmann (1939/1958) in fact proposed a differentiating

function to balance its synthetic counterpoint.
autonomous is not distinguished from becoming an anonymous and substi-

tutable occupant of a social role. "Individuality" is quite compatible with
being a means to the "greater" good of utility to the system, and when it is
construed primarily in systemic terms, it actively promotes such subordina-
tion (Horkheimer, 1974).
With regard to moral autonomy, Loevinger adopts a relativistic, utili-
tarian ethics within which the moral is conceived teleologically as the non-
rational control and socialization of irrational impulses. Such a "conflict
resolution" approach cannot transcend conventionality (Blasi, 1975) and,
in fact, runs counter to any questioning of illegitimate authority (Zahayke-
vich, 1982, 1983). Autonomy, in ego psychology, is therefore a spurious
freedom, lacking emancipation (Broughton, 1986b). This fits with the fact
that the synthetic function places the wholeness of the individual over and
above the wholeness of society, thereby neglecting benignly the stratified
condition of the latter. Political apathy is certainly more "adaptive" than
public moral opposition (Habermas, 1975b). Perhaps this is why the "re-
nunciation of the unattainable" is posited by Loevinger (Loevinger &
Wessler, 1970, Vol. 2) as one of the hallmarks of the higher levels of ego
development for women, if not for all people.
Developmental Theory and Ideology

For all five criteria discussed-individuality, self-awareness, complexity,
wholeness, and autonomy-the shortcomings are closely related, reflecting
the problems of liberal individualism in its various aspects. 20 The higher
ego levels, therefore, cannot stake any claim to being more rational than
the lower ones. Adaptation has never been either an adequate guarantee
or an acceptable arbiter of objectivity (Blanshard, 1939). In addition,
Loevinger's higher ego levels lack the potential for genuine sociality, resis-
tance to illegitimate authority, or concern for freedom from oppression. As
they are examined in detail, these higher levels, which at first appear to
represent the pinnacle of personal maturity, come to seem much more con-
ventional. Loevinger has been careful to point out that her higher levels
2°It is important to realize that individualism is a much more sophisticated world

view than is often realized (Broughton, 1978). For example, it takes as one of its
basic notions that individuals naturally and successfully tend to form associations
with each other (Waterman, 1981), and that out of this cooperative interaction of
individuated beings emerges the possibility of democratic order (McPherson,
1962). It is in this sense that individualism is compatible with Loevinger's social
virtues of the autonomous person (respect for individuals, empathic understanding,
etc.), much as it is with Kohlberg's social role-taking processes (Broughton,
1986b). For astute critical analyses of "liberal psychology," see Unger (1975) and
Sullivan (1984).
(from "Conscientious" upwards) are postconformist rather than postcon-

ventional. However, as even Kohlberg (1969) has made clear, transcending
conformity does not lead to the transcendence of conventionality.
The lesson of Loevinger's scheme may be that we have collectively
underestimated the sophistication and power of conventional thinking.
This conclusion fits Kohlberg's finding, in the domain of moral develop-
ment, that much of what had been taken previously as principled judgment
was in fact purely conventional in structure (Colby, 1978; Gibbs, 1979).
Loevinger's developmental scheme is conventional throughout in the
sense that it depicts the phase-by-phase acquisition of conventional liberal
individualism. Loevinger herself manifests a remarkably consistent ego,
subscribing whole-heartedly to all of the basic precepts of liberalism, and
setting these up as empirically identified yet self-evident end points of
development. The consequence is that none of her subjects, or at least
none of their "egos," transcends the basic precepts of liberalism.
Developmental theories employ a sleight of hand by which they appear
to describe the emergence of truly higher orders of human functioning,
while actually naturalizing and legitimizing a specific, preferred world view
-one that, by no coincidence, happens to correspond to the dominant
form of scientific, moral, and juridical order in Western democracies. In-
sofar as liberalism embodies a critique of the conventions of other world
views (e.g., conservatism), rejecting them as less advanced than its own,
socialization into the normative "autonomous" character structure gives
the impression of a revolutionary, postconventional maturity and indepen-
dence. A lot hangs on that felicitous prefix post and its rhetorical overtones
of superiority.
Another way of exposing this legerdemain is to point out that neither
postconformist nor postconventional implies postideological. It is precisely
the ideological, self-deceptive nature of conventional mentality that allows
it to present itself as something that it is not: as supraconventional. One
may criticize the dignification of the transit into mature liberal ideology in
just the same way that one may question the way in which ego develop-
ment theory celebrates the adolescent and early adult transit into con-
formist and conventional mentalities as a progressive movement. A theory
that begins by accepting the necessity of colluding with the established
order in the interest of gaining new levels of personal development cannot
then escape the implications of its own authoritarianism.
The mass popularity of liberal psychology, which has now attained an
impressive hegemony in Western democracies, and the scarcity of critical
response to it, can be attributed to its success in conjuring up a diminished
vision of "ideology." "Ideology" tends, within the liberal mainstream
psychology, to be reduced to the purely irrational level of mere beliefs or
political biases (Broughton, 1979b). Under that distorted interpretation
(the "ideology of ideology" as I have called it elsewhere [Broughton,
1976]), ideology seems irrational, and therefore nonstructural and
nonfunctional-in other words, like sex or class, irrelevant to a scheme of

developmental progression.
Such an impoverished assumption about ideology does an injustice to
the phenomenon. The power and significance of ideology reside in its abil-
ity to penetrate silently into every aspect of human development, even to
the point of making the notion of "development" itself fundamentally
ideological. 21
Collapse of Meaning
Loevinger's ambition has been no less immodest than a remodeling of
psychoanalytic theory in its entirety. Repeating the traditional strategy in
positivist psychology of questioning the mechanical connotations and spe-
culative status of the "drive" concept, she has deliberately displaced the
vicissitudes of instinct and the structural model embodying them by forging
a theory that is explicitly cognitive rather than psychosexual (Loevinger,
1962, p. 116). As her own programmatic statement points out (Loevinger,
1976a, p. 69), she has copied Sullivan's way of substituting for the motiva-
tional dynamics of Freudian theory the processes of psychologically repre-
senting interpersonal relations. 22
Loevinger's attempt to replace the id with the ego, to construe the ego
apart from defense, and to establish secondary process by displacing pri-
mary process are all at odds with the original dialectical intention of these
conceptual relationships.23 She has been consistently eager to collapse
basic distinctions and dismiss central concepts in psychoanalysis despite (a)
her confusion about what claims they represent24; (b) her inability to pro-
vide any conceptually grounded rejection of those notions; (c) her lack
of any reasoned contesting of the rather extensive body of clinical and
experimental evidence adduced by Freud and many others25 ; and (d)
the absence, in her own research, of any provision for dismissing psycho-
analytic concepts on empirical grounds.
21 This "ideology-critique" of the concept of development is presented more fully in

my recent collection, Critical Theories of Psychological Development (Broughton,
1987).
22See Sullivan (1953) and Loevinger (1985).
23See, for example, Horowitz (1977) and Ingleby (1983). In many ways, this sub-
version of the critical edge to psychoanalytic theory is reminiscent of Jung's original
departure from Freudian concepts (see, for example, Rieff [1963, 1966]). For a
detailed account of how this antidialectical trend has been played out in neo-
Freudian and post-Freudian psychology, see Jacoby (1975).
24 Loevinger is honest enough to admit her confusion, for example, regarding the
oedipal complex (Loevinger, 1962, p. 107).
25 See, for example, Edelson (1984), Fisher and Greenberg (1977/1985), Kline
(1981), Kubie (1952), and Sears (1943).
In an effort to claim that her theory represents an improved version of

psychoanalytic thinking, because it adds a sensitivity to the growth of
meaning and understanding, she includes Ricoeur in her "New Ego Para-
digm," which supposedly supersedes the Freudian "Drive Paradigm."
There is a great irony in this, since Ricoeur's theory hinges on the tension
between hermeneutics and energetics sustained by the psychoanalytic
tradition. His writings on psychoanalysis demonstrate the importance of
retaining the life historical dimension of the person, in which the "appro-
priation of tradition" plays as important a role as it does in social history.
The biographical approach to personality hinges on the demythologizing of
symptoms, anxieties, and defenses in terms of their origins-an "archaeol-
ogy of the subject" that Loevinger (1976b) has explicitly rejected in favor
of her "teleology of the subject" (p. 293). This ego-psychological ap-
proach, Ricoeur (e.g., 1970, p. 462) has shown, forecloses the mnemonic
significance of biography by attempting (a) to build a developmental
psychology of personality26; (b) to conceive the goal of life as "mastering
experience,"27; (c) to replace energic motivational constructs with structu-
ral, cognitive ones28 ; and (d) to displace the symbolic nature of meaning
with a rationalized, organizational one. 29
As Ricoeur warned, the humanistic approach of ego psychology, with its
desperate faith in the seamless civilizing of brute nature through scientific
mastery, collapses the internal divisions that the psyche reifies in its de-
fensive attempts and removes the sensitivity of theory to patterns of
self-deception and self-coersion that we know to be central to personality
formation. 3o The clinically observed phenomena of the identification and
amelioration of repressive and other defensive distortions, and the raising
of unconscious content to a conscious level are, within Loevinger's
framework, not comprehensible as comprising or contributing to develop-
mental advances. Conversely, as Loevinger (1976a, p. 385) herself has
asserted, the kind of autonomy attained at the higher ego levels is quite
26 1 have elaborated the Ricoeurian argument in favor of "theories of dissimula-

tion" over "theories of genesis" in relation to academic developmental psychology
elsewhere (Broughton, 1980b, 1983b).
27This phrase is Loevinger's (1976a, p. 404).
28Habermas (1975a) has also criticized the absence of any clear conception of need
or the expression of need in Loevinger's theory.
29Loevinger's scoring system sets explicit "taboos" on the symbolic interpretation
of sentence completion responses (Loevinger & Wessler, 1970, p. 21). Ricoeur's
concern that symbolic meaning and interpretation not be denied is echoed in
Habermas's (1970, 1971) hermeneutic revision of the theory of psychoanalytic ther-
apy, although his resolution differs in interesting ways from Ricoeur's (see Licht-
man, in press).
30The way in which distortion is endemic to the self-conceptualization of the ego
has been documented by Gecas (1982), Greenwald (1980), Marcus and Nurius
(1986), and Rosenberg (1981).
compatible with neurosis. In fact, according to the argument above, move-

ment into the higher ego levels would seem to be premised on a relatively
high level of defense formation. To quote Holt (1974):
Obsessive-compulsive characteristics such as wordiness, balancing of alternatives,
and introspective rumination are likely to be scored as indications of a fairly high
level of ego development. Indeed, in a number of ways the Conscientious level is a
good description of what is clinically called a well-compensated compulsive per-
sonality. (p. 312)31
Having given up the unconscious, dynamic, and self-deceptive dimen-
sion of meaning that would allow personality to be demythologized,
Loevinger is condemned to accept the individual's self-presentation at face
value. Personality is reduced to mind, and mental life is merely the arti-
culation of belief systems. The Washington Sentence Completion Test
thereby confounds development with the complexity of compensatory
neurotic mechanisms. The synthetic function, unconstrained by judgments
of morality or authenticity, is free to weave the synthetic fabric of a false
self.
Personality and Domination

It is interesting to note that, despite the debt owed to the postwar research
on the authoritarian personality (for example, Loevinger, 1962, 1983), the
socially critical insights ofthat work (e.g., Adorno, 1950) have been totally
obscured in the latest incarnation of ego development theory. 32 Loevin-
ger's approach to personality development screens out not only the
domination of inner nature, but also the domination of outer nature and
society.
In brief, at every point, the "New Ego Paradigm" tends either to ignore
the material reality of society or to psychologize it. Any societal source of
conflict is treated as an extrinsic and immutable given. "The autonomous
person. . . has the courage to acknowledge and deal with conflict rather
than ignoring it or projecting it onto the environment" (Loevinger, 1976a,
p.23).
Strangely, for a theory that is not accidentally based on research with
women, the costs of this "autonomy" are particularly clear with regard to
the ideal feminine role. "Autonomous" responses to the sentence stem "A
wife should ___ ," for example, are: "maintain her distinct identity, as
31 Holt's suggestion has since been confirmed empirically by Gold (1980).

32The historical link between theory and research on ego development and theory
and research on the authoritarian personality, and the gradual erasure of this link
as the former took shape, have been elaborated by Broughton and Zahaykevich (in
preparation). Compare the historical account of Samelson (1986).
that's probably what attracted her spouse, if she wants to keep him," and
"devote her life to her husband primarily, her children secondarily. She
should also be an informed intellectual companion and a lover too"
(Loevinger & Wessler, 1970, Vol. 2, p. 183). At the "Integrated" stage, a
woman is exhorted to make sense out of her conflictual roles "so that she is
free to make her husband happy" (Loevinger & Wessler, 1970, Vol. 2,
p.184).
Thus, for the sake of maturity, women are encouraged to take individual
responsibility for the inner reconciliation of conflicts around sex and gen-
der, rather than "projecting" them onto some fictive external reality, such
as the institutionalization of male dominance, presumably. The criteria of
ideal character development (Table 9.1) are unambiguous: "reconciling
inner conflicts" and "renunciation of the unattainable." Resistance to and
subversion of prevailing realities would imperil the requirement of ever-
increasing organizational cohesion and adaptation, and so constitute re-
gressive tendencies. It is testimony to the synthetic power of the system
and the seamless integration of psychology with social oppression that a
measure focused on women, and constructed by women, should orient
women away from the reality of their SUbjugation and back to the servicing
of men.
Conclusion
As Habermas (1975a) has pointed out, Loevinger's version of ego develop-
ment fails to include the way in which the needs of the individual are
brought forward through dialogue and action. When Loevinger arranges
the introjection of the origins of chaos and constraint into the person, she
excludes needs from the public symbolic universe, construing them (and
urging us to construe them) as "personal" and as part of an ahistorical
human nature rather than a culture in a process of social formation at a
particular historical juncture. In naturalizing the process by which indi-
viduals are cultivated, ego development theory draws attention away from
mores, institutions, and forms of interaction that foster the misinterpreta-
tion of needs and thereby undo developmental potentials. 33
Empirical research on personality development (e.g., Fowler, 1974;
Keniston, 1968; Zahaykevich, 1983) indicates that it is precisely the capac-
ity to perceive societal realities as contradictory and unjust, rather than as
"complex" or "holistic," that permits the expression of need, and the
33 A theoretical framework for understanding the way in which ideologically frozen

relations of dependence obstruct and pervert development has been articulated by
Habermas (1979) and colleagues (Dobert, Habermas, & Nunner-Winkler, 1987).
Empirical evidence of such inhibitive processes has been provided by the work of
Zahaykevich (1987).
transformation of understanding into dialogue and action that allows indi-

viduals (and groups) to develop beyond the conventional acceptance of
established ideology and social order. Ego development theory privatizes
experience, segmenting individual from individual, and dissolving any
motive for transformative speech, or action in the idealizing mental
transformations of the synthetic function.
Ironically, Loevinger's theory may not be untrue, despite being ideo-
logical. The ego development that it describes is most likely a more or less
accurate plotting of the apparently spontaneous, step-by-step production
of organized, adapted monads that exhibit a compliant balance between
their internal preoccupation with system governance and the external
norms of system stability. Insofar as Loevinger's theory tells us anything, it
is the narrative of a great tragedy. What the theory cannot account for is
the critically minded person who is socially, politically, and historically
aware, and who does not accept that particular dramatic work as the only
possible form of theater.
This conclusion does not imply the romantic belief in a naturally good
individual hemmed in by adverse social forces. The compulsive features of
persons at the "Conscientious" level and above are understandable, not as
the unfortunate consequence of extrinsic pressures imposed by society, but
as a consequence of the desperate quest for the impossible goal of abstract,
isolated individuality-as the cost of having a totally synthetic personality.
Genuine freedom cannot be restored by trying to insert sociality into the
person in the form of cognitive representation of interpersonal relations,
since that internalization carries into the self precisely those limitations of
human interaction that stabilize social orders by lending themselves to
institutionalization (Marcuse, 1955).
Insofar as a projective test taps a relatively unreflective sphere of mind,
the research findings on ego development serve as evidence of how deeply
the ideological justification of these limited relations has penetrated into
the psychology of the individual. To the extent that academic and profes-
sional psychology accept these findings at face value, this ideological jus-
tification will come to confront us as an apparently natural given, sanctified
in the norms of developmental psychology. An ideal balance will then be
struck between a fabricated reality and a synthetic individual-each well-
adapted to the other, and each contributing to the other's survival.
Acknowledgments. The origin of my interest in the current topic was a

graduate course taught in 1968 by Larry Kohlberg, whom we all now sadly
miss. My research into the related topic of self-concepts was encouraged by
Jane Loevinger herself, among others. The collegial relationship between
Larry and Jane was a major influence on all of us at the Center for Moral
Education at Harvard in the early 1970s. My first occasion to write about
ego development was provided by Paul Piccone, and the first attempt to do
so was fostered by Dan Candee.
Helpful remarks regarding an earlier draft of this chapter were pro-
vided by Jeff Atlas, Gus Blasi, Rainer Dobert, Howard Gadlin, Charles
Hecksher, Bob Holt, Jane Loevinger, Gil Noam, Ed Sullivan, and David
Wright. Most of these were not in agreement with the arguments presented
here, but despite their misgivings, they were generous enough to offer
constructive criticisms. Ray McDermott provided valuable bibliograhic in-
formation. I am especially grateful to Rich Ryan, of the University of
Rochester, whose friendly willingness to converse on issues pertaining to
personality and society has been instrumental in my renewal of interest in
the topic of self. Contact with Craig Barclay, Jim Connell, Dale Dannefer,
and Ed Deci at the University of Rochester has also contributed to the
writing of this chapter.
REFERENCES
Adorno, T.W. (1950). Types and syndromes. In T.W. Adorno, E. Frenkel-
Brunswik, D. Levinson, & N. Sanford. The authoritarian personality. New York:
Harper & Row.
Adorno, T.W. (1968). Sociology and psychology (part II). New Left Review, 47,
79-97.
Adorno, T.W., Frenkel-Brunswik, E., Levinson, D., & Sanford, N. (1950). The
authoritarian personality. New York: Harper & Row.
Baldwin, J.M. (1915). Genetic theory of reality. New York: Putnam.
Beilin, H. (1987). The new functionalism. In B. Inhelder & D. de Caprona (Eds.),
Piaget today. Hillsdale, NJ: Erlbaum.
Belenky, M.F., Clinchy, B.M., Goldberger, N.R., & Tarule, J.M. (1986).
Women's ways of knowing: The development of self, voice, and mind. New York:
Basic Books.
Benjamin, J. (1987). The decline of the Oedipus complex. In J.M. Broughton
(Ed.), Critical theories of psychological development. New York: Plenum.
Blanshard, B. (1939). The nature of thought (2 vols). London, England: Allen &
Unwin.
Blasi, A. (1975, August). Role-taking and the development of social cognition.
Paper presented at the American Psychological Association, Chicago.
Blasi, A. (1976). The concept of development in personality theory. In J. Loevin-
ger, Ego psychology: Concepts and theories. San Francisco: Jossey-Bass.
Blasi, A. (1978, August). The varieties of consciousness and the "bicameral" con-
sciousness. Paper presented at the American Psychological Association,
Toronto.
Blasi, A. (1983). The self and cognition: The roles of the self in the acquisition of
knowledge. In B. Lee & G.G. Noam (Eds.), Developmental approaches to the
self New York: Plenum.
Borgatta, E.F. (1961). Making a sentence test: An approach to objective scoring of
sentence completions. Genetic Psychology Monographs, 63,3-65.
202 lohn M. Broughton and Marta K. Zahaykevich
Broughton, I.M. (1976). Review of 1. Gabel, False consciousness. Telos, 29, 229-
237.
Broughton, I.M. (1978). Criticism and moral development theory. Journal Supple-
ments Abstract Service, 8.
Broughton, I.M. (1979a). The limits of formal thought. In R.A. Mosher (Ed.),
Adolescent education. Berkeley: McCutchan.
Broughton, I.M. (1979b). Dialectics and moral development. In P. Scharf (Ed.),
Readings in moral education. Minneapolis: Winston.
Broughton, I.M. (1980a). The history of the self: From substance to function. In
R.W. Rieber & K.W. Salzinger (Eds.), Psychology: Historical and theoretical
perspectives. New York: Academic Press.
Broughton. I.M. (1980b, April). Theories of genesis and theories of dissimulation.
Paper presented at the Conference on Adult Development and Adult Education,
Teachers College, Columbia University.
Broughton, I.M. (1981). Piaget's structural developmental psychology III. Func-
tion and the problem of knowledge. Human Development, 24(4), 257-285.
Broughton, I.M. (1983a). Women's rationality and men's virtues: A critique of
gender dualism in Gilligan's moral development theory. Social Research, 50(3),
597-624.
Broughton, I.M. (1983b). The cognitive developmental approach to adolescent self
and identity. In B.Lee & G.G. Noam (Eds.), Developmental approaches to the
self. New York: Plenum.
Broughton, I.M. (1985). The surrender of control: Computer literacy as political
socialization ofthe child. In D. Sloan (Eds.), The computer in education: A criti-
cal perspective. New York: Teachers College Press.
Broughton, I.M. (1986a). The political psychology of faith. In C. Dykstra &
s. Parks (Eds.), Faith and Fowler. Birmingham, AL: Religious Education Press.
Broughton, I.M. (1986b). The genesis of moral domination. In S. Modgil &
C. Modgil (Eds.), Lawrence Kohlberg: Consensus and controversy. Philadelphia:
Taylor & FrancislFalmer.
Broughton, I.M. (Ed.). (1987). Critical theories of psychological development. New
York: Plenum.
Broughton, I.M. (1988). Machine dreams: Graduate student fantasies about com-
puters. In R.W. Rieber (Ed.), Communication, individual, and society: Tribute
to Gregory Bateson. New York: Cambridge University Press.
Broughton, I.M., & Riegel, K.F. (1977). Developmental psychology and the self.
Annals of the New York Academy of Sciences, 291, 149-167.
Broughton, I.M., & Zahaykevich, M.K. (1977). Review of 1. Loevinger, Ego
psychology: Concepts and methods. Telos, 32, 246-253.
Broughton, I.M., & Zahaykevich, M.K. (1980). Personality and ideology in ego
development. In V. Trinh van Thao & 1. Gabel (Eds.), Dialectics in the social
sciences. Paris: Anthropos.
Broughton, I.M., & Zahaykevich, M.K. (in preparation). From authoritarian
personality to ego development: A critique of Loevinger's theory of personality
development.
Brunswik, E. (1947). The conceptual framework of psychology. In International
Encyclopedia of Unified Science (Vol. 1, Part 2). Chicago: University of Chicago
Press.
Buck-Morss, S. (1987). Piaget, Adorno and dialectical structures. In J.M. Brough-

ton (Ed.), Critical theories of psychological development. New York: Plenum.
Cameron, N. (1938). A study of thinking in senile deterioration and schizophrenic
disorganization. American Journal of Psychology, 51,650-644.
Candee, D. (1974). Ego developmental aspects of new left ideology. Journal of
Personality and Social Psychology, 30, 620-630.
Chasseguet-Smirgel, J. (1984). Creativity and perversion. New York: Norton.
Colby, A. (1978). Evolution of a moral-developmental theory. In W. Damon
(Ed.), New directions in child development: Vol. 2. Moral development. San
Francisco: Jossey-Bass.
Damon, W. (Ed.) (1978). New directions in child development: Vol. 1. Social cogni-
tion. San Francisco: Jossey-Bass.
Damon, W. (1983). Social and personality development: Infancy through adoles-
cence. New York: Norton.
Dobert, R., Habermas, J., & Nunner-Winkler, G. (1987). The development of the
self. In J.M. Broughton (Ed.), Critical theories of psychological development.
New York: Plenum.
Edelson, M. (1984). Hypothesis and evidence in psychoanalysis. Chicago: Univer-
sity of Chicago Press.
Fast, I. (1985). Event theory: A Piaget-Freud integration. Hillsdale, NJ: Erlbaum.
Fisher, S., & Greenberg, P.R. (1985). The scientific credibility of Freud's theories
and therapy. New York: Columbia University Press (originally published in 1977
by Harvester, Hassocks, England).
Flavell, J.H., & Ross, L. (Eds.). (1981). Social cognitive development: Frontiers
and possible futures. Cambridge, England: Cambridge University Press.
Fowler, J. (1974). The pilgrimage in faith of Malcolm X. The Foundation: Journal
of the Gammon Theological Seminary, Atlanta, 79,11-32.
Fowler, J. (1981). Stages offaith. New York: Harper & Row.
Gecas. V. (1982). The self-concept. Annual Review of Sociology, 8, 1-33.
Gergen, K.J., Hepburn, A., & Comer D. (1986). The hermeneutics of personality
description. Journal of Personality and Social Psychology, 6, 1261-1270.
Gibbs, J.e. (1979). Kohlberg's moral stage theory: A Piagetian revision. Human
Development, 22, 89-112.
Giddens, A. (1976). Functionalism: Apn!s la lutte. Social Research, 43(2), 325-
336.
Gilligan, C. (1982). In a different voice: Psychological theory and women's develop-
ment. Cambridge, MA: Harvard University Press.
Gold, S.N. (1980). Relations between level of ego development and adjustment
patterns in adolescence. Journal of Personality Assessment, 44(6),630-638.
Gouldner, A. (1970). The coming crisis in Western sociology. New York: Basic
Books.
Green, R.D., Ford, M.P., & Flamer, G.B. (1971). Measurement and Piaget. New
York: McGraw-Hill.
Green, M., & Rieber, R. W. (1980). The assimilation of psychoanalysis in America.
In R.W. Rieber & K.W. Salzinger (Eds.), Psychology: Historical and theoretical
perspectives. New York: Academic Press.
Greenwald, A.G. (1984, August). Totalitarian egos and totalitarian societies.

Invited address, American Psychological Association, Toronto.
Gutman, H. (1964). Structure and function. Genetic Psychology Monographs, 70,
3-56.
Habermas, J. (1970). Toward a theory of communicative competence. In H.-P.
Dreitzel (Ed.), Recent sociology (vol. 2). New York: Macmillan.
Habermas, J. (1971). Knowledge and human interests. Boston: Beacon.
Habermas, J. (1975a). Moral development and ego identity. Telos, 24, 41-55.
Habermas, J. (1975b). Legitimation crisis. Boston: Beacon.
Habermas, J. (1979). Communication and the evolution of society. Boston: Beacon.
Harvey, O.J., Hunt, D.E., & Schroder H. (1961). Conceptual systems and per-
sonality organization. New York: Wiley.
Hartmann, H. (1958). Ego psychology and the problem of adaptation. New York:
International Universities Press (originally published 1939).
Hauser, S. (1976). Loevinger's model and measure of ego development: A critical
review. Psychological Bulletin, 33 (5), 928-955.
Hegel, G.W.F. (1975). Logic. New York: Oxford University Press (originally pub-
lished 1816).
Hess, RD., & Handel, G. (1959). Family worlds. Chicago: University of Chicago
Press.
Hobhouse, L.T. (1913). Development and purpose. London: Macmillan.
Holt, RR. (1974). Review of Measuring ego development. Journal of Nervous
and Mental Diseases, 158(4), 310-316.
Horkheimer, M. (1947). Rise and decline of the individual. In Eclipse of reason.
New York: Oxford University Press.
Horowitz, G. (1977). Repression. Toronto: University of Toronto Press.
Husserl, E. (1977). Cartesian meditations: An introduction to phenomenology.
Hague: Martinus Nijhoff (originally published 1929).
Ingleby, D.J. (1983). Freud and Piaget: The phoney war. New Ideas in Psychology,
1 (2), 145-152.
Jacoby, R (1975). Social amnesia: A critique of contemporary psychology from
Adler to Laing. Boston: Beacon Press.
James, W. (1890). Principles of psychology. New York: Holt.
Jung, C.G. (1921). Psychological types: Collected works (vol. 6). Princeton, NJ:
Princeton University Press.
Kaplan, B. (1967). Meditations on genesis. Vita Humana (Human Development),
10,65-87.
Keniston, K. (1968). Young radicals: Notes on committed youth. New York: Har-
court, Brace & World.
Kohlberg, L. (1958). The development of modes of moral thinking and choice in the
years ten to sixteen. Unpublished doctoral dissertation, University of Chicago.
Kohlberg, L. (1969). Stage and sequence. In D. Goslin (Ed.), Handbook of
socialization research. Chicago: Rand McNally.
Kohlberg, L. (1973). Continuities and discontinuities in childhood and adulthood
moral development. In P. Baltes & K.W. Schaie (Eds.), Lifespan developmental
psychology: Personality and socialization. New York: Academic Press.
Kohlberg, L. (1978). Revisions in the theory and practice of moral development. In
W. Damon (Ed.), New directions in child development: Vol. 2. Moral develop-
ment. San Francisco: Jossey-Bass.
Kohlberg, L. (1980). Meaning and measurement of moral judgment: Heinz Werner

memorial lectures. Worcester, MA: Clark University Press. (Reprinted in
L. Kohlberg (1984), The psychology of moral development. New York: Harper &
Row.)
Kohlberg, L., & Armon, C. (1984). Three types of stage models used in the study
of adult development. In M.L. Commons, F.A. Richards, & c. Armon (Eds.),
Beyond formal operations: Late adolescent and adult cognitive development. New
York: Praeger.
Kohlberg, L., & Mayer, R. (1972). Development as the aim of education. Harvard
Educational Review, 42, 449-496.
Kovel, J. (1974). Erik Erikson's psychohistory. Social policy, March-April, 60-64.
Kline, P. (1981). Fact and fantasy in Freudian theory, (2nd ed.). New York:
Methuen.
Kubie, L.S. (1952). Problems and techniques of psychoanalytic validation and
progress. In E. Pumpian-Mindlin (Ed.), Psychoanalysis as science. Stanford:
Stanford University Press.
Kuhn, M.H., & McPartland, T.S. (1965). An empirical investigatiou of self
attitudes. American Sociological Review, 19,68-76.
Kuhn, T.S. (1962). The structure of scientific revolutions. Chicago: University of
Chicago Press.
Labov, W. (1974). The art of sounding and signifying. In Language and its social
setting. Washington, DC: Anthropological Society of Washington.
Lakatos, I. (1978). The methodology of scientific research programmes. Cambridge,
England: Cambridge University Press.
Langer, J. (1969). Theories of development. New York: Holt, Rinehart & Winston.
Leites, N. (1971). The new ego. New York: Science House.
Lichtman, R. (in press). Habermas' hermeneutics of psychotherapy. New Ideas in
Psychology.
Loevinger, J. (1962). Measuring personality patterns of women. Genetic Psycholo-
gy Monographs, 65, 53-136.
Loevinger, J. (1966). The meaning and measurement of ego development. Amer-
ican Psychologist, 21(3), 195-206.
Loevinger, J. (1969). Theories of ego development. In L. Breger (Ed.), Clinical-
cognitive psychology: Models and integrations. Englewood Cliffs, NJ: Prentice-
Hall.
Loevinger, J. (1976a). Ego development: Conceptions and theories. San Francisco:
Jossey-Bass.
Loevinger, J. (1976b). Origins of conscience. In M.M. Gill & P.S. Holzman (Eds.),
Psychology versus metapsychology: Psychoanalytic essays in memory of George
S. Klein. New York: International Universities Press [Psychological Issues, 9(4),
Monograph 36].
Loevinger, J. (1979). Construct validity of the sentence completion test of ego
development. Applied Psychological Measurement, 3(3),281-311.
Loevinger, J. (1983). On ego development and the structure of personality. De-
velopmental Review, 3,339-350.
Loevinger, J. (1984). On the self and predicting behavior. In R.A. Zucker, J.
Aronoff, & A. Rabin (Eds.), Personality and the prediction of behavior. New
York: Academic.
Loevinger, J. (1985). A century of character development. In S. Koch and D.E.
Leary (Eds.), A century of psychology as science. New York: McGraw Hill.

Loevinger, J. (1986). On Kohlberg's contributions to ego development. In S.
Modgil and C. Modgil (Eds.), Lawrence Kohlberg: Consensus and controversy.
Philadelphia: Falmer/Taylor & Francis.
Loevinger, J., & Blasi, A. (1976). Issues in defining stages and types. In J. Loevin-
ger, Ego development: Conceptions and theories. San Francisco: Jossey-Bass.
Loevinger, J., & Wessler, R. (1970). Measuring ego development. San Francisco:
Jossey-Bass. (2nd ed. 1978).
Lonergan, B.J.F. (1968). The subject. Milwaukee: Marquette University Press.
Looft, W.R. (1973). Socialization and personality throughout the lifespan: An
examination of contemporary psychological approaches. In P. Baltes & K.W.
Schaie (Eds.), Lifespan developmental psychology: Personality and socialization.
New York: Academic Press.
Lorge, I., & Thorndike, E.L. (1941). The value of the responses in a completion
test as indications of personal traits. Journal of Applied Psychology, 25, 191-199.
Lukes, S. (1973). Individualism. New York: Harper & Row.
Marcus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41(9),
954-969.
Marcuse, H. (1955). Critique of neo-Freudian revisionism. In Eros and civilization.
Boston: Beacon Press.
Martindale, D. (1960). The nature and types of sociological theory. Boston:
Houghton Mifflin.
Marx, K. (1976). A contribution to the critique of political economy. Peking: For-
eign Languages Press (originally published 1859).
McAdams, D.P. (1985). Power, intimacy and the life story. Homewood, IL:
Dorsey.
McClelland, D.C. (1951). Personality. New York: Dryden.
McPherson, C.B. (1962). Possessive individualism: From Hobbes to Locke. New
York: Oxford University Press.
Mead, G.H. (1932). Mind, self and society. Chicago: University of Chicago Press.
Moore, J.S. (1933). The problem of self. Philosophical Review, 62,487-499.
Murray, H.A. (1951). Toward a classification of interactions. In T. Parsons & E.A.
Shils (Eds.), Toward a general theory of action. Cambridge, MA: Harvard Uni-
versity Press.
Nunberg, H. (1931). The synthetic function of the ego. International Journal of
Psychoanalysis, 13, 129-150.
Nunberg, H. (1955). The psychology of the ego. In Principles of psychoanalysis:
Their application to the neuroses. New York: International Universities Press.
(First published in German by Hans Huber Verlag, 1932)
Parsons, T., & Shils, E.A. (Eds.). (1951). Toward a general theory of action. Cam-
bridge, MA: Harvard University Press.
Payne, A.F. (1928). Sentence completions. New York: New York Guidance Clinic.
Pepper, S. (1966). Concept and quality. Lasalle, IL: Open Court.
Peterfreund, E. (1971). Information, systems and psychoanalysis. Psychological
Issues, 71(1-2) (Monograph 25-26).
Rapaport, D., Gill, M., & Schafer, R. (1968). Diagnostic psychological testing.
Redmore, c., Loevinger, J., & Tamashiro, R. (1978). Measuring ego development:
Scoring manual for men and boys. Unpublished manuscript, Washington Uni-
versity, St. Louis.
Redmore, C., Wright, D., & Rashbaum T. (1974). Measuring ego development:
Scoring manual for men and boys. Unpublished manuscript, Washington Uni-
versity, St. Louis.
Reisman, K. (1974). Noise and order. In Language and its social setting. Washing-
ton, DC: Anthropological Society of Washington.
Ricoeur, P. (1970). Freud and philosophy. New Haven, CT: Yale University Press.
Rieff, P. (1963). Introduction to S. Freud, The history of the psychoanalytic move-
ment. New York: Collier.
Rieff, P. (1966). Triumph of the therapeutic: Uses offaith after Freud. New York:
Harper & Row.
Rosenberg, M. (1981). The self-concept: Social product and social force. In M.
Rosenberg & RH. Turner (Eds.), Social psychology: Sociological perspectives.
New York: Basic Books.
Rotter, J.B. (1951). Word association and sentence completion methods. In H.H.
Anderson & G.L. Anderson (Eds.), An introduction to projective techniques.
New York: Prentice-Hall.
Samelson, F. (1986). Authoritarianism from Berlin to Berkeley: On social psychol-
ogy and history. Journal of Social Issues, 42(1),191-208.
Sanford, R.N. (1943). Physique, personality, and scholarship. Monographs of the
Society for Research in Child Development, 8(1).
Sears, RR. (1943). Survey of objective studies of psychoanalytic concepts. Bulletin
51, Social Science Research Council, New York.
Smelser, N.J., & Erikson, E. (1980). Themes of work and love in adulthood. Cam-
bridge, MA: Harvard University Press.
Snarey, J., Kohlberg, L., & Noam, G.G. (1983). Ego development in perspective:
Structural stage, functional phase, and cultural age period models. Developmen-
tal Review, 3, 303-338.
Spiegelberg, H. (1969). The phenomenological movement: A historical introduc-
tion. Hague: Martinus Nijhoff.
Sullivan, H.S. (1953). Interpersonal theory of psychiatry. New York: Norton.
Sullivan, E.V. (1984). A critical psychology: Interpretation of the personal world.
New York: Plenum.
Swensen, C.H. (1980). Ego development. In RH. Woody (Ed.), Encyclopedia of
clinical assessment (vol. 1). San Francisco: Jossey-Bass.
Symonds, P.M. (1947). The sentence completion test as a projective technique.
Journal of Abnormal and Social Psychology, 42,320-329.
Taylor, C. (1971). Interpretation and the sciences of man. Reivew of Metaphysics,
13,1-51.
Tendler, A.D. (1930). A preliminary report on a test for emotional insight. Journal
of Applied Psychology, 14, 123-136.
Turiel, E. (1983). Domains and categories in social-cognitive development. In
W.F. Overton (Ed.), Relationship between social and cognitive development.
Unger, RM. (1975). Knowledge and politics. New York: Free Press.
Vincent, L.R., & Vincent, K.R (1979). Psychological Reports, 44,408-410.
Waterman, A. (1981). Individualism and interdependence. American Psychologist,
36(7), 762-773.
Wright, D. (1975). Human development and sociology. Unpublished doctoral dis-
sertation, Washington University, St. Louis.
Zahaykevich, M.K. (1982, April). Critical approach to adult women's development.
Paper presented at the Conference on Adult Life Transitions, Teachers College,

Columbia University.
Zahaykevich, M.K. (1983). Moral development and autonomy in Soviet human
rights activists. Unpublished doctoral dissertation, Teachers College, Columbia
University.
Zahaykevich, M.K. (1987, April). Mother-daughter dialogue and the individuation
of adolescent girls: A speech act analysis. Paper presented at Society for Research
in Child Development, Baltimore.
Zavalloni, M. (1986). The affective representational circuit as the foundation of
identity. New Ideas in Psychology, 4(3),333-350.
Part III Identity
10
Common Processes Underlying
Ego Identity, Cognitive/Moral
Development, and Individuation
JAMES E. MARCIA
This chapter discusses the interrelationships among three major theo-

retical streams in developmental personality theory: psychosocial (ego
psychoanalytic), cognitive/moral, and object relations. The starting point
for looking at this proposed interrelationship is with the psychosocial con-
cept of ego identity. The first section of this chapter concerns some theo-
retical and empirical justification for relating ego identity status to cogni-
tive development, moral development, and the object relational concept of
individuation. The second section of the chapter deals with these three
approaches from the standpoint of their similar developmental goals,
underlying processes, and child-rearing implications.
Ego Identity, Object Relations, and

Cognitive/Moral Development
EGO IDENTITY STATUS

The most important single concept, and the only truly structural one, in
Erikson's (1959) theory of psychosocial development is ego identity.
Although the more formal aspects of his theory, especially the notion of an
epigenetic ego developmental sequence, may be a more significant con-
tribution than any particular stage-specific content, among the specific con-
tents of the eight proposed life-cycle stages, identity is the most central.
One effect of the prominence that has accrued to the identity concept has
been the extension of the final significant developmental point upward into
adolescence from the classical psychoanalytic oedipal period. Similarly, be-
cause of their emphasis on the centrality of introjects for the development
of a self-structure, the object-relations theorists have pushed the develop-
mental sequence back to infancy as a significant phase-again in distinc-
tion to the classical analytic oedipal position.
212 James E. Marcia
Erikson's theory was conceived within, and intended to be applicable to,

the practice of psychoanalytic psychotherapy. However, the theory also
furnishes a potentially testable scheme of personality development. Re-
search into Erikson's theory has progressed furthest on the crucial concept
of identity, and the paradigm most frequently used has been the identity
status approach (Bourne, 1978a, 1978b; Marcia, Waterman, and Matte-
son, in preparation).
The identity status approach to the measurement and validation of the
construct, ego identity, involves categorizing individuals into one of four
identity "styles" or statuses based upon the two process criteria of explora-
tion of alternatives and degree of subsequent commitment to some alterna-
tives. The specific content areas in which exploration and commitment are
determined are occupational choice, religious and political ideology, and
sexual-interpersonal values. If an individual is committed, he or she may
be classified as Identity Achievement if the commitments are preceded by
exploration, and as Foreclosure if no exploration has been undertaken. If
an individual is deficient in commitment, he or she may be classified as
Identity Diffusion if little or no commitment is present and not much con-
cern is evinced, or as Moratorium if his or her commitments are vague and
he or she is struggling to form them. Construct validity for the identity
statuses has been established in numerous studies over the past 20 years.
(See reviews in Bourne, 1978a, 1978b; Marcia, 1980; Marcia, Waterman,
& Matteson, in preparation.)
RESEARCH IN IDENTITY
What has emerged as surprising in the identity status research is the unex-
pected longevity of the constructs and their relationship to constructs from
other theoretical realms such as cognitive/moral development and object
relations. The connection between identity and formal operations per se is
not clear-cut; some studies find a correlation between the two and some do
not. It appears that the closer a cognitive measure approaches a real-life
situation, (e.g., cognitive integrative complexity [Slugoski, Marcia, &
Koopman, 1984]), the greater its relationship to identity (Marcia et al., in
preparation). This hypothesis is borne out most clearly in the consistent
relationships that have been found between levels of moral reasoning and
identity (Hult, 1979; Leiper, 1981; Podd, 1972; Poppen, 1974; Rowe &
Marcia, 1980; Skoe, 1986). With respect to object relations, Kroger (1985),
using Hansburg's (1980) index of separation-individuation, found that per-
sons high in identity were less anxiously attached and more personally
secure in dealing with separation issues than were persons in lower identity
statuses. Looking at intimacy development, a concept closely related to
identity, Levitz-Jones and Orlofsky (in press) found a relationship between
mature handling of separation issues and intimacy status.
10. Ego Identity, Cognitive/Moral Development, and Individuation 213
DEVELOPMENTAL LINKAGES
Development of Ego Identity
As noted before, the identity statuses are defined on the basis of the twin
criteria of exploration and commitment. Exploration refers to a willingness
to consider possible future directions other than just those that have been
parentally given. Exploration involves a departure or separation from
one's origins, an active taking into account of one's special abilities and
needs. Commitment refers to an evental adherence to a course of action, a
kind of settling down. In more formal theoretical terms, commitment is the
beginning of structure coalescence. The identity structure itself, an orga-
nized composite of these commitments, is the result of ego synthesis (a
primary autonomous ego function). It is important to note here that this
initial identity structure, if it is not a foreclosed one, can be expected to be
reformulated throughout the life cycle. The process involved again is ex-
ploration and commitment; however, now what is departed from is not
one's parental plans (i.e., one's unquestioned identifications), but one's
own self-fashioned identity structure.
Identity development, then, can be expected to involve a lifelong pro-
cess of change. This change proceeds from a position of initial structure,
hypothesized to take shape for the first time at late adolescence. When this
initial structure is either too rigid (as a function of its being parentally
conferred rather than individually constructed-Foreclosed), or too
amorphous (as a result of alienation from parents-Diffused), identity
reformulation is less likely to occur throughout the life cycle. Although
longitudinal research on this issue is scanty, that which does exist sup-
ports the idea of "stuckness" in identity development for the Foreclosure
and Diffusion statuses (Waterman, 1982).
Some speculations on the childhood antecedents of identity formation
were advanced by Schiedel and Marcia (1985). They found that identity
seemed to relate to masculine sex-typing (Bern, 1977), especially for
women; that intimacy seemed to relate to feminine sex-typing, especially
for men; and that high identity coupled with high intimacy was related to
androgyny. They suggested that the origins of at least the exploration
aspect of identity might lie in the oedipal period when children learn the
role requirements of their gender. Their hypothesis was that the extent of
guilt- and anxiety-free experimentation that was allowed during the oedi-
pal period had implications for the development of a more flexible superego,
especially the ego ideal aspect.
Identity and Object Relations

The object-relations theorists considered most relevant for this essay are
Margaret Mahler and John Bowlby. Also important for her extension of
Bowlby's theory via empirical study is Mary Ainsworth (Ainsworth, Ble-
214 James E. Marcia
har, Waters, & Wall, 1978). In general, the developmental pattern that
these theorists describe is one in which the child proceeds from embedded-
ness to differentiation to individuation. The result of this process is the
child's sense of a separate and autonomous self, independent of the
mother-child matrix. This separate self is a necessary basis for the later
development of a pychosocial identity. An example of the relationship
between a sense of separate self and a psychosocial identity is found in the
contrast between the disturbance of identity found in the borderline pa-
tient and the identity crisis of a late adolescent. In the borderline case, the
individual has split off separate selves in order to cope with ambivalence,
and she or he maintains this unintegrated (unmetabolized) state into adult-
hood via splitting and projective identification (Kernberg, 1975). In other
words, in the case of the borderline whose difficulties lie in Mahler's
"rapprochement" phase, the essential substructure for later psychosocial
identity formation (an integrated self) is missing. In the case of the late
adolescent in an identity crisis, where basic sense of self is present, the
struggle, usually time-limited, is to achieve a "fit" between various
expressions of that self (e.g., occupational, ideological, and sexual-inter-
personal) and social institutions.
Intervening, psychologically and temporally, between the attachment-
separation-individuation sequence of early childhood and the late adoles-
cent sequence of exploration-commitment-identity is the individual's rela-
tionship with his family. The first major separation is from one's mother;
the second major separation is from one's family (and their introjected
representatives). The young child leaves the mother to explore the world;
the late adolescent leaves the external and internal family to make his
unique place in that world. What is clear from Bowlby's (1969), Ainsworth's
(Ainsworth et aI., 1978), and Mahler's (Mahler, Pine, & Bergman, 1975)
work with young children and from our work on identity development
(Adams, Ryan, Hoffman, Dobson, & Nielsen, 1985; Cooper, Grotevant,
& Condin, 1983; Grotevant & Cooper, 1985) is that secure attachment
precedes individuation and identity. The identity status that has the most
difficulty in life, the Identity Diffusion, is also the status that feels most
alienated from parental figures. Moreover, the anxious or ambivalent
attachment that results in clinging children is very likely mirrored by the
tenacious hold that our Foreclosed-identity adolescents maintain on their
conditionally loving parents. This is not to say that Foreclosures are neces-
sarily deficient in identity (as Diffusions are), anymore than anxiously
attached young children lack attachment. It is the quality of the identity
and of the attachment that is less than optimal.
The link between ego identity and the object-relations concept of indi-
viduation lies in the role that the parental introjects play in the adolescent's
secure sense of self and in his or her freedom to question, explore, and
differentiate himself or herself from those introjects. Parenting figures to
whom the individual is securely attached provide the "raw material" for
the introjects that support a secure sense of self. These parental figures,
experienced at the oedipal period by the child both in their reality aspect as
well as through the filter or set of their introjected images, provide the
conditions under which sexuality and sex roles are dealt with. In other
words, the child enters the oedipal period of sex-role socialization with not
only her "reality" parents and her fantasies of them, but also her percep-
tions of them as a result of all of her previous preoedipal experience with
them. She perceives them not as they "are," but at least partially as her
introjected images of them dictate.
Recalling the Schiedel and Marcia (1985) hypothesis that adolescent
exploration may recapitualte oedipal-period exploration, the assumption is
that parents to whom the child has been securely attached provide stable
and differentiated introjects at the oedipal period. These introjects will
allow for anxiety- and guilt-free exploration of sexuality and sex roles. At
adolescence, it is proposed that they will provide internally, in the form of
support and encouragement, those conditions necessary for adolescent
exploration and commitment leading to psychosocial identity formation.
Identity, Cognitive/Moral Development, and Object Relations
Turning to cognitive developmental theory, a link among psychosocial
developmental, object relational, and cognitive developmental theory is
furnished by the work of Irving Sigel (1984). Much of Sigel's work has
concentrated upon the task of helping parents establish conditions that
enhance children's cognitive development, primarily the ability to "dis-
tance." By "distancing," Sigel means the ability to detach oneself from the
concrete object (apple in a tree) as the sole representative of a category
(apples), so that other experiences (picture of apple in book, printed word
"apple") come to have functional equivalence. Of course, what this in-
volves is an interiorization of the world by the child. Paying attention only
to externals, the three varieties of apple noted above are very different
experiences. But if the child can come to "distance" himself from the con-
crete experience by internalizing the concept of apple, then the three sepa-
rate experiences become functionally equivalent. This progressive in-
teriorization of the world is part of the general process of cognitive de-
velopment described by Piaget and Inhelder (1985) and results, somewhat
paradoxically, in both an increasing independence from the external world
and more efficient functioning within the external world. To anticipate a
future point, the development of individuation and the formation of an
identity are also the results of a process of interiorization.
What are the parental strategies, then, that maximize children's distanc-
ing capabilities? Sigel has demonstrated that parental question-asking and
question-sanctioning, as opposed to question-answering, techniques are
most effective. He makes it clear that it is not a strategy of parental "per-
missiveness" (in Baumrind's [1967] sense), but one of active parental in-
volvement and attention to the child's epistemological and metaphysical
216 James E. Marcia
efforts. It is in these attempts to modify parental behavior that concepts

derived from a psychotherapy-based theory such as object relations can
inform more experimentally based cognitive developmental theory. Some
parents take to promoting distancing more easily than others. Clearly,
some parents find it easier to respect their children's cognitive exploratory
processes than do other parents. Parents, after all, operate on the basis of
their introjects, and, hence, their own parents' prescriptions for parenting.
They look to their own parents, both internally as introjects and externally
as models, for approval of themselves in their own parenting styles. The
extent to which they do this is probably a function of their identity status
(e.g., Achievements would be expected to be less introject-bound than
Foreclosures) .
Turning back to identity, we have some evidence for a relationship
between parenting styles, construed within a cognitive development
framework, and identity development. Caroline Newberger (1980) has
developed a measure of parental child-rearing ideology based upon a
Piagetian-Kohlbergian model. Three levels of ideology or philosophy
are discriminated that can be described roughly as self-centered, role-
centered, and mutuality. Mothers at lower levels were found to be more
likely to be abusive to their children. Partol (1980), expanding upon
Newberger's work, found that mothers high in identity were more likely to
be high in child-rearing ideology levels. Although it remains to be demon-
strated empirically, it is likely that parents at advanced levels of child-
rearing ideology also will be more able to establish those optimal condi-
tions for "distancing" described by Sigel.
SUMMARY
Although the foregoing discussion has wandered a rather discursive path

through theoretical concepts and empirical findings, the general sense of
direction that ought to emerge is that the processes, goals, and optimal
child-rearing conditions related to identity formation, cognitive/moral de-
velopment, and individuation are similar. Perhaps one reason for the lack
of previous acknowledgment of these interconnections is that the concepts
being discussed here refer to different points in the life cycle. Object rela-
tions tends to focus on early childhood; cognitive/moral development has
been most contributive to childhood and early adolescence; and identity is
a concept most relevent to middle and late adolescence. In any case, if this
proposed similarity is valid, then, aside from focusing on different prob-
lems and different age groups, the three theoretical streams may have more
in common than has been immediately apparent. The next section of this
chapter deals specifically with the commonalities among these three
theoretical approaches, using identity as the focal integrative concept.
Goals
EGO IDENTITY
The psychosocial developmental sequence preceding identity in late
adolescence represents a progressive movement away from the position of
dependence upon others for gratifications and direction, and a movement
toward self-obtained satisfactions and self-direction. The watershed stage
of late adolescence marks both the end of childhood dependence and the
beginning of adult responsibility. The singular achievement of late adoles-
cence, the formation of an identity, is accomplished via a synthesis of pre-
vious childhood identifications, so that an individual maintains a continuity
with his or her past, a meaning for the present, and a direction for the
future. In the most optimal form of identity development, Identity
Achievement, the individual's identity is more or less available to con-
sciousness and is self-formed. This is in contrast to the less-than-optimal
identity resolution. Foreclosure, wherein an identity, is conferred and the
individual is left only with. the task of becoming aware of the conferred
elements. In both cases, whether identity is constructed (Achievement) or
conferred (Foreclosure), it is seen as an inner organization of the indi-
vidual's needs, abilities, values, personal history, and plans. This inner
organization, or structure, enables the individual to recast the world in her
or his own terms and to endow aspects of that world with personal mean-
ing. The alternative to this internal, meaning-conferring structure is the
Diffusion individual's vulnerability to the vicissitudes of external pressure,
a sense of inner emptiness, and directionlessness. In short, the formation
of an identity structure represents an advance in interiorization that en-
ables more effective, efficient, self-directed action in the external world.
The individual now possesses, inside and organized, in the form of an
identity, what previously existed outside as conflicting demands for occu-
pational directions, ideological loyalties, and sexual-interpersonal stands.
By becoming more internal, the identity-achieved person is able to func-
tion externally in a more efficient and directed fashion.
COGNITIVE/MoRAL DEVELOPMENT
The goal of both cognitive and moral development is also a progressive
interiorization. The phases of cognitive development, as described by
Piaget, begin with an individual's almost total dependence upon the exter-
nal world (in the sensorimotor stage) and end with the individual's ability
(in formal operations) to recreate the external world, internally, and to go
on to create alternative external worlds internally. The stages of develop-
ment of moral thought, as described by Kohlberg, involve a hierarchy of
individually constructed rationales for the resolution of moral dilemmas.
218 James E. Marcia
The lowest level of reasoning relies upon environmentally induced pain

or pleasure; the highest level consists of a personally constructed value
system, superceding given laws, and "universally" applicable. Moral thought
based upon the principle of care (Gilligan, 1977) similarly involves a move-
ment from reliance upon immediate environmental consequences to the
individual's construction of a rationale that involves concern for both self
and others. Hence, in the development of cognition and in both justice-
and care-oriented moral thought, there is an assumed progression from
dependence upon immediate environmental consequences and on the
world of concrete objects and rules given by external figures to internal
self-constructed conceptions ofthe world and of values. Both epistemology
(and, hence, metaphysics) as well as ethics, become issues of individual
construction and not environmental givens.
OBJECT RELATIONS
According to some object-relations theorists l (Bowlby, 1969; Mahler et
aI., 1975), an infant's first form of relationship is either symbiosis (em-
beddedness) or attachment. In either case, the position of the child vis-a-
vis the mother2 is so intertwined that they practically constitute a single
organism. In one of the several paradoxes of psychological development,
the more securely this attachment is effected, the more successfully dif-
ferentiation proceeds. The movement away from secure attachment (called
separation, practicing, exploration) occurs as the young child leaves the
proximity of the mother and engages the world of toys and strangers with-
out seeking her constant attention. This foray is marked by periodic re-
turns to "home base" (rapprochement efforts) for emotional refueling.
This back-and-forth movement from the mother has two consequences: the
mother becomes internalized by the child as an introject (usually with the
help of a transitional object), so that the child comes to hold the mother
inside, rather than needing her constant external presence; and, with this
internalization and the increased opportunities for exploration that it en-
ables, the child begins to develop a sense of an autonomous self (individua-
tion). Greenberg and Mitchell (1983) put it nicely:
Within a relatively short time, the infant becomes a child with a unique personality.
He is an individual living in a world that, within limits, he has already structured in
ways that make it comprehensible to him. He experiences in his own way; he reacts
1 The work of Mahler and Bowlby, as well as the research of Ainsworth et al. (1978)
and Hansburg (1980), will be blended here for the sake of argument. This is not
intended to blur differences, but, by focusing on similarities, to make a point-
without, I think, unduly distorting the different theories.
2"Mother," here, refers neither to the birth mother nor to parent gender, but to
the primary person who fulfills the mothering function.
10. Ego Identity, CognitivelMoral Development, and Individuation 219
in his own way; he acts in his own way. He has, in short, become a person ... The
task of the psycho-analytic developmental theorist is to chart [this] path from form-
lessness to form (p. 270)
In terms of the general issues being discussed here, the child emerges from
total embedded dependence upon the environment via interiorization of
objects (introjects) into a position of autonomous interdependence.
SUMMARY
Identity, formal operations, principled moral thought, and individuation,

positive end-states proposed by psychosocial developmental, cognitive/
moral developmental, and object relational theories have in common a
general movement away from environmental dependence and a movement
toward interiorization of external reality that both frees the individual from
the necessity for direct contact with that reality and, at the same time,
allows her or him to deal more effectively with it.
Processes
EGO IDENTITY
Goals are probably the least important of the similarities among the three
theoretical streams under discussion. The psychological processes assumed
to account for change and development are the most important. It was
stated previously that one possible factor accounting for the productivity
and longevity of the identity status paradigm lies not in its use as a develop-
mental stage outcome measure but in the generalizability of its process
criteria of exploration and commitment. In terms of personality develop-
ment, it is assumed that the physiological, cognitive, and social changes
constituting adolescence provide a naturally occurring disequilibrating
event (an identity "crisis") that can be used by the individual as a catalyst
for leaving childhood occupational, ideological, and sexual-interpersonal
positions and exploring new ones. As has been discussed elsewhere
(Marcia, 1983), not all adolescents use (or are socially enabled to use)
this period in the service of change and integration. Informal social con-
straints (e.g., economic demands), formal social constraints (e.g., rites of
passage), and intrapsychic constraints (e.g., strict superego sanctions) may
preclude genuine exploration. In this case, an identity is said to be fore-
closed rather than achieved, and that identity is assumed to be less flexible,
more brittlely fragile, and in greater need of content-consistent social sup-
port.
Exploration, as a defining variable in identity status determination,
refers to cognitive and behavioral investigation and testing of alternative
occupations, beliefs, and styles of self-presentation and relationships. It is
220 James E. Marcia
assumed that a structure exists from which to depart in order to begin

exploration. In the case of no preexisting structure, the individual starts
from a point of diffusion and is more likely than not to be unable to main-
tain the tension of genuine exploration or to forego gratification, as one
must in order to make commitments. The preexisting structure is the
agglomeration of part-identifications of which Erikson (1959) speaks. For
most individuals, exploration ends when an internal integration has been
accomplished (via ego synthesis) of conflicting needs and abilities, and ex-
ternal social forms have been found that are compatible with that internal
integration. It is assumed that the integrative process occurs concurrently
with the exploratory process and that the subsequent commitments reflect
an outcome of this exploratory-integrative process. Those external com-
mitments, then, are taken to be reflections of an internal organization-an
identity. Since one aspect of identity is its similarity to theory-a theory
one holds about oneself-the identity structure, especially if it is self-
constructed and not parentally conferred, suffers the same fate as does a
testable theory. It is eventually disconfirmed, in part, and undergoes revi-
sion. (This process of self-theory revision has been described in detail by
George Kelly (1955) in his work on personal construct theory.)
Revision or reformulation of an identity structure proceeds via the
same "adolescent" procedure of exploration and commitment. And the
reformulated structure subsequently undergoes the same revisional fate as
did the inital structure. Hence, "adolescence" is lifted from its position as a
label for a life cycle era and becomes a metaphor for an identity reformula-
tion phase, a number of which can be expected across one's life cycle. At
such moments of change, one not only appears to others and to oneself as
"adolescent" , one is, in a metaphorical sense, an "adolescent."
The essential point here, with respect to identity formation and refor-
mulation, is that the process involved is one in which an existing structure
is called into question, an exploratory period undergone during which an
internal integrative process is constantly active, and subsequent commit-
ments are made that both reflect and contribute to a new identity structure.
This new structure has a continuity with the preexisting one and furnishes
some of the elements for its eventual successor.
COGNITIVE/MORAL DEVELOPMENT
Equally, and perhaps more, important than the description of specific

stages of cognitive development, is Piaget's specification of the means by
which the individual moves from one stage to the other. In general, this
process also applies to changes in levels of moral thought. The individual is
assumed to possess certain structures that process and organize experience.
These structures begin as primarily physical operations on external reality
(sensorimotor schemata) and eventually develop into primarily mental
operations on internal and external reality (formal operational concepts).
The "structures" are not actually "ideas" or "concepts," rather they are
more like mechanisms for processing experience that become progressively
more sophisticated and efficient as they become capable of converting
actions into ideas.
Cognitive structural change involves three aspects: assimilation, dis-
equilibration, and accommodation. Assimilation refers to the processing of
information by means of the existing structure or operation. Disequilibra-
tion occurs when there is a noticed discrepancy between the experience
data and the individual's means of organizing or processing those data. In
other words, there is information that is not assimilable or a problem that is
not solvable, given the cognitive structure available. In a fairly neutral
affective situation, this disequilibratory experience leads to a change either
vertically or horizontally in the current operating structure-an accom-
modation. A cognitive structure, having undergone accommodation, can
now handle data (solve problems) that it was previously incapable of
doing. The consequences of this, of course, are that the experiences de-
fined as "data" and the data perceived as "problems" also change. Chang-
ing a cognitive structure means changing the experiential world.
Change in levels of moral thought depend upon cognitive development,
so that advancement in the latter is necessary, but not sufficient, for
advancement in the former. It is assumed that developmental changes of
level of moral thought are initiated by individuals' experiences of moral
dilemmas that exceed their existing reasoning capacities by some optimal
extent-not too little to be assimilable, not too much to be insolvable.
To return to the issues under comparison, changes in both cognitive and
moral development involve the existence of a structure that is experienced
as inadequate to handle a problem and a subsequent change in that struc-
ture that allows for successfully resolving both the current problem and
subsequent ones similar to it. There is a general movement from a
"secure" (assimilative) position to an "insecure" (disequilibrative) one,
and a subsequent structural reformulation (accommodative) that restores
some sense of "security" or equilibrium.
OBJECT RELATIONS
The object relational approach taken by Mahler and Bowlby, in general,
views the infant as beginning in a state of nonrelatedness that soon gives
way to attachment or symbiosis. Following a period of secure attachment,
differentiation occurs, culminating in "psychological birth" (i.e., the child
can walk and walk away). Walking away toward other persons and other
persons and other more interesting places and things has the effect of
separating the child from the parent. This separation, however, makes evi-
dent to the child his or her smallness in a large world, and the parent is
again sought out as a guide, protector, and emotional refueling station.
This alternation of separation and rapprochement leads to the child's grow-
222 James E. Marcia
ing sense of self as more of the world is autonomously explored within the
context of a dependable attachment. Eventually, the parent, with the help
of a totemic representative, the transitional object, becomes internalized
(introjected) and can be referred to by the child in the absence of the
parent in external reality. Ultimately, then, the child becomes individuated
(i.e., a unique, autonomous, interdependent person) as a function of the
quality of the introjects. Again, paradoxically, one becomes independent
as a result of initially successful dependency upon parental figures, and a
result of subsequently successful dependency upon their internal repre-
sentatives.
As in the case of identity development and cognitive/moral develop-
ment, individuation follows a course of moving from a stable, secure
position to an exploratory phase, to a resolution characterized by an in-
teriorization of elements that were once external. And this interiorization,
again, enables the individual to function more efficiently externally.
Whether or not one can, one need not go home again, having internalized
"home." As with identity and cognitive/moral development, the interplay
of attachment, exploration, and individuation, although most crucial in the
initial early-childhood phase, continues to operate later in the life cycle.
Child-Rearing Implications
Rather than following the format of the preceding two sections in which
the three theories under consideration are discussed separately, this sec-
tion considers all three theories together. To do otherwise would be repeti-
tious, because the implications for parental styles are the same from all
three perspectives. Identity formation is facilitated by parents who are
comfortable with their adolescents' attachment and dependency needs and
with their efforts at separation and exploration. Cognitive and moral
development are enhanced by parents who can maintain their own view
of reality and their own value positions, thus offering a well-defined con-
text, which, because of its stability, can afford the flexibility of permitting,
sanctioning, and even encouraging children's questioning (Sigel, 1984),
doubting (Chandler, in press), and moral uncertainty (Gilligan, 1977,
Kohlberg, 1973). Individuation proceeds from a parental environment that
is rhythmically synchronized with a child's early needs, yet capable of dis-
engagement in the face of the child's efforts at separation.
It has been proposed that identity, at the late-adolescent phase of the
life cycle, might serve as an integrative concept in viewing psychosocial
developmental, cognitive/moral developmental, and object-relations
theory. At an earlier place in the life cycle, Irving Sigel's concept of "dis-
tancing" strategies, and especially his colleagues' and his efforts at teaching
parents these strategies, provide an integrative view of techniques that can
enhance development on both cognitive and interpersonal fronts. In order
to emphasize this integrative function, one simply has to ask the question,
"How must parents view their children in order to promote distancing?"
The answer intended to this question goes beyond (and beneath) the
specific techniques noted by Sigel, McGillicuddy-Delisi, and Johnson
(1980). Surely those parents functioning at higher levels of child-rearing
ideology (Newberger, 1980) would be more likely not only to tolerate, but
to welcome their children's questions about the world and about values.
And it is equally likely that egocentric or rule-bound parents might be
somewhat intolerant of children's attachment needs, more intolerant of
their separation and exploration needs, and certainly unaccepting of alter-
nating attachment-exploration cycles.
SUMMARY
The formation of an ego identity is hypothesized to be completed for the
first time at late adolescence, and, if optimally formed, is expected to
undergo reformulation throughout the rest of the life cycle. The processes
involved in identity formation and reformulation are exploration and com-
mitment. Research on the identity statuses, a paradigm for investigating
ego identity development, has shown relationships to exist between identi-
ty, an ego psychoanalytic concept, and constructs from cognitive/moral de-
velopmental theory, and between identity and constructs from object-
relations theory. It is proposed that identity development, cognitive/moral
development, and object relations share similar goals and underlying pro-
cesses, and have similar implications for child-rearing procedures. In
general, the common goals involve a progressively organized interioriza-
tion of the external world that allows for more efficient and effective func-
tioning. The common processes are characterized by a movement from
security and structure, through a period of exploration and loosening of
structure, to a position of restructuring that resolves the disequilibrating
issues on a higher level than was previously possible. The common child-
rearing implications are that development in all three spheres is facilitated
by a parental environment that furnishes a sense of security, respect for
differences, and encouragement of questioning and exploration.
REFERENCES
Adams, G.R., Ryan, J.H., Hoffman, J.J., Dobson, W.R., & Nielsen, E.C. (1985).
Ego identity status, conformity behavior, and personality in late adolescence.
Journal of Personality and Social Psychology, 47, 1091-1104.
Ainsworth, M.D.S., Blehar, M., Waters, E., & Wall, E. (1978). Patterns of
attachment. Hillsdale, NJ: Erlbaum.
Baumrind, D. (1967). Child care practices anteceding three patterns of preschool
behavior. Genetic Psychology Monographs, 75, 43-88.
Bern, S.R. (1977). On the utility of alternative procedures for assessing psycho-
224 James E. Marcia
logical androgeny. Journal of Consulting and Clinical Psychology, 45, 196-205.

Bourne, E. (1978a). The state of research on ego identity: A review and appraisal:
Part I. Journal of Youth and Adolescence, 7, 223-251.
Bourne, E. (1978b). The state of research on ego identity: A review and appraisal:
Part II. Journal of Youth and Adolescence, 7, 371-392.
Bowlby, J. (1969). Attachment and loss: l. Attachment. New York: Basic Books.
Chandler, Michael. (in press). The Othello effect: An essay on the emergence and
eclipse of skeptical doubt. Human Development.
Cooper, C.R., Grotevant, H.D., & Condin, S.M. (1983). Individuality and con-
nectedness in the family as a context for adolescent identity formation and role-
taking skill. In New Directions in & Child Development (pp. 43-59) H.D.
Grotevant & C.R. Cooper (Eds.), Adolescent development in the family: San
Francisco: Jossey-Bass.
Erikson, E.H. (1959). Identity and the life cycle. Psychological Issues, Monograph
No. 1.
Gilligan, C. (1977). In a different voice: Women's conception of the self and of
morality. Harvard Educational Review, 47,481-517.
Greenberg, J.R. & Mitchell, S.A. (1983). Object relations in psychoanalytic theory.
Grotevant, H.D. & Cooper, C.R. (1985). Patterns of interaction in family rela-
tionships and the development of identity exploration in adolescence. Child
Development, 56,415-428.
Hansburg, H.G. (1980). Adolescent separation anxiety: A method for the study of
adolescent separation problems (Vol. I). New York: Robert E. Krieger Co.
Hult, R.E. (1979). The relationship between ego identity status and moral reason-
ing in university women. The Journal of Psychology, 103, 203-207.
Kelly, G. (1955). The psychology of personal constructs (2 vols.). New York:
Norton.
Kernberg, D.F. (1975). Borderline conditions and pathological narcissism. New
York: Aronson.
Kohlberg, L. (1973). Continuities in childhood and adult moral development revi-
sited. In P.B. Baltes & K.W. Schaie (Eds.), Life-span developmental psychology:
Personality and socialization. New York: Academic Press.
Kroger, J. (1985). Separation-individuation and ego identity status in New Zealand
university students. Journal of Youth and Adolescence, 14, 133-147.
Leiper, R.N. (1981). The relationship of cognitive developmental structures to the
formation of identity in young men. Unpublished doctoral dissertation, Simon
Fraser University, Burnaby, British Columbia, Canada.
Levitz-Jones, E.M. & Orlofsky, J.L. (in press). Separation-individuation and in-
timacy capacity in college women. Journal of Personality and Social Psychology.
Mahler, M.S., Pine, F., & Bergman, A. (1975). The psychological birth of the
human infant. New York: Basic Books.
Marcia, J.E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of
adolescent psychology. New York: Wiley, pp. 159-187.
Marcia, J.E. (1983). Some directions for the investigation of ego identity develop-
ment in early adolescence. Journal of Early Adolescence, 3, 215-223.
Marcia, J.E., Waterman, A.S., & Matteson, D.R. (in preparation). Studies in ego
identity and intimacy: A handbook for psychosocial research.
Newberger, C.M. (1980). The cognitive structure of parenthood: Designing a de-
scriptive measure. In New directions in child development (Vol. 7, pp. 45-67).

San Francisco: Jossey-Bass.
Partol, S. (1980). The correlates of parental awareness. Unpublished doctoral dis-
sertation, Graduate School of Education, Boston University.
Piaget, J., & Inhelder, B. (1958). The growth of logical thinking from childhood to
adolescence. New York: Basic Books.
Podd, M.H. (1972). Ego identity status and morality: The relationship between two
developmental constructs. Developmental Psychology, 6, 497-507.
Poppen, P.J. (1974). The development of sex differences in moral judgment for col-
lege males and females. Unpublished doctoral dissertaion, Cornell University.
Rowe, I., & Marcia, J.E. (1980). Ego identity status, formal operations, and moral
development. Journal of Youth and Adolescence, 9, 87-99.
Schiedel, D.G., & Marcia, J.E. (1985). Ego identity, intimacy, sex role orienta-
tion, and gender. Developmental Psychology, 18, 149-160.
Sigel, I.E. (1984). Reflections on action theory and distancing theory. Human De-
velopment, 27, 188-193.
Sigel, I.E. McGillicuddy-Delisi, A.V., & Johnson, J.E. (1980). Parental distanc-
ing, beliefs, and children's representational competence within the family context.
ETS RR-80-21 (Educational Testing Service, Princeton, New Jersey).
Skoe, E. (1986). The development and partial validation of a care-based measure of
moral development. Unpublished doctoral dissertation, Simon Fraser University,
Burnaby, British Columbia, Canada.
Slugoski, B.R., Marcia, J.E., & Koopman, R.F. (1984). Cognitive and social in-
teractional characteristics of ego identity statuses in college males. Journal of
Waterman, A.S. (1982). Identity development from adolescence to adulthood: An
extension of theory and a review of research. Developmental Psychology, 18,
342-358.
11
Identity and the Development
of the Self
AUGUSTO BLASI
Over 40 years ago, Erikson (1946, 1950) introduced in psychology the con-
cept of identity. This concept must have resonated with the psychologists'
intuition or experience. In fact, identity was soon accepted as representing
a very important step in normal human development and gave origin to a
long list of empirical studies, even though Erikson's overall theory of per-
sonality development was frequently left aside and, at times, even dismissed.
Quite soon, however, there were complaints that Erikson's language was
imprecise and frequently metaphorical, that his discussion of the concept
was fragmented and not always coherent, and that, therefore, it was very
difficult to translate the concept of identity into empirical operations.
These problems should not be minimized. The major difficulty, however,
may lie in the intrinsic complexity of "identity" as Erikson understands it
(see, in particular, Bourne, 1978; Erikson, 1950, 1959, 1964, 1968a, 1968b;
Stewart, 1984). In it, in fact, one can abstract the following elements, con-
nected to each other by a network of dialectical relations:
1. Identity is an answer to the question, "Who am I?"
2. In general, the answer consists of achieving a new unity among the
elements of one's past and the expectations about one's future
3. giving origin to a fundamental sense of sameness and continuity.
4. The answer to the identity question is arrived at by realistically
appraising oneself and one's past;
5. By considering one's culture, particularly its ideology, and the ex-
pectations that society has for oneself,
6. while at the same time questioning the validity of both culture and
society and the appropriateness of others' perceptions and expecta-
tions (crisis).
7. This process of integration and questioning should occur around cer-
tain fundmental areas, such as one's future occupation, sexuality, and
religious and political ideas,
8. and should lead to a flexible but durable commitment in these areas;
9. such that it guarantees, from an objective perspective, one's produc-
tive integration in society,
11. Identity and the Development ofthe Self 227
10. and, subjectively, a sense of basic loyalty and fidelity,

11. as well as deep, preconscious, feelings of rootedness and well-being,
self-esteem, and purposefulness.
12. The sensitive period for the development of identity are the adolescent
years.
Erikson might have looked at these characteristics, perhaps with the ex-
ception of the last one, as together establishing a unified concept and,
therefore, as being all needed to determine whether or not a certain
psychological phenomenon should count as identity. However, one may
not wish to follow Erikson on this point. One may want to distinguish those
statements that aim at describing identity from those that concern the fac-
tors and the processes that are necessary for its development and from
those dealing with its functional consequences. The description of identity
as a conscious and preconscious experience of oneself as a differentiated
individual, enjoying a sense of fundamental unity, seems to be conceptual-
ly independent of, and open to, other questions (e.g., whether there are
specific domains of human experience around which identity must neces-
sarily be established; what and how many these domains are).
Even when Erikson's "identity" is restricted to its descriptive features, it
still remains a very complex concept. Erikson (1968a) himself was openly
skeptical about the possibility of capturing it in psychological research.
In psychologists' attempts to operationalize interesting but almost unman-
ageable ideas (Erikson's "identity" is certainly one), there frequently is
the tendency to resort to simplifying strategies. A complex theoretical
framework may be translated into the framework of observations and self-
reports; only some of the elements may be selected, those that are thought
to be more important and that can be handled by traditional assessment
methods. These elements may then be treated as separate dimensions
rather than as components of a unitary phenomenon. Finally, conceptual
judgment may be substituted with numerical calculus. As a result, rating
scales are constructed and algorithms are formulated in order to combine
more or less inconsistent information.
This is not the appropriate context for a detailed discussion of the vari-
ous measures of identity. In general, the most frequently used measures
(for a list, see Bourne, 1978; Marcia, 1980; Waterman, 1982) share similar
drawbacks. They focus on the adolescent's attitudes and decisions concern-
ing such specific areas as occupation, political ideology, and religion, to
which gender roles, interpersonal relationships, and life-style have recently
been added. In these measures, then, the processes of developing one's
identity seem to be assimilated to the process of making important deci-
sions. Following Marcia's (1966) operationalizations of Erikson's defini-
tion, adolescents' and young adults' attitudes are typically explored in
terms of the categories of crisis and commitment; on the basis of these
processes, responses are classified according to four identity statuses (for
reviews, see Marcia [1980J and Waterman [1982]). However, these pro-
228 Augusto Blasi
cesses tend to be distorted and trivialized, particularly by the objective

assessment procedures. Crisis tends to lose the Eriksonian meaning of
questioning the perceptions and expectations of one's culture and of de-
veloping an autonomous position with regard to one's society, and is prac-
tically interpreted as having thought about certain issues and having con-
sidered alternatives. Commitment loses the meaning of investing one's
own self in certain lifelong projects, and is simply interpreted as having
made a firm decision.
Briefly, what appears to be missing in the standard identity measures is
the basic identity question, "Who am I?", and the specific role that crisis
and commitment around certain domains play for each person in answering
that question. In other words, what is overlooked is the person's relation to
himself (rather than to his occupational choices, political ideas, or marital
roles) and his experience of himself as an individual and unified entity.
A concrete example of how contemporary approaches to the study of
identity have lost the special sense of self that identity was supposed to
convey is provided by the transformation of one central characteristic, the
capacity of identity to establish a deep sense of unity in the person's
perception and experience of himself or herself. While Erikson believed
that identity should be worked through around one's future career and
one's relation to political action and religious beliefs, he never thought that
one's preoccupation with a multiplicity of issues would be at the expense of
the sense of unity. There is no identity if the sense of self becomes frag-
mented among the many domains. Instead, this is precisely the common
outcome of the measures of identity: inevitably, an individual will appear
to have various degrees of crisis and commitment from issue to issue, and
even from item to item representing the same issue. One's identity, then,
becomes a profile of scores or a summary score, obtained by averaging the
many ratings or by some other algorithm.
The central argument of this paper is that the meaning of Erikson's con-
cept can be recovered by taking the perspective of the self as subject,
namely, of that characteristic mode that each person has of experiencing
himself of herself. In the following section the concept of self as subject is
discussed. A second section then describes an attempt to apply this concept
to Loevinger's ego development data and outlines three different identity
modes that resulted from this attempt.
The Self-as-Subject
THE EXPERIENCE OF SELF
In every intentional action that we perform, in fact in every experience that
we undergo, we experience ourselves, in the process of acting and ex-
periencing, as related to our actions and experiences. What in this paper is
11. Identity and the Development of the Self 229
called self-as-subject or simply self refers to such an immediate, though

indirect and "unintended," experience (for a more detailed discussion, see
Blasi, 1983, in press). Without this relation of ourselves to our actions,
embedded in each action and experience, there would be neither action
nor experience, but only processes impersonally occurring within the
organism.
Moreover, there are different ways of relating to our actions and experi-
ences. For example, some may experience their sexuality as something that
happens to them and that is passively suffered; others may experience it as
something that they have mastered and can control; others, finally, may
relate to their sexual self as something to be chosen, decided, actively con-
structed and reconstructed. Similarly, while different people may have
what seem to be the same opinions, they may relate to them in very dif-
ferent ways: to some, opinions are like ready-made "objects," heard and
given back, but not really appropriated; to others, opinions are carefully
considered and chosen and are felt, in the very moment in which they are
expressed, as being unquestionably their own, perhaps intrinsically con-
nected with the sense of who they are.
In analyzing what constitutes the self-as-subject, it is possible to concep-
tually isolate and describe four dimensions that are present in all experi-
ences and characterize the very subjectivity of experience. The first dimen-
sion is agency and refers to the degree to which an action is unreflectively
grasped as one's own and oneself is grasped as its source. The second,
identity with one~elf, refers to that immediate realization of oneself in
action as being identical to the aware, grasping self, a realization that is the
foundation of reflectivity and distancing. The third dimension is the experi-
ence of unity among the various actions that constitute one's present activ-
ity: thinking, selecting words, talking, hearing one's voice, and so forth, all
belong to the same agent, to the experiencing and acting "I". Finally, there
is otherness, the awareness of oneself as separate agent, different, in per-
forming one's action, from other entities, minimally from the object on
which one acts. The self-as-subject, in sum, is understood here as the spe-
cific way by which one experiences oneself as agent, unified in one's agency,
different from other agents, and capable to stand back vis-a-vis oneself in
reflection.
INTENTIONAL ACTION AND THE SELF

This broad description of the self-as-subject is derived from the analysis of
what it means to experience, particularly to experience oneself in inten-
tional action. The self, in fact, is not grasped primarily in contemplation
and in the process of reflecting on oneself; instead, it is grasped indirectly
though immediately in and through every intentional action, having as
direct objects external or even internal events.
It is useful, then, to see what is involved in intentional action and what
230 Augusto Blasi
is the role of the self in it. This analysis seems to be particularly neces-
sary, because of the long tradition in psychology of reducing the self to
self-concepts and of more recent attempts to assimilate the self to cog-
nitive schemata, cognitive structures, or to certain features of computer
programs (e.g., in Greenwald & Pratkanis, 1984; Markus and Sentis,
1982; Pratkanis & Greenwald, 1985). It may help to begin with a set of
global distinctions between three aspects of intentional action: namely,
products, operators, and stances.
Products refer to the intended results of human activity. They can be
external and concrete (e.g., a table, a piece of music, a novel, a smile, a
comforting word). They can also be internal and less tangible, such as
wishes, fantasies, ideas, and theories. Perceptions and concepts, and also
self-perceptions and self-concepts, are products, in the sense used here.
Under operators are included all of the various processes and tools by
which one constructs the products of intentional action. Some of them are
seeing and hearing, comparing and reasoning, classifying, wanting, con-
trolling. Some operators, perhaps most of our cognitive tools, are con-
structed by us; therefore, they can be equally considered as products or as
operators.
An important subcategory of operators includes the cognitive structures
of Piagetian theory and the schemata that are described in the information-
processing approach, namely, those mental tools by which objects are
selected, processed, transformed, and finally understood. For instance, a
child who is attempting to resolve a conservation problem resorts, more or
less consciously, to certain rules of quantity conservation, which, according
to Piaget's theory, are related to even more encompassing operational
structures. The general logical structure and the more specific cognitive
rules are instruments unconsciously used by the child to arrive at a satisfac-
tory answer (e.g., in this case, about the amount of plasticine in two balls).
Some cognitive schemata are used by people to understand themselves and
their social experiences and, therefore, have been called self-schemata
(Markus & Sentis, 1982) or self-theories (Epstein, 1973).
The fact that cognitive schemata, including self-schemata, should be
understood as operators (namely, as instruments to process information
and to acquire knowledge) manifests how inadequate and ambiguous is the
distinction originally proposed by William James (1890/1950) between the
self-as-known and the self-as-knower or self-as-subject. In fact, cognitive
structures and self-schemata are not typically objects of direct knowledge;
if anything, they are on the side of the subject or of the knower. Therefore,
it is possible to conclude that the self-as-subject is simply the structural
organization of self-concepts or the basic cognitive structures that are in-
strumental to the construction and organization of self-concepts (Epstein,
1973; Frondizi, 1971; Kelly, 1955; Sarbin, 1962). For instance, Markus and
Sentis (1982) recently wrote:
From our perspective, the self can be conceptualized as a system of setf-

schemata . ... A schema is a conceptually advantageous analog for the self be-
cause it can potentially represent what James (1915) called the two distinct aspects
of the self-"the self as the knower and the self as that which is known." (p. 45)
However, there are two different ways of understanding "the knower,"
and here lies the ambiguity. "The knower" may refer to the possessor
of knowledge, namely, the subject who attends, compares and organizes
information, constructs concepts, and finally understands and knows.
According to a second, admittedly somewhat inappropriate, meaning, "the
knower" refers to the processes, concepts, and logical organizations-the
computer's program-used by the agent to arrive at knowledge. This
seems to be the meaning implied by Markus and Sentis (1982), when they
write that "self-schemata search for information that is congruent with
them and direct behavior so that it is commensurate and consistent with
them" (p. 45).
One may ask, however, whether schemata search for information, direct
behavior, and try to remain consistent with themselves or, rather, whether
the knower (in the first meaning) uses schemata as tools to achieve these
results. To assimilate schemata or cognitive structures, then, to James's
"knower" does not resolve the question about the nature of the self, but
only results in a serious confusion between two very different categories. In
fact, operators and subjects have very different relations with their prod-
ucts: operators have with their outcomes either a cause-to-effect relation or
a relation of logical implication, whereas the relation of the subject to the
products of his or her activity is intentional in nature. 1
This distinction can be clarified by considering stances, the third aspect
of intentional action. Let us begin concretely with descriptions of inten-
tional actions: "I am sorry I could not finish my paper, but I am sure that I
will send it next week"; "I hope I will graduate next year"; "I did say that,
but I did not mean to insult you"; "I believe that your car is safe, but I am
still afraid to drive it"; "I have been trying to refrain from drinking, and I
am very proud of myself."
1 Another example of the same confusion concerns the unity aspect of the self. When
the self-as-subject is not distinguished from either self-concepts or from ego pro-
cesses or self-schemata, the unity of the self is confused with the organization of
self-concepts or the coherence of self-schemata (Greenwald, 1982). Here, too, the
objective and the subjective dimensions are hopelessly confused; in fact, within the
subjective, there is a confusion between that sense of unity that results from ab-
sence of internal conflict, and the sense of being one subject. The functional orga-
nization of a smoothly operating machine and the coherence of a logical structure
(in the objective dimension) differ from the sense of lack of internal tension, which
differs from the self-consistency of a single agent (Blasi & Oresick, 1985). Conflict
can be subjectively experienced only when there is the experience of being a unit-
ary self.
232 Augusto Blasi
In its philosophical meaning, intentionality is present here on two levels:

in the experience of not being able to finish something, of apologizing,
trying, hoping, refraining, believing, and so forth, that these statements
reflect, as well as in verbally describing these experiences.
Each one of the above statements could be analyzed in two ways,
according to different theoretical biases: one could focus on cognitive-
structural operators and try to describe the specific conceptual organization
that should account for the statement; alternatively, one may prefer func-
tional operators and describe a network of causal relations. Taking, for
instance, the last statement, "I have been trying to refrain from drinking
and I am very proud of myself," a cognitive-structural analysis will even-
tually result in a set of concepts and self-concepts: I, drinking, trying, re-
fraining, pride, and so forth. Then one may outline the semantic organiza-
tions to which these concepts belong, and finally attempt to reconstruct the
basic logical structure and self-schemata that would allow someone to
meaningfully utter that proposition and to believe it as true.
According to the functional analysis, one would speak of a habit of
drinking, of negative unpleasant states produced by it, of inhibitory
mechanisms, and of an emotional state, pride, as the last effect in the
causal chain, reinforcing the inhibitory activity and eventually causing the
extinction of the habit.
However, what the person is saying and experiencing is not conveyed by
either one of these analyses. What both analyses leave out is precisely the
person's stance, namely, his or her relations to the goal of not drinking, to
the operations of trying, and to the feeling of pride; the person's affective
relation to him- or herself, to the statement and the act of saying it in the
very moment that he or she says it.
One cannot bridge the gap between the person's description and the
psychological accounts by adding new elements to the model, cognitive or
functional, or by reshuffling the components of the logical and the causal
networks. These models may be perfectly adequate to do what they are
supposed to do, except that they totally miss the stance aspect, and thus the
intentionality of action. To say that an action is intentional is to say that the
action as a whole expresses the subject's stance with regard to the object.
The subject's stance-his believing, desiring, controlling, or hoping-is
not one component among others, but permeates every aspect of the action
and gives unity to it.
Stances may vary enormously from each other, but they also share two
fundamental characteristics. First, they occur in consciousness. Their
presence in consciousness, however, is not similar to the passive way in
which trees, people, and birds are in our field of consciousness when we
look around us. In fact-and this is the second crucial characteristic-all
stances express, in the subject's consciousness, a relation of agency and
ownership of the subject to the various stances that are expressed in action.
This is why we could not properly say, as a consistent functionalist or a
consistent information-processing theorist would have to say, that affection

exists in the mother and the baby; that yesterday a bit of knowledge was
elicited in Charles; and that regret just occurred in me. Nevertheless we
could say that there is similarity between the mother and the child; that a
state of anxiety flooded Charles's mind; and that a feeling of sudden fatigue
came over me. Affection, knowing, and regretting express a person's spe-
cific stances toward certain objects; similarity, anxiety, and fatigue do not.
In sum, the intentionality of our actions is expressed by the stance that
we as subjects take toward the objects of the action, or by the specific way
we relate to each one of these objects. There are very many possible
stances, all qualitatively different from each other. We cannot confuse the
experiences of loving, loathing, desiring, searching, inhibiting, or making.
But, as varied as they are, these experiences have something in common,
namely, what could be called a stance-like characteristic. If one attempts to
analyze what is this central characteristic of stances, one will find again the
four basic dimensions of the self-as-subject; agency, unreflected identity
with oneself, unity, and otherness. All stances manifest the indirect, unin-
tended awareness of the subjective self as the agentic source and owner of
intentional actions. To be a self, in the sense adopted in this paper, means
to be able to relate to the objects of one's actions in a stance-like manner,
namely as a distinct, whole agent and possessor of one's actions.
IDENTITY AND THE SELF
These four dimensions of subjectivity order a large variety of experiences

and become headings for a large number of psychological processes and
characteristics. Self-analysis, self-appropriation, self-understanding, free-
dom, self-control, self-mastery, self-change, ownership, responsibility,
commitment, self-consistency, integrity, internal compartmentalization,
pride, defiance, search for uniqueness, and autonomy, for example, seem
to outline a potentially enormous domain of psychological inquiry.
Identity, as described by Erikson, is one such construct and should be
understood from the perspective of the self-as-subject. Identity corre-
sponds to a special mode (or some special modes) of experiencing oneself-
as-subject. In fact, the realistic appraisal of oneself in asking the identity
question corresponds to a certain mode of reflectiveness and distancing
from oneself. The attempt to integrate the various components of one's
personality and to find a principle of order whereby past, present, and
future form a coherent biography defines a mode of unity in experiencing
oneself. Aspects of this unity are also fidelity (the characteristic virtue of
identity, according to Erikson) and the necessity to be consistent with one's
ideals. The attempt to question one's culture and the previously accepted
ideology and to distance oneself from others' definitions and expectations
defines a special kind of otherness and of experienced individuality. Final-
ly, the feelings of being in charge of constructing one's identity, of being
234 Augusto Blasi
responsible for who one is, and of being committed to one's choices, are all
part of a specific sense of agency. Even the sense of loyalty for the tradi-
tions of one's culture and the attempt to respond to the expectations
of one's society by finding in it one's own niche are not passive responses,
but reflect the general commitment and responsibility that characterize a
properly developing identity.
It should be added that the perspective of the self-as-subject, as central
and as necessary as it is for its understanding, cannot exhaust the study of
identity. In fact, this approach leaves out those concrete aspects-specific
competencies, preoccupations, ideals, and so forth-that each person
selects as the center of his or her world and around which he or she con-
structs the sense of a unified self. A full understanding of identity requires
the study of two components: the set of concrete contents (perhaps a spe-
cific focus on religion, or on one's career, or on political ideology) and the
set of attitudes (of choice and commitment, of integration, individuation,
etc.) by which the concrete contents become the substance of one's subjec-
tive identity. Two different people may have an equally strong sense of
being a unified self; however, for the first, the sense of wholeness may be
built around music as a career and an aesthetic orientation, while religion
may be the center of the second person's sense of integration. The attitu-
dinal component, namely the experience of oneself as an individual whole,
is what makes a certain psychological phenomenon to be identity, or the
identity of a certain developmental variety; the content component,
instead, accounts for the fact that different people, all having arrived at a
subjective sense of identity, dramatically differ from each other on the
basis of their identities.
Patterns of Identity in Loevinger's Ego Stages
In the previous section, a concept of self-as-subject is presented as being

irreducible to both self-concept and self-schemata. It is argued that the
self-as-subject (namely, the experience of oneself as a unified, self-
reflecting agent) constitutes the proper perspective to understand the
Eriksonian concept of identity.
Of course, an important question that psychologists must ask about
theoretical frameworks and concepts concerns their empirical viability and
fruitfulness. An attempt was made, therefore, to test, and perhaps demon-
strate, the possibility of empirically capturing differences in the ways the self
is experienced by different people. Loevinger's ego development catego-
ries as reported in the scoring manuals (Loevinger, Wessler, & Redmore,
1970; Redmore, Loevinger, & Tamashiro, 1978) served as the basic mate-
rial for this exercise (Blasi, 1983). Eight of Loevinger's 36 sentence com-
pletion items were selected as frequently yielding responses that are
relevant to the self-as-subject. 2 For each item and for the eight items as a
whole, the response categories that are assigned to the same ego stage were
analyzed together and reinterpreted, by essentially asking two questions:
What kind of statements are typically given by people who are classified at
different stages? What kind of experiences of self do these statements
suggest? In other words, in which specific ways do people at different ego
stages experience themselves as agents, as wholes, as different from others,
and as distancing from themselves in their self-reflection?
In sum, Loevinger's ego development stages were used as rudimentary
guidelines for discovering different patterns of self-experience. Loevinger's
scoring manuals allow this type of analysis, because the coding categories
represent a rather narrow semantic field, are very close to the level of
meaning intended by the original responses, and are accompanied by a
number of examples reflecting the modal tone of the statements as well as
the main variations.
Following this procedure, each stage was redescribed in order to bring
out the characteristics of distancing and reflectivity, of agency, control,
and mastery, of separateness and individuality, and of unity and self-
appropriation, that seem to constitute, at that stage, the specific quality of
experiencing oneself in action (see Blasi, 1983).
To each of the last three stages (in Loevinger's terminology, Con-
scientious-Conformist or 1-3/4, Conscientious or 1-4, and Autonomous
or 1-5) there seems to correspond a specific mode of constructing and living
one's identity.3 Thus, if one follows this approach and emphasizes the self-
as-subject, one finds not one but three kinds of psychological identity,
which will be called here Identity Observed, The Management of Identity,
and Identity as Authenticity.
IDENTITY OBSERVED
From the perspective of the self-as-subject, Loevinger's Conscientious-
Conformist stage seems to mark the birth of the reflected self and the be-
2These eight items can be grouped in four pairs, representing four important
themes of the subjective self: "The thing I like about myself is ... " and "I am ... "
can be considered as self-concept items; "When I get mad ... " and "When I am
nervous ... " elicit concern about self-control; "What gets me into trouble ... "
and "My main problem is ... " relate to self-criticism; "When they avoided
me. . ." and "When I am criticized. . ." bring up preoccupations about social
rejection and self-protection.
3In Loevinger's (Loevinger, 1976; Loevinger, Wessler, & Redmore, 1970) se-
quence, the Conscientious-Conformist or 1-3/4 stage, is considered a transitional
stage; additionally, there is a later transitional stage, 1-4/5, and a more advanced
stage after the Autonomous, the Integrated stage or 1-6. In the present reanalysis,
the transitional 1-4/5 and the Integrated stages were combined with the Auton-
omous stage to draw an outline of the post-Conscientious identity.
236 Augusto Blasi
ginning of identity. Prior to this stage, one's "identity" is made up of one's

external appearance, social or family relations (The thing I like about my-
self is ... "having a brother," " ... having friends," " ... my heritage"),
specific, frequently role-related actions (" ... my table manners," " ... I
clean the house"), and simple, socially approved, feelings and traits (I am
" ... considerate," " ... a good friend," " ... gentle"). Most of the re-
sponses refer to aspects that just happen to the person and do not involve
much of a subjective experience. There is a concept of oneself, made up of
a list of actions, roles, relations, and simple traits, but not of a unifying self.
But at the Conscientious-Conformist ego stage (strictly speaking, a tran-
sitional stage) identity is understood and experienced in a dramatically
different way. For the first time, the self is reflected on and spoken of as a
whole entity: I am ... "an active person," " ... an insecure person"; My
main problem is ... "myself"; When they avoided me ... "something is
wrong with me." This new entity is made up of traits and no longer of
actions, and, most important, these traits are internal and psychological: I
am ... "insecure," " ... bashful," " ... nervous"; My main problem
is ... "I am too emotional," " ... not patient enough." Social roles, rela-
tions, and demographic characteristics are no longer mentioned.
In other words, there seems to be the discovery of a new substance,
which is differentiated from the global entity that was the person. To use a
metaphor, up to this stage the self was diffused and submerged in each of
the person's actions, more a necessary condition for the experience than a
reality of its own. Now the self begins to be extracted from each action and
to be concentrated; space is created around it; it is isolated and differ-
entiated from other internal processes; it is experienced as the subject of
feelings and processes, a substance in its own right or, better, as the core
and the intimate nucleus of the person.
This new quasi-substance is the person's identity; that is, it is considered
as the real and genuine part of oneself and is opposed to one's external
characteristics as reality is opposed to appearance.
As a result, new feelings are expressed that suggest the person's caring
for his or her inner core: self-confidence, confusion, disappointment with
oneself, pride, and self-respect (social respect was already an issue at
previous stages). In completing the stem, "When I am criticized ... ,"
subjects now answer: " ... I am insulted," " ... I feel rebellious," " ... I
defend my views," " ... I stand up for what I believe."
One important characteristic of this mode of experiencing identity, that
differentiates it from later modes, is that identity seems to be discovered as
existing there, inside oneself, ready-made, a given of one's nature. There-
fore, the inner self becomes the focus of attention and reflection, but not,
as in later modes, the object of construction, decisions, and responsibility.
It is true that agentic characteristics become now, for the first time, an
important part of the self-concept; people describe themselves as being
active, desiring, trying, contributing, attempting, controlling, and so forth.
Agency, however, cooperates with an identity that already exists and is not
applied to shaping one's identity.
The two main characteristics of this mode, the isolation of the inner self
as one's identity and its ready-made nature, explain the surprising impor-
tance that, at this stage, acquire the categories of sincerity and phoniness
or fakeness; The thing I like about myself is " ... that I am honest,"
" ... frank," "not phony," "not fake" are very frequent responses. On
reflection, these concerns fit perfectly well with this mode of identity: if one
splits the inner self from external appearance and if the inner self is also the
real self, one is faced with the problem of expressing it outwardly, through
the observable part. As Trilling (1972) points out, the importance of
sincerity can only be perceived when the person is able to differentiate the
behavioral surface and the deeper central stream, and when one is con-
cerned with discovering and revealing one's hidden true reality, minimally
for the sake of honest interaction.
THE MANAGEMENT OF IDENTITY
The crucial difference between this and the previous mode of experiencing
identity is that, now, identity is not simply there to be observed, but must be
achieved in action. One's efforts aim, at the same time, at proving who one
is and at shaping oneself according to one's ego ideal.
Agency is at the center of self-experience and identity is extended into
action. Thus, "efficiency traits," interests, and abilities become an impor-
tant part of the self-concept and a basis for pride and self-criticism: The
thing I like about myself is... "flexibility," "... adaptability,"
". . . perseverance," ". . . determination," ". . . I can play the piano,"
" ... I am very interested in art," or vice versa, My main problem
is ... "procrastination," " ... indecision," " ... being disorganized,"
" ... my lack of motivation."
More important, agentic characteristics are applied to the management
of oneself: The thing I like about myself is ... "I try to solve my prob-
lems," " ... I'm trying to make something of myself," " ... I am changing
to a better person," " ... I strive to reach the ultimate goal"; When I am
criticized" ... it may be an indication that some type of self-improvement
may be necessary," ". . . I want to find out why and correct it." In these
responses the sense of direction is as evident as the sense of striving and
even struggling.
The basic reason for this transformation seems to lie in a mode of identi-
ty in which inner standards, ideals for oneself, and a personal philosophy
replace a precise and detailed outline. An inner landscape can only be
observed, while standards and ideals have to be worked out in daily action.
At this stage, in fact, responses mentioning goals, ideals, and convictions
are as frequent as those reporting desires, worries, and thoughts, which
characterized the previous stage.
238 Augusto Blasi
Understandably, duty and responsibility now become central categories.

In fact, achievements in the outside world as well as inner accomplishments
are seen as expressions and as extensions of one's identity. Actions are now
understood to be one's own in a much stronger sense: not only because the
agent is their intentional source, but because he or she invests them with
his or her projected identity.
The crucial differentiating characteristic from the later identity mode is
that the goals, standards, and ideals that constitute one's identity are now
an unquestionable given. One's purpose and responsibility lie exclusively
in following and carrying them through with intelligence and determina-
tion. One is only the manager of one's identity.
IDENTITY AS AUTHENTICITY
With respect to identity, Loevinger's Autonomous stage (1-5) seems to be
characterized by a double movement, a negative and a positive movement.
On the negative side, there appears to be a loss of the old certainties,
even those that concerned ultimate goals and ideals for self. There is, in-
stead, the realization that one harbors conflicting desires and irreconcilable
goals: I am ". . . a very confused person in a world of very confused peo-
ple," " ... introvert, shy, with a desire to be friendly and outgoing,"
" ... a combination of many dichotomies"; "I seek satisfaction from life in
love and career and yet I cannot find both." As a result, those "efficiency
traits" in which one used to take pride seem to have lost much of their
importance.
Also on the negative side, there is an attempt to free oneself from social-
ly accepted ideas, an effort to carefully distinguish one's reality from social
stereotypes, and a concern to protect one's independence, particularly in
the context of close and intimate relationships. Thus, one reads: The thing
I like about myself is " ... I don't wonder constantly how others see me.
Most often I couldn't care less," " ... I am completely free of materialistic
and 'high' aiming goals"; My main problem is " ... being too concerned
with seeking others' approval."
Interestingly, in spite of these negative moves, the sense of self and the
sense of unity and direction are not lost; at least, there are no indications
that radical questioning and conflicts produce a sense of fragmentation.
The unity of the self that was previously given by specific clear-cut goals
and ideals now seems to be accomplished by a broad stance toward oneself
and the world. In some, but only in some, responses, there is a clear sense
of searching: My main problem is ". . . trying to decide what I want to do
with the rest of my life," " ... what I am going to do or be," " ... deciding
what my purpose in life is and what I can do to achieve this purpose." More
general characteristics of this stance are openness to the wide world and to
reality and the affirmation of one's ultimate autonomous responsibility to
recognize the truth, whatever it might be.
There are many indications of such an openness. One is the identifica-

tion with the universal humanity: The thing I like about myself is ". . . my
sensitivity to the world around me, to people's feelings, and my sense of
humor," ". . . I like all humanity," ". . . striving to experience as many
things as I can and get a broad view of how the human race lives"; I am
". . . too much in love with everything. . . ," ". . . hopeful that the state
of man will become universally better."
Another indication is the frequently expressed respect for truth and con-
cern with distorting reality: My main problem is " ... my imagination,"
" ... a tendency to daydream too much," " ... being satisfied with day-
dreams instead of dealing with things as they really are"; What gets me into
trouble is ". . . first impressions I have of people and cannot change even
when they are untrue." For the first time, honesty with oneself and self-
deception appear as response categories: The thing I like about myself is
" ... I don't try to lie to myself," " ... that I am honest with myself,"
" ... what I consider my honesty and integrity, perhaps false"; When I am
criticized" ... I often take offense and close my mind to it, even though I
know I shouldn't," " ... I often become defensive, because I can't remem-
ber that people are trying to help me," " ... I try to overcome my habit of
defense."
This respect for reality is not only mentioned, but appears in the direct
recognition of unpleasant truths: My main problem is ". . . I am basically
dishonest and I get in trouble, because I dislike dishonesty and stifle my
impulses," " ... I want people to think that I am better than I am. I try to
put on this impression but it shows. . ."; When they avoided me ". . . I
took it for granted I was no longer needed," " ... there was usually good
reason ... ," " ... I decided that they knew something about me which I
was not aware of."
In sum, the negative movement at this stage should not be interpreted as
an attempt to establish a radical, and empty, individuality; instead, when
seen in the context of the positive movement, it seems to be an effort to
break away from the narrow constraints of social stereotypes and accepted
"truths" in order to be free to search for and find the general truth. One's
identity and one's sense of unity consists in such a broad, contentless,
stance.
This pattern of characteristics seems to correspond to that authenticity
that, according to Trilling (1972), is a discovery of modern Western his-
tory. As Trilling describes it, authenticity suggests,
a more strenuous moral experience than "sincerity" does, a more exigent concep-
tion of self and what being true to it consists in, a wider reference to the universe
and man's place in it, and a less acceptant and genial view of the social circum-
stances of life. (p. 11)
Authenticity is outlined particularly in contrast to sociocultural construc-

tions:
240 Augusto Blasi
At the behest of the criterion of authenticity, much that was once thought to make
up the very fabric of culture has come to seem of little account, mere fantasy or
ritual, or downright falsification. Conversely, much that culture traditionally con-
demned and sought to exclude is accorded a considerable moral authority by reason
of the authenticity claimed for it .... (p. 11)
Many responses manifest an attitude that had never appeared in pre-
vious stages, namely a sense of humor, amusement, and even irony toward
oneself: " ... I laugh at myself," " ... I conceal my temper by laughing at
my idiocy," ". . . I try to muster my sense of humor to alleviate my anger";
When I get angry". . . I am impossible to live with and love every minute
of it." This attitude seems to be as fitting this new mode of identity as
sincerity and the seriousness of duty and responsibility were appropriate in
the two previous modes. This self-directed humor, in fact, allows one to
maintain one's fundamental loyalty to truth and, thus, to acknowledge and
appropriate internal divisions, irreconcilable desires, and even one's dar-
ker aspects, without giving up one's integrity and the hope for a more basic
unity.
Summary and Conclusion

It has been argued that the empirical study of identity has tended to
seriously distort Erikson's understanding of this concept. In his various
analyses, Erikson was inclined to bring together, without clearly separating
them, descriptive statements and theoretical explanatory statements; state-
ments about the subjective experience that characterizes identity, those
about the conditions and the processes of its formation, and those about its
adaptive implications. The distortion, introduced by standard assessment
procedures, seems to be a result of their focusing on the formative pro-
cesses of crisis and commitment. These procedures almost totally neglect
the subjective experience of identity, divorcing those formative processes
from that type of self-experience that, according to Erikson, they are sup-
posed to produce.
The main purpose of this chapter, then, is conceptual, an attempt to
recover the appropriate perspective for understanding psychological
identity. Its basic proposition is that whatever one may wish to say about
identity must be anchored to its subjective core, the description of a special
mode of experiencing oneself as agent and as an individual unified entity;
in other words, that the appropriate perspective for recovering the mean-
ing of Erikson's concept is the perspective of the self-as-subject.
The analysis of Loevinger's ego development categories, using as
framework the dimensions of the self-as-subject, yields three quite differ-
ent modes of experiencing identity. These modes are presented more as a
demonstration that this type of analysis is possible than as actual findings.
Even so, they seem to show that Erikson's description may have to be
revised. First, there may be different forms of identity (only partially over-
lapping with the various statuses as they are presently defined), whose
development may extend far beyond the years of adolescence and even
youth. Second, some of the major characteristics of Erikson's description
may not be inseparable components of one cluster, but may be at least
partially independent of one another and may follow different develop-
mental schedules. For instance, Erikson considers the sense of inner same-
ness, the capacity for commitment, and the questioning of one's society as
parts of the same cluster. Instead, the sense of oneself as having a unified
intrinsic nature (as in the Identity Observed mode) may appear first. The
importance of commitments in constructing one's identity requires a dif-
ferent sense of agency and a different relation to oneself (as in the Man-
agement of Identity mode) and may appear later. Finally, the Identity
as Authenticity mode seems to involve something akin to what Erikson
called crisis and may develop last, if at all.
The descriptions of the identity modes presented here mayor may not
stand up after further and more careful investigation. In any event, ques-
tions concerning the precise patterns of identity and the number of the
basic patterns through development are strictly empirical; their answer
should be left to an intelligent reading of the relevant empirical informa-
tion. However, relevant data can only be obtained when the correct
approach has been secured. This chapter is meant to be a step in that direc-
tion.
REFERENCES
Blasi, A. (1983). The self as subject: Its dimensions and development. Unpublished
manuscript, University of Massachusetts, Boston.
Blasi, A. (in press). The self as subject in the study of personality. In D.J. Ozer,
J.M. Healy, Jr., & A.J. Stewart (Eds.), Perspectives in personality: Self and
emotion. Greenwich, CT: JAI Press.
Blasi, A., & Oresick, R.J. (1985). Emotions and cognitions in self-inconsistency. In
D. Bearison & H. Zimiles (Eds.), Thought and emotion (pp. 147-165). Hills-
dale, NJ: Erlbaum.
Bourne, E. (1978). The state of research on ego identity: Review and appraisal.
Journal of Youth and Adolescence, 7,223-251.
Epstein, S. (1973). The self-concept revisited, or a theory of a theory. American
Psychologist, 28, 404-416.
Erikson, E.H. (1946). Ego development and historical change. In R. Eissler (Ed.),
The psychoanalytic study of the child. (Vol. 2, pp. 359-395). New York: Interna-
tional Universities Press.
Erikson, E.H. (1950). Childhood and society. New York: Norton.
Erikson, E.H. (1959). Identity and the life cycle. Psychological Issues, 1, (whole
No.1).
Erikson, E.H. (1964). Insight and responsibility. New York: Norton.
Erikson, E.H. (1968a). Identity: Youth and crisis. New York: Norton.
Eriskon, E.H. (1968b). Identity, psychosocial. In D.L. Sills (Ed.), International
242 Augusto Blasi
encyclopedia of the social sciences (Vol. 7, pp. 61-65). New York: Macmillan
and Free Press.
Frondizi, R (1971). Nature of the self' A functional interpretation. Carbondale, IL:
Southern Illinois University Press.
Greenwald, A.G. (1982). Is anyone in charge? Personalysis versus the principle
of personal unity. In J. Suls (Ed.), Psychological perspectives on the self (Vol. 1,
pp. 151-181). Hillsdale, NJ: Erlbaum.
Greenwald, A.G., & Pratkanis, A.R. (1984). The self. In R.S. Wyler and T.K.
Srull (Eds.), Handbook of social cognition (Vol. 3, pp. 129-178). Hillsdale, NJ:
Erlbaum.
James, W. (1950). The principles of psychology. New York: Dover Publications
(originally published 1890).
Kelly, G.A. (1955). The psychology of personal constructs. New York: Norton.
Loevinger, J., Wessler, R, & Redmore, C. (1970). Measuring ego development: II.
Scoring manual for women and girls. San Francisco: Jossey-Bass.
Marcia, J.E. (1966). Development and validation of ego identity status. Journal of
Personality and Social Psychology, 3,351-358.
adolescent psychology (pp. 159-187). New York: Wiley.
Markus, H., & 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.
Pratkanis, A.R., & Greenwald, A.G. (1985). How shall the self be conceived?
Journal for the Theory of Social Behavior, 15, 311-329.
Redmore, C., Loevinger, J., & Tamashiro, R.T. (1978). Measuring ego develop-
ment: Scoring manual for men and boys. Unpublished manuscript, Washington
University.
Sarbin, T.R. (1962). A preface to a psychological analysis of the self. Psychological
Review, 59, 11-22.
Stewart, A. (1984). Coding definitions of three aspects of development of the adult
self. Unpublished manuscript, Boston University.
Trilling, L. (1972). Sincerity and authenticity. Cambridge, MA: Harvard University
Press.
341-358.
Waterman, A.S. (Ed.). (1985). Identity in adolescence: Processes and contents. In
New directions for child psychology (No. 30). San Francisco: Jossey-Bass.
12
Self-Theorists, Identity Status, and
Social Cognition
MICHAELD. BERZONSKY
Erikson's (1959, 1968) theory of personality provides the context for most
recent work on identity formation. The extent to which identity research
actually is based on Erikson's theoretical views is at least questionable, as
Looft (1973) noted:
Erikson's concepts are being operationalized and tested by increasing numbers of
researchers .... Most typically, however, any mention of Erikson's theory in
research reports is to be found in the "Discussion" section; it is used as a sort of
after-the-fact framework in which to discuss data already obtained. (pp. 40-41)
Looft was referring specifically to the utilization of Marcia's (1966)

identity-status approach. The status paradigm continues to be the most
prevalent means of operationalizing identity formation (Berzonsky, 1981;
Bourne, 1978; Marcia, 1980; Waterman, 1982) and it continues to be criti-
cized. For example, Blasi (1987) argues that status classifications distort
and trivialize Eriksonian identity processes. Blasi (1987) is correct, of
course. But similar criticisms could be leveled at most attempts to oper-
ationalize psychological processes and constructs; at the very least, some
degree of conceptual richness gets lost in the translation. Yet, operational-
ize we must, if empirical research is going to be the basis for investigating
identity (see Berzonsky, 1986a, 1986b). When Marcia's paradigm is con-
sidered as a heuristic for generating research, its track record is quite good,
even impressive.
Identity statuses are usually conceptualized as differential outcome vari-
ables. In this chapter, I offer an alternative process view. The focus is on
differences among the statuses in the way self-relevant information is en-
coded, elaborated, and structured. Identity is characterized as a self-theory
(Epstein, 1973). Recent work in the philosophy of science is used to articu-
late this identity-as-theory model. The position that individuals classified in
terms of Marcia's identity statuses operate as different types of theorists is
then developed. The identity-as-theorist view includes a norm orientation
(dogmatic theorists), diffuse orientation (ad hoc theorists), and an informa-
tion orientation (scientific theorists). Relevant research on social cognition
244 Michael D. Berzonsky
is reviewed and considered in terms of these processing orientations. I COn-

clude by briefly considering a unified view of identity.
Structure of Identity: A Self-Theory

Ego identity according to Erikson (1968) enables an individual to achieve a
sense of inner sameness and continuity over time. A synchronic dimension
provides coherence to the contemporaneous elements in one's life, and a
diachronic dimension signals the temporal integration of personal past,
present, and future. Identity is a unified configuration that emerges from,
but is not reducible to, earlier identifications and life experiences. This
self-structure has been conceptualized in various ways (Berzonsky, 1986a;
Greenwald, 1980; McAdams, 1985). In this chapter, the structure of identi-
ty is conceptualized as a self-generated theory about the self. By analogy,
the self-theory serves the same functions and is composed of the same
elements as a scientific theory.
It is a theory that the individual has unwittingly constructed about himself as an
experiencing, functioning individual. ... [It contains] major postulate systems for
the nature of the world, for the nature of the self, and their interaction. Like most
theories, the self-theory is a conceptual tool for accomplishing a purpose. [Major
purposes are] to optimize the pleasure/pain balance of the individual over the course
of a lifetime . .. and to organize the data of experience in a manner that can be coped
with effectively. (Epstein, 1973, p. 407)
A formal scientific theory is evaluated in terms of pragmatic usefulness.
The self-theory functions in the same pragmatic manner: Does it solve the
problems that it was constructed to handle? Does it generate novel predic-
tions that otherwise would not be made? Does it offer a parsimonious and
effective means of abstracting, coding, integrating, and interpreting rel-
evant information? Whether or not one's self-theory is a "true" repre-
sentation is irrelevant and unanswerable (Berzonsky 1986a, 1986b; see
Waterman, 1986, for an alternative view). As Schlenker (1985) suggests,
personal theoretical beliefs and principles have utility to the degree they
are believable (logically coherent and empirically valid) and personally
beneficial (provide explanations and interpretations that serve one's values
and objectives). Pragmatic utility, of course, is constrained by the physical
and social reality within which a person functions (Berzonsky, 1986b).
THEORY DEVELOPMENT
Contemporary philosophers of science emphasize the constructive nature

of knowing (Kuhn, 1970; Lakatos, 1970; Popper, 1970; Suppe, 1977). For
instance, Thomas Kuhn's (1962) widely read classic on paradigms and sci-
entific revolutions convincingly argued against a linear "accretion-model"
12. Self-Theorists, Identity Status, and Social Cognition 245
of objective data-driven scientific discovery. Instead, he proposed that de-

velopment proceeded dialectically via alternating periods of paradigm-
dominated "normal science" punctuated by chaotic theoretical "revolu-
tions. "
Scientific Research Programs

For present purposes, the conception of "scientific research programs"
provided by Lakatos (1970) offers a useful metaphor for characterizing
changes in personal self-theories. While his account is not identical with
Kuhn's, the "gist" is retained but cast in terms designed to minimize
conceptual vagueness. A research program is composed of three distinct
levels: an irrefutable "hard core" of assumptions and propositions, a re-
visable "protective belt" of auxiliary hypotheses that are responsive to
evidence, and a "research heuristic" or methodological rules for solving
problems. Progressive research programs maintain the hard core by adap-
tively modifying the protective belt of theoretical ideas in innovative
ways whenever anomalies emerge. A revised protective belt in a progres-
sive program does more than account for the successes of its predecessor
and the anomalies that did it in, however. The new theory also generates
novel predictions that can be verified; scientific progress ensues. Research
programs may also be degenerating; anomalies are handled by ad hoc
defensive manuevers that raise more long-term problems than they "solve"
in the short run. A point may be reached when hard-core assumptions must
be abandoned; a new core must be sought. Of course, there may be differ-
ences in the tenacity and creativity with which the core is defended; nega-
tive heuristics, for instance, may be employed to avoid potentially anoma-
lous findings (Lakatos, 1970).
Personal Research Programs

This philosophy-of-science view provides a metaphor for answering the
basic question of identity formation: How do individuals develop identities
that are believable and personally beneficial? We hold an irrefutable
"core" of assumptions about life, reality, knowledge, and so forth. Desired
images of the self (Schlenker, 1982) are also included. This core is not
necessarily explicit. The self-theory serves as a protective buffer as in-
dividuals attempt to maintain self-esteem, maximize the pleasure-pain
balance, make sense out of personal reality, cope effectively with the
problems and obstacles encountered in day-to-day living, and the like.
While physical and social reality afford constraints that all self-theories
must consider, the nature of hard-core propositions and identity images
provides considerable latitude in determining what constitutes a personal
problem, what are relevant data, how effective a solution is, and so on.
A person's overall identity is composed of self-specific content areas:
"me" as a student, friend, athlete, brother/sister, and so forth (Berzonsky
& Barclay, 1987; Epstein, 1973; James, 1890/1950; Kelly, 1955). The
personal salience (Stryker, 1968) of content domains or self-schemata (cf.
Kihlstrom & Cantor, 1984; Markus & Sentis, 1982) will vary with core
beliefs. Research programs or personal problem-solving efforts within
these content domains may be developing at different rates and may be
subject to different environmental andlor social demands. Some may be
degenerative (the religious self), while others are progressing (the voca-
tional self). Asynchronous personal development may be observed across
self-domains. In this view, a resolved Eriksonian identity crisis represents a
change in hard-core assumptions and a qualitative restructuring of the
overall program.
Identity-as-Theorist: A Process Orientation

A scientific approach to theory construction can be conceptualized as an
alternating spiral of deductive and inductive reasoning (Cattell, 1966).
Existing theoretical principles and expectations provide a basis for inter-
preting information and solving problems; data are assimilated into the
existing theoretical structure. A theory also generates novel hypotheses
and deductions about the world. These predictions, tested against publical-
ly verifiable data, are either confirmed or they yield anomalous findings. In
progressive research programs, anomalies induce accommodations or revi-
sions in the theoretical structure that, in turn, generate new deductions,
and the spiral continues.
A diagrammatic representation of this feedback loop as it applies to self-
theory construction appears in Figure 12.1. Individuals live and exist within
a social and physical reality. As one interacts in daily life, self-relevant
information and problems are encountered. The self-theory or identity
structure contains cognitive schemata and scripted behavioral strategies
(e.g., Neisser, 1976; Schank & Abelson, 1977) for handling experiences,
problems, stressors, and so forth. However, to adapt resourcefully one has
to monitor information about the success or effectiveness of a given
approach. If assimilation is unsuccessful, a state of dissonance exists and
accommodative efforts to modify and revise relevant aspects of the self-
theory would commence. These are the functions that comprise the TOTE
unit of Miller, Galanter, and Pribram (1960): the test-operate-test-exist
sequence. An assimilative approach is the initial reaction used to interpret
and cope with environmental situations. Assimilative processes are
assumed to be relatively automatic (Shiffrin & Schneider, 1977) or "mind-
less" (Langer, Blank, & Chanowitz, 1978). While such processing can be
efficient (see Hansen, 1985), it can also lead to theory-driven distortions of
reality at times (see below). Given objective reality monitoring, however,
a point will be reached when "data" from the situation necessitate an
accommodative response. A balanced, flexible deployment of assimilative
Self-Theory
(Identity Structurel
Accommodative A.slmllatlve
Proce •• es Proce •• e.
Social and
Physical Reality
FIGURE 12.1. Self-Regulatory Model of Self-Theory Utilization and Revision.
and accommodative processes provides the basis for resourceful personal

adaptation (see Berzonsky & Barclay, 1981; Block, 1982; Showers & Can-
tor, 1985).
The cognitive component of the feedback loop is focal in this chapter.
A behavioral component is also assumed. Schemata for strategic problem
solving as well as schemata for cognitive representations and interpreta-
tions are available. The self-theory contains procedural and declarative
knowledge (Winograd, 1975) or operative and figurative aspects (Piaget,
1964). Accommodative cognitive restructuring will ordinarily be attention
demanding (Kahneman, 1973). Behavioral adjustments, however, may
occur without mentally effortful cognitive accommodations. For instance,
one may be extremely vulnerable to environmental demands and distrac-
tions and be unable or unwilling to delay personal satisfaction (Block &
Block, 1980). Alternatively, momentary behavioral compliance may result
from hedonic concerns without long-term changes in one's cognitive belief
structure (Kelman, 1958).
While person-as-scientist metaphors abound in psychology (e.g., Heid-
er, 1958; Kelly, 1955; Kelley, 1967), current research in social cognition
portrays the normative person as a "cognitive miser" exerting minimal men-
tal effort and utilizing a variety of self-serving confirmation biases to main-
tain existing beliefs (Fiske & Taylor, 1984). This corresponds to an over-
emphasis on the assimilation side of Figure 12.1. In addition there are
hedonically driven ad hoc theorists trying to make do in the short run, but
with limited concern about distal implications-an overemphasis on the
accommodation side of Figure 12.1. I now attempt to illustrate that these
orientations to self-theory construction, revision, and development can be
mapped onto current theoretical treatments of identity status as oper-
ationalized by Marcia's (1966) scheme.
Marcia's Identity-Status Paradigm

Marcia (1966) simultaneously considered two Eriksonian identity dimen-
sions, effortful self-exploratory crises and personal commitments. A crisis
period is more than stress and confusion; deliberate self-examination and
an active processing of self-relevant information must occur. Commitment
is evidenced by a stable self-structure; there is personal investment in what
one does and believes.
STATUS CLASSIFICATIONS
An identity-status interview is used to assess the presence or absence of
self-reported crises and commitment. Two of the four status classifications
involve self-reflection: Achievers and Moratoriums. They differ in their
commitments. Achievers are personally committed; they have reportedly
resolved an identity crisis. Moratoriums are currently attempting to resolve
a crisis. The remaining two statuses are distinguished by the absence of
effortful self-examination: Foreclosures and Diffusions. Foreclosures have
a firm self-definition that was not derived through personal decision mak-
ing and role experimentation. They have more automatically adopted
norms, values, and expectations prescribed by significant others, especially
their parents. Diffusions have not internalized social identity prescriptions,
and they are not attempting to forge a personal self-definition.
Data accumulated over the past two decades indicate that the scheme
is a reasonably valid method for assessing interindividual differences in
identity formation (see Berzonsky, 1981, 1985; Bourne, 1978; Marcia,
1980; Waterman, 1982). This differential-status approach implies that
identity is an outcome variable. While interstatus developmental progress
may occur, personal stability is expected once identity has been achieved.
Recent evidence, however, has revealed anomalous intraindividual
changes in identity status (Adams & Fitch, 1982; Marcia, 1976). Theoreti-
cally regressive and inconsistent status changes may to some degree reflect
measurement error. However, an alternative process-oriented interpreta-
tion of these anomalous longitudinal findings is also possible. The statuses
may represent, or at least be associated with, different styles of personal
decision making and problem solving (Berzonsky, 1986c, 1987; Berzonsky
& Barclay, 1987). Self-explorers, Moratoriums and Achievers, are in-
formation oriented (cf. Petty & Cacioppo, 1986); they actively seek out,
elaborate, and evaluate relevant information before making decisions and
commitments. Foreclosures are more norm oriented (cf. Ajzen & Fishbein,
1977), focusing on the normative expectations held for them by significant
referent others, especially parental figures. Uncommitted Diffusions utilize
a diffuse orientation (cf. Block & Block, 1980); they tend to delay and
procrastinate until the hedonic cues in the immediate situation dictate a
course of behavior. These orientations suggest different approaches to self-
theory construction and revision.
SELF-REFLECTIVE STATUSES: SCIENTIFIC PERSONAL THEORISTS
Moratoriums are currently engaged in an ongoing process of self-

reflection. Achievers have been, at least in the past, engaged in self-
reflection. Moratoriums are the focus of this section. An Achiever's firmly
committed identity structure may result in the sort of biased processing
suggested below to characterize Foreclosures (see Berzonsky, 1986c, 1987,
for further discussion). Of course, all thinking and self-reflection is biased
by existing hard-core assumptions and theoretical views. The issue is indi-
vidual differences in anomaly detection and the manner in which self-views
are conserved.
Self-reflection involves making inferences and hypotheses about one's
mental states and psychological dispositions. However, it is not assumed
here that individuals have direct introspective access to these internal
states. One's "true self" or daimon is not discovered via self-examination,
as some have suggested (see Waterman, 1981, 1984). Instead, deliberate
self-analysis is conceptualized as an indirect attempt to construct a believ-
able and personally beneficial account about behavior', feelings, attitudes,
and so forth. The literature on social cognition is replete with illustrations
of individuals telling (e.g., Nisbett & Wilson, 1977; Wilson, 1985), feeling
(e.g., Schachter & Singer, 1962; Wilson, 1979), and remembering (e.g.,
Conway & Ross, 1984; Sherman, 1980) more than they "really know"
about themselves. Self-inferences can be "correct" in the sense that any
theoretical account can be believable and pragmatically beneficial. Yet,
there will be times when the act of self-reflection leads one astray (e.g.,
Wilson, Dunn, Bybee, Hyman, & Rotondo, 1984), at least until views are
tested and accommodative revisions made.
This information orientation to identity issues (Berzonsky, 1987; Ber-
zonsky & Barclay, 1987) implies a tendency to seek out, elaborate, and
evaluate self-relevant information in an objective fashion. Issue-relevant
elaboration has been found to influence attitude change (Petty & Caciop-
po, 1986). For instance, subjects experimentally induced to objectively
process a persuasive message, change their attitudes in the direction of the
advocacy when strong compelling arguments are presented. However, the
same elaboration-facilitating conditions lead to counterarguing and negligi-
ble attitude change when specious arguments are presented (Petty &
Cacioppo, 1979). The quality of the information that comprises a persua-
sive message is the main determinant of attitude change when the likeli-
hood of objective elaboration is high. Situational factors found to influence
the objective processing of persuasive messages include personal relevance
(Petty, Cacioppo, & Goldman, 1981) and individual responsibility (Petty,
Harkins, & Williams, 1980).
Individual differences in the extent to which persuasive messages are
objectively evaluated have also been found (Cacioppo & Petty, 1982;
Cacioppo, Petty, & Kao, 1984). According to the model in Figure 12.1,
self-constructs must be tested behaviorally and feedback must be moni-
tored and processed if adaptive functioning is to occur. Individual differ-

ences in responding to environmental demands, assuming a skeptical and
tentative attitude toward theoretical constructions, and displaying a will-
ingness to test and revise in light of contradictory evidence distinguish the
more objective efforts of a scientific personal theorist from a dogmatic
theorist's more defensive efforts (see also Berzonsky & Barclay, 1981;
Showers & Cantor, 1985). To use Petty and Cacioppo's (1986) terms,
objective theorists and motivated to evaluate and ascertain the "validity"
of information, and biased theorists are motivated to defend preexisting
beliefs.
FORECLOSURES: DOGMATIC PERSONAL THEORISTS

Foreclosures are rigidly committed to a set of beliefs and values that have
not been personally examined and evaluated. This belief structure drives
thinking in a decidedly biased fashion. Greenwald's (1980) portrayal of the
normative self as a thought-controlling totalitarian society, seems to apply
best to a foreclosed self-theorist. Confirmation-biased information
searches designed to defend and conserve established belief structures
typify such thinking (Greenwald, 1980; Swann, 1985).
Research on belief perseverance indicates that firmly held beliefs do
indeed produce a biased assimilation of information (Ross, Lepper, &
Hubbard, 1975). Instead of undermining confidence and certainty, discon-
firming evidence may actually strengthen preexisting beliefs. For instance,
Lord, Ross, and Lepper (1979) had subjects, who either strongly opposed
or supported capital punishment, read two "studies," one of which con-
firmed and one that disconfirmed the subjects' belief. Confirmation biases
were revealed when the "methodological soundness" of the two studies
was rated; both opponents and proponents rated the particular study that
supported their belief as being superior to the other. The mixture of con-
firming and disconfirming evidence resulted in a greater polarization of
attitudes. Lord et al. (1979) suggest that individuals with strong opinions
subject "disconfirming" evidence to a biased evaluation, while they accept
at face value data that "confirms" the position to which they are com-
mitted. Consistent with the present position on self-reflective theorizing,
instructing subjects to monitor their cognitive processing with a consider-
the-opposite strategy has been found to have a corrective effect on the
biased assimilation of evidence (Lord, Lepper, & Preston, 1984).
Research by Chanowitz and Langer (1981) is also relevant to foreclosed
beliefs. They investigated the subsequent utilization of beliefs to which
subjects had prematurely committed themselves (i.e., beliefs foreclosed to
alternative explanations). Subjects were given fictional information about
a perceptual deficit, "field dependence." The self-relevance of the in-
formation was manipulated by leading the subjects to believe that the
deficit was very common or rare in the population (80% versus 10%). Fol-
lowing "perceptual assessments" revealing that all subjects were "field

dependent," an actual field-dependence test was administered. Subjects
in the self-irrelevant (10%) condition found significantly fewer embedded
objects than subjects in the self-relevant (80%) condition or control sub-
jects in an offset no-treatment condition. The explanation offered by Cha-
nowitz and Langer (1981) is consistent with our discussion; the presumably
self-irrelevant information was not objectively evaluated, which lead to a
foreclosed or premature commitment that precluded alternative explana-
tions. When the information subsequently became self-relevant (after the
"perceptual assessment"), it was rigidly maintained and it influenced be-
havior. However, subjects in the self-irrelevant condition who were ex-
perimentally induced to think critically about the information by consider-
ing possible compensatory strategies, did not demonstrate a performance
decrement on the field-dependence test (Chanowitz & Langer, 1981).
The foreclosed or dogmatic self-theorist's one-sided beliefs have not
been personally evaluated. Biased assimilation-side social cognition is most
likely to occur (see Figure 12.1). In comparison to a more objective pro-
cessor, the biased processor finds compelling pro attitudinal evidence to
be more convincing, but high-quality counterattitudinal evidence to be less
convincing. With specious evidence, the opposite effects will occur (Petty
& Cacioppo, 1986).
DIFFUSIONS: AD Hoc PERSONAL THEORISTS

Diffusions lack firm commitments and they are not expending a lot of men-
tal effort attempting to formulate a personal belief structure. They display
a tendency to procrastinate and to avoid confronting personal problems
(Grotevant & Adams, 1984). Their behavior tends to be situation specific,
governed by relatively immediate environmental consequences. For in-
stance, Berzonsky (1986c) found that highly diffuse subjects held extemal-
control expectancies and that a diffuse orientation was positively associated
with self-monitoring scores. High self-monitors deal with situations by
asking themselves, "Who does this situation expect me to be and how can I
be that person?" (Snyder, 1979, p. 102). This implies a concern to manage
one's self-presentation so as to maximize the rewards afforded by the con-
temporaneous situation. The skill with which one engages in such strategic
interactions may not be a linear function of diffuseness, however. A diffuse
orientation is not being equated with Snyder's (1979) self-monitoring con-
struct. The absence of inner-directedness and the immediate expression of
hedonistic motives and affect is being stressed (Block & Block, 1980).
In terms of the model in Figure 12.1, a diffuse orientation to personal
theorizing is predominantly accommodation sided. One is willing to change
chameleon-like from situation to situation. Such alterations, however, tend
to involve transient behavioral compliance rather than stable cognitive
restructuring (see Hoffman, 1977; Kelman, 1958).
Identity Status and Social Cognition

To what extent is research on the identity statuses consistent with these
proposed differences in social-cognitive orientations? Since most available
research was not designed in terms of these distinctions, post hoc inter-
pretations are required. A brief review of studies examining social-
cognitive correlates of Marcia's (1966) identity statuses is presented,
followed by a more detailed consideration of status differences in inter-
nalizing self-presentation behaviors.
SOCIAL-COGNITIVE CHARACTERISTICS OF THE IDENTITY STATUSES

Foreclosures have consistently been found to score the highest on mea-
sures of Authoritarianism (Marcia, 1966, 1967; Marcia & Friedman, 1970;
Schenkel & Marcia, 1972). This is consistent with the view that Fore-
closures, hold inflexible self-belief systems that are resistant to change
and intolerant of ambiguous information. Self-reflective Moratoriums and
Achievers have been found to display significantly greater integrative com-
plexity in reasoning than Foreclosures or Diffusions (Slugoski, Marcia, &
Koopman, 1984), indicating that Foreclosures and Diffusions have difficul-
ty adopting a self-determined perspective within which multiple and con-
flicting sources of information can be integrated. Likewise, status compari-
sons on measures of attentional and interpersonal style (Read, Adams, &
Dobson, 1984) revealed that Foreclosures were least apt to evaluate and
integrate information from multiple perspectives, and that Foreclosures
and Diffusions failed to attend to relevant information because of a re-
stricted attentional focus. Achievers and Moratoriums, in contrast, were
found to process greater quantities of information, report more confidence
in their own views, and make fewer impulsive errors when information
load increased (Read et al. 1984). Finally, Grotevant and Adams (1984)
found that total achievement scores on an identity-status interview were
positively associated with a reported tendency to think about and confront
problems and to rely on self-directed problem-solving strategies. Diffuse-
ness was related to a reported tendency to avoid facing personal problems,
and both Foreclosure and Diffusion scores were correlated with other-
directed problem-solving strategies. Foreclosures indicated a marked de-
pendence on others as well.
This admittedly sketchy picture is consistent with the position that the
statuses reflect different orientations to self-theory construction. Identity-
status differences in the effect that public self-expressions may have on
private self-conceptions will now be considered.
IDENTITY STATUS AND INTERNALIZATION OF SELF-PRESENTATIONS

Research indicates that private self-conceptions may be influenced by
public self-presentation behaviors (see Schlenker, 1985). For instance,
Jones, Rhodewalt, Berglas, and Skelton (1981, Study 3) had subjects, in a

simulated job interview, present themselves in either a self-enhancing or
self-deprecating fashion. Self-esteem assessments were then obtained.
Subjects were given high or low choice to participate (choice condition),
and they either improvised their own interview responses (self-reference
condition) or followed a prepared script. Self-deprecating subjects showed
decreased self-esteem, provided they had freely chosen to participate. Self-
deprecating subjects in the no-choice conditions showed no internalization.
Self-enhancing subjects displayed elevated self-esteem, but only in the self-
reference conditions. Jones et al. (1981) offered a dual-process explanation
of these presentation-induced carry-over effects on self-esteem: Positive
changes resulted from self-inferences (Bern, 1972), whereas the arousal of
cognitive dissonance (Festinger, 1957) produced the negative changes.
This explanation is based on two different views of self-conceptions.
Opposing Conceptions of the Self

Self-views have been conceptualized in differing ways, some of which seem
to be contradictory. One position (e.g., Bern, 1972) maintains that self-
conceptions exist in a variable state of flux; they result from, rather than
cause, social interactions. (This view seems to describe Marcia's uncom-
mitted statuses.) The accessibility of particular self-views is enhanced by
situational cues and recent behavior. Jones et al. (1981) suggested that role
playing increased the salience of particular self-views and the subsequent
self-evaluations on the self-esteem measure were, therefore, biased in the
direction of the self-presentation. Personally improvised positive be-
haviors, presumably self-representative, would most likely produce this
effect.
An alternate view maintains that self-conceptions are firmly structured.
(This view appears to characterize Marcia's committed statuses.) Cognitive
dissonance would result from self-discrepant behavior for which an indi-
vidual is personally accountable (Wicklund & Brehm, 1976). Jones et al.
(1981) suggested that dissonance, created by the self-discrepant negative
presentations in the choice condition, produced the deflated self-esteem. A
crucial consideration in evaluating their intriguing but post hoc account is,
"Can the conditions in which each view will be applicable be predicted?"
Jones et al. (1981) turned to the attitude-change literature for an answer.
Specificially, they proposed that the domains within which self-perception
and dissonance theories apply depend on whether a presentation falls with-
in an individual's latitude of acceptance or rejection.
Latitudes of Acceptance and Rejection

According to Hovland, Harvey, and Sherif (1957), self-relevant views and
dimensions involve a range of values. The latitude of acceptance is defined
as the range of values on a dimension (e.g., assertive/submissive) consid-
ered to be self-descriptive. Unacceptable values comprise the latitude of
rejection, and the latitude of noncommitment is the area in which the indi-
vidual would equivocate. Fazio, Zanna, and Cooper (1977) provided evi-
dence that dissonance processes mediated change when attitude-discrepant
behavior fell within a subject's latitude of rejection. However, self-
perception processes were found to be relevant when behaviors fell within
a subject's latitude of acceptance or noncommitment. Therefore, Jones et
al. (1981) speculated that dissonance was aroused by a self-deprecating
presentation for which subjects felt personally responsible because it fell
within their latitude of rejection. Self-perception processes, however,
were responsible for changes induced by a self-referenced positive pre-
sentation that presumably fell within the latitude of acceptance. Recent
research is consistent with this latitude-of-acceptance/rejection account
(Rhodewalt & Agustsdottir, 1986; Trudeau, 1986).
Identity statuses who vary in the firmness of their personal commitments
may also differ in the width of their latitudes of acceptance, noncommit-
ment, and rejection. Personal involvement has been found to increase the
latitude of rejection (Sherif & Sherif, 1967). Therefore, one would expect
uncommitted Moratoriums and Diffusions to be more susceptible to self-
perception effects (but less apt to experience dissonance) than committed
Achievers and especially Foreclosures. In addition, the different orienta-
tions to self-theory construction proposed above suggest status differences
in the effect that public self-presentations will have on private self-views. A
recent experiment by Berzonsky, Schlenker, and McKillop (1987) investi-
gated some of these issues.
Internalizing Public Self-Expressions

Berzonskyet al. (1987) assigned subjects, pretested on measures of identi-
ty status and sociability, to one of six self-presentation conditions. They
were instructed to create either a positive or negative impression of their
"sociability"-ostensibly an important quality for the "job" for which they
were being interviewed-in one of three presentation conditions: 1) a
face-to-face interview; 2) a written questionnaire; and 3) an anonymous
written questionnaire. All subjects were given a choice "to help us out"
(none refused), and they were also asked not to be "clearly untruthful."
Internalization was operationalized as posttest-pretest change scores on
the sociability measure.
Overall, subjects internalized positive presentations and resisted inter-
nalizing negative ones. Interview format did not have a significant effect.
When subjects present themselves in a self-representative positive way, the
behavior would generally fall within their latitude of acceptance, and self-
inferences will produce internalization. A negative role-playing experi-
ence, however, would generally fall in their latitude of rejection and create
dissonance. Since dissonance can be resolved by means other than self-
belief revision (e.g., objective counterarguing, biased processing, etc.),
internalization is resisted. Assuming that committed and uncommitted

statuses differ in their personal latitudes of acceptance, rejection, and
noncommitment, self-presentation effects would vary with identity-status
scores.
Status x Internalization correlations calculated within the combined
positive and negative presentation conditions revealed increased suscep-
tibility among uncommitted Moratoriums in the negative conditions
[r (55) = -.27, P < .05.], and enhanced resistance among rigidly commit-
ted Foreclosures in the positive conditions [r (51) = -.31, P < .05]. No
other significant relationships were found. The general tendency among all
subjects to internalize positive behaviors was not enhanced by a self-
reflective moratorium state of noncommitment. Likewise, the normative
tendency to resist internalizing a negative self-expression was not increased
by a nonreflective foreclosed belief structure.
While these data do not provide a definitive basis for resolving process-
relevant issues, they appear to be consistent with the self-theory orien-
tations presented above. The findings suggest that the process by which
public self-presentations influenced private self-conceptions differed by
identity status.
Uncommitted Moratoriums may have very wide latitudes of acceptance
or noncommitment on specific personality dimensions, thereby increasing
the likelihood that a presentation is perceived to be self-reflective. Also,
given an ongoing process of self-reflection, the elaborating, rehearsing,
and interpreting of previously experienced events may be biased, at least
initially, by feedback from their recent behavior (Bern, 1972; Jones et al.,
1981).
Biased assimilation and resistance to internalization may be a general-
ized response to negative self-presentations, and being foreclosed did
not increase it. Why, however, did the Foreclosures resist the positive
role? Possible explanations include: latitude-of-rejection differences on the
dimension (they were most apt to view the role as self-incongruent and
experience dissonance); concern about being accountable for a positive
image that they may not be able to live up to (Schlenker, 1982); and react-
ance to a perceived threat to their freedom to act as they choose (Brehm,
1966).
Status x Internalization correlations within the positive and negative
face-to-face conditions alone revealed that diffuseness was related to in-
creased sociability following an actual positive encounter [r (16) = + .48,
p < .05] and inversely but nonsignificantly related following a negative
encounter [r (15) = - .22, p> .10]. Although no other correlations were
significant, Moratoriums tended to internalize both actual presentations:
positive face-to-face presentation, r (16) = + .18, p> .10; negative pre-
sentation, r (15) = - .18, p > .10.
These findings support the view that Diffusions may be especially re-
sponsive to situational affective cues. Immediate consequences would be
minimal in the written and anonymous conditions. In an actual public set-

ting, however, highly diffused individuals may attend to the impression
that they are making. Such self-monitoring would influence the informa-
tion that is encoded and available in memory. Selective retrieval would
tend to occur if they are subsequently asked to make self-judgments. The
tendency to monitor their ongoing behavior, however, may vary with the
instrumental value of the setting.
The meaning of these "internalization" scores may vary with identity
status. For self-reflective Moratoriums, they may represent "hypotheses"
to be considered, tested, revised, and so forth. For Diffusions, they may
represent momentary verbal "compliance" with a perceived situational
demand. For Foreclosures, they may represent the defensive reaction of
biased processing.
Conclusions: Toward a Unified View of Identity

While social-cognitive research is consistent with a self-theorist account,
confusion arises because process may be confounded with structure in
Marcia-type identity-status classifications. Consequently, I did not focus
on Achievers who are both self-reflective and firmly committed. Identity
appears to involve three interdependent components: process, the means
by which identity content is encoded, elaborated, and integrated; structure,
the way identity is organized; and content, the information from which
identity is constructed (Berzonsky, 1986a, 1986c, 1987; Berzonsky & Bar-
clay, 1987). The process component is emphasized in the present chapter.
(Berzonsky and Barclay [1987] provide further discussion of the model.)
The process component is assumed to involve at least three levels. The
most basic units are the actual behavioral patterns and cognitive responses
that individuals perform as they deal with their lives. Social-cognitive
strategies are integrated collections of behaviors and cognitive responses.
Identity style refers to the strategy that an individual prefers to use or
characteristically employs.
It is assumed that Marcia's statuses best correspond to self-reported dif-
ferences in preferred identity styles: Moratoriums and Achievers are in-
formation oriented; Foreclosures are more norm oriented, and diffusions
tend to delay and utilize a diffused orientation. By adolescence, virtually
all individuals would be capable of utilizing the social-cognitive strategies
that comprise each of the three styles. Environmental demands may, there-
fore, induce individuals to employ, say, elaborative objective analyses of
self-relevant information, even though they may not typically utilize that
style. Thus, situational factors like personal relevance (Petty et aI., 1981),
individual accountability (Petty et aI., 1980), freedom of choice (Jones et
al. 1981), and the like may override style preferences. Of course, the
extent to which subjects are susceptible to situational demands and ex-
perimental manipulations may vary with style differences in processing

orientation (Berzonsky et aI., 1987).
The structural component and specific content domains may also in-
fluence the way in which self-relevant information is processed. Firmly
committed individuals have well-organized and integrated self-structures
that "drive" processing and assimilation in an efficient but schema-biased
fashion (Hansen, 1985; Kihlstrom & Cantor, 1984; Markus, 1977; Rogers,
1981). We may find information-oriented Achievers (who are also firmly
committed) responding like Foreclosures in some situations (Berzonsky,
1986c, 1987). Interstatus changes that appear to be theoretically inconsis-
tent or regressive (e.g., Adams & Fitch, 1982; Marcia, 1976) may reflect
style preferences being overridden by environmental or structural factors.
It is simply inefficient to continually reconsider and reevaluate issues
that have been resolved, especially when these self-views are believable
and personally beneficial.
This view is consistent with research on life-span identity crises (e.g.,
Gould, 1978; Levinson, 1978; Vaillant, 1977). A normative orientation in
adulthood usually is effective. A life crisis may induce the use of a more
elaborative information-oriented strategy. Perhaps there are times when a
defensive or "degenerative" program of personal problem solving necessi-
tates revisions of "hard-core" beliefs (Lakatos, 1970). Longitudinal studies
of individuals undergoing major life transitions would report "anomalous"
changes in identity classifications, unless these process factors were taken
into account.
REFERENCES
Adams, G.R., & Fitch, S.A. (1982). Ego stage and identity status development: A
cross-sequential analysis. Journal of Personality and Social Psychology, 43,574-
583.
Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analy-
sis and review of empirical research. Psychological Bulletin, 84, 888-918.
Bern, D.J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in
experimental social psychology (Vol. 6, pp. 1-62). New York: Academic Press.
Berzonsky, M.D. (1981). Adolescent development. New York: Macmillan.
Berzonsky, M.D. (1985). Diffusion within Marcia's identity-status paradigm: Does
it foreshadow academic problems? Journal of Youth and Adolescence, 14, 527-
538.
Berzonsky, M.D. (1986a). Discovery versus constructivist interpretations of iden-
tity formation: Consideration of additional implications. Journal of Early
Adolescence, 6, 111-117.
Berzonsky, M.D. (1986b). Identity formation, metaphors, and values: A rejoinder
to Waterman. Journal of Early Adolescence, 6, 123-126.
Berzonsky, M.D. (1986c). A measure of identity style: Preliminary findings. Paper
presented at the meeting of the Society for Research on Adolescence, Madison,
Wisconsin.
Berzonsky, M.D. (1987). Identity style: Conceptualization and measurement.

Unpublished manuscript, State University of New York at Cortland, Cortland,
New York.
Berzonsky, M.D., & Barclay, C.R. (1981). Formal reasoning and identity forma-
tion: A reconceptualization. In J.A. Meacham & N.R. Santilli (Eds.), Social
development in youth: Structure and content (pp. 64-87). Basel: Karger.
Berzonsky, M.D., & Barclay, C.R. (1987). Toward a unified view of identity forma-
tion: Individual differences in self construction. Unpublished manuscript, State
University of New York at Cortland, Cortland, New York.
Berzonsky, M.D., Schlenker, B.R., & McKillop, K., Jr. (1987). Identity status,
se/f-presf!ntation, and self-conceptions. Paper presented at the Biennial Meetings
of the Society for Research for Child Development, Baltimore, Maryland.
Blasi, A. (1987). Identity and the development of the self. In D.K. Lapsley & F.C.
Power (Eds.), Self, ego, and identity: Integrative approaches (pp. 226-242). New
York: Springer-Verlag.
Block, J. (1982). Assimilation, accommodation, and the dynamics of personality
development. Child Development, 53, 281-295.
Block, J.H., & Block, J. (1980). The role of ego-control and ego-resiliency in the
organization of behavior. In W.A. Collins (Ed.), Minnesota symposia on child
psychology (Vol. 13, pp. 39-101). Hillsdale, NJ: Erlbaum.
Bourne, E. (1978). The state of research on ego identity: A review and appraisal:
Part II. Journal of Youth and Adolescence, 7,371-392.
Brehm, J.W. (1966). Response to loss offreedom: A theory of psychological react-
ance. New York: Academic Press.
Cacioppo, J.T., & Petty, R.E. (1982). The need for cognition. Journal of Personal-
ity and Social Psychology, 42, 116-131.
Cacioppo, J.T., Petty, R.E., & Kao, C. (1984). The efficient assessment of need for
cognition. Journal of Personality Assessment, 48,306-307.
Cattell, R.B. (1966). Psychological theory and scientific method. In R.B. Cattell
(Ed.), Handbook of multivariate experimental psychology (pp. 1-18). Chicago:
Rand McNally.
Chanowitz, B., & Langer, E.J. (1981). Premature cognitive commitment. Journal
of Personality and Social Psychology, 41, 1051-1063.
Conway, M., & Ross, M. (1984). Getting what you want by revising what you had.
Epstein, S. (1973). The self-concept revisited: Or a theory of a theory. American
Psychologist, 28, 404-416.
Erikson, E.H. (1959). Identity and the life cycle. Psychological Issues Monograph
Series (No.1). New York: International Universities Press.
Erikson, E.H. (1968). Identity: Youth and crisis. New York: Norton.
Fazio, R.H., Zanna, M.P., & Cooper, J. (1977). Dissonance and self-perception:
An integrative view of each theory's proper domain of application. Journal of
Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford Uni-
versity Press.
Fiske, S.T., & Taylor, S.E. (1984). Social cognition. Reading, MA: Addison-
Wesley.
Gould, R.L. (1978). Transformations: Growth and change in adult life. New York:
Simon and Schuster.
Greenwald, AG. (1980). The totalitarian ego: Fabrication and revision of personal
Grotevant, H.D., & Adams, G.R. (1984). Development of an objective measure
to assess ego identity in adolescence: Validation and replication. Journal of
Youth and Adolescence, 13,419-438.
Hansen, R.D. (1985). Cognitive economy and commonsense attribution proces-
sing. In J.H. Harvey & G. Weary (Eds.), Attribution in contemporary psycholo-
gy (pp. 65-85). New York: Academic Press.
Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.
Hoffman, M.L. (1977). Moral internalization: Current theory and research. In L.
Hovland, C.I., Harvey, O.J., & Sherif, M. (1957). Assimilation and contrast
effects in reactions to communication and attitude change. Journal of Abnormal
and Social Psychology, 55, 244-252.
James, W. (1950). The principles of psychology. New York: Dover (originally pub-
lished 1890).
Jones, E.E., Rhodewalt, F., Berglas, S., & Skelton, J.A. (1981). Effects of
strategic self-presentation on subsequent self-esteem. Journal of Personality and
Social Psychology, 41,407-421.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ: Prentice-Hall.
Kelley, H.H. (1967). Attribution theory in social psychology. In D. Levine (Ed.),
Nebraska symposium on motivation (Vol. 15, pp. 192-238). Lincoln: University
of Nebraska Press.
Kelly, G.A (1955). The psychology of personal constructs. New York: Norton.
Kelman, H.C. (1958). Compliance, identification, and internalization: Three pro-
cesses of attitude change. Journal of Conflict Resolution, 2, 51-60.
Kuhn, T.S. (1962). The structure of scientific revolutions. Chicago: University of
Chicago Press.
Kuhn, T.S. (1970). The structure of scientific revolutions (2nd ed.). Chicago:
University of Chicago Press.
Lakatos, I. (1970). Falsification and the methodology of scientific research pro-
grammes. In I. Lakatos & A Musgrave (Eds.), Criticism and the growth of
knowledge (pp. 91-195). Cambridge, England: Cambridge University Press.
Langer, E.J., Blank, A, & Chanowitz, B. (1978). The mindlessness of ostensibly
thoughtful action: The role of "placebic" information in interpersonal inter-
action. Journal of Personality and Social Psychology, 36, 635-642.
Levinson, D.J. (1978). The seasons of a man's life. New York: Ballantine.
Looft, W.R. (1973). Socialization and personality throughout the life span: An
examination of contemporary psychological approaches. In P.B. Baltes & K.W.
Schaie (Eds.), Life-span developmental psychology: Personality and socialization
(pp. 25-52). New York: Academic Press.
Lord, C.G., Lepper, M.R., & Preston, E. (1984). Considering the opposite:
A corrective strategy for social judgment. Journal of Personality and Social
Psychology, 47, 1231-1243.
Lord, C.G., Ross, L., & Lepper, M.R. (1979). Biased assimilation and attitude
polarization: The effects of prior theories on subsequently considered evidence.

Marcia, J.E. (1966). Development and validation of ego identity status. Journal of
Marcia, J.E. (1967). Ego identity status: Relationship to change in self-esteem,
"general maladjustment" and authoritarianism. Journal of Personality, 35, 118-
133.
Marcia, J.E. (1976). Identity six years after: A follow-up study. Journal of Youth
adolescent psychology (pp. 159-187). New York: Wiley.
Marcia, J.E., & Friedman, M.L. (1970). Ego identity status in college women.
Journal of Personality, 38, 249-263.
Markus, H. (1977). Self-schemata and processing information about the self. Jour-
nal of Personality and Social Psychology, 35, 63-78.
Markus, H., & Sentis, K. (1982). The self in information processing. In J. Suls
(Ed.), Psychological perspectives on the self (Vol. 1, pp. 41-70). Hillsdale, NJ:
Erlbaum.
McAdams, D.P. (1985). Power, intimacy, and the life story: Personological inquires
into identity. Homewood, IL: Dorsey.
Miller, G.A., Galanter, E., & Pribram, K.H. (1960). Plans and the structure of
behavior. New York: Holt, Rinehart & Winston.
Neisser, U. (1976). Cognition and reality. San Francisco: Freeman.
Nisbett, R., & Ross, L. (1980). Human inference and shortcomings of social judg-
ment. Englewood Cliffs, NJ: Prentice-Hall.
Nisbett, R., & Wilson, T.D. (1977). Telling more than we can know: Verbal re-
ports on mental processes. Psychological Review, 84,231-259.
Petty, R.E., & Cacioppo, J.T. (1979). Issue-involvement can increase or decrease
persuasion by enhancing message-relevant cognitive responses. Journal of Per-
sonality and Social Psychology, 37, 1915-1926.
Petty, R.E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persua-
sion. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol.
19, pp. 123-205). New York: Academic Press.
Petty, R.E., Cacioppo, J.T., & Goldman, R. (1981). Personal involvement as a
determinant of argument-based persuasion. Journal of Personality and Social
Petty, R.E., Harkins, S.G., & Williams, K.D. (1980). The effects of group diffu-
sion of cognitive effort on attitudes: An information processing view. Journal of
Piaget, J. (1964). The development of mental imagery. In R.E. Ripple & V.N.
Rockcastle (Eds.), Piaget rediscovered (pp. 21-32). Ithaca, NY: Cornell Uni-
versity Press.
Popper, K.R. (1970). Normal science and its dangers. In I. Lakatos & A. Musgrave
(Eds.), Criticism and the growth of knowledge (pp. 51-58). Cambridge, England:
Cambridge University Press.
Read, D., Adams, G.R., & Dobson, W.R. (1984). Ego-identity, personality and
social influence style. Journal of Personality and Social Psychology, 46, 169-177.
Rhodewalt, F., & Agustsdottir, S. (1986). Effects of self-presentation on the phe-
nomenal self. Journal of Personality and Social Psychology, 50, 47-55.
Rogers, T.B. (1981). A model of the self as an aspect of the human information
system. In N. Cantor & J.F. Kihlstrom (Eds.), Personality, cognition, and so-
cial interaction (pp. 193-214). Hillsdale, NJ: Erlbaum.
Ross, L., Lepper, M.R., & Hubbard, M. (1975). Perseverance in self-perception
and social perception: Biased attribution processes in the debriefing paradigm.
Schachter, S., & Singer, J. (1962). Cognitive, social, and physiological determi-
nants of emotional state. Psychological Review, 69,379-399.
Schank, RC., & Abelson, RP. (1977). Scripts, plans, goals, and understanding:
An inquiry into human knowledge structures. Hillsdale, NJ: Erlbaum.
Schenkel, S., & Marcia, J.E. (1972). Attitudes toward premarital intercourse in
determining ego identity status in women. Journal of Personality, 40,472-482.
Schlenker, B.R (1982). Translating actions into attitudes: An identity-analytic
approach to the explanation of social conduct. In L. Berkowitz (Ed.), Advances
in experimental social psychology (Vol. 15, pp. 193-247). New York: Academic
Press.
Schlenker, B.R. (1985). Identity and self-identification. In B.R Schlenker (Ed.),
The self and social life (pp. 65-99). New York: McGraw-Hill.
Sherif, M., & Sherif, C.W. (1967). Attitude as the individual's own categories: The
social judgment-involvement approach to attitude and attitude change. In C.W.
Sherif & M. Sherif (Eds.), Attitude, ego-involvement, and change (pp. 105-139).
New York: Wiley.
Sherman, S.J. (1980). On the self-erasing nature of errors of prediction. Journal of
Personality and Social Psychology, 39, 211-22l.
Shiffrin, R.M., & Schneider, W. (1977). Controlled and automatic human informa-
tion processing: II. Perceptual learning, automatic attending, and general
theory. Psychological Review, 84, 127-190.
Showers, C., & Cantor, N. (1985). Social cognition: A look at motivated strategies.
Annual Review of Psychology, 36, 275-305.
Slugoski, B.R, Marcia, J.E., & Koopman, RF. (1984). Cognitive and social
interactional characteristics of ego identity status in college males. Journal of
Snyder, M. (1979). Self-monitoring processes. In L. Berkowitz (Ed.), Advances in
experimental social psychology (Vol. 12, pp. 85-128). New York: Academic
Press.
Stryker, S. (1968). Identity salience and role performance. Journal of Marriage and
the Family, 30, 558-564.
Suppe, F. (1977). The structure of scientific theories (2nd ed.). Urbana, IL: Uni-
versity of Illinois Press.
Swann, W.B., Jr. (1985). The self as architect of social reality. In B.R. Schlenker
(Ed.), The self and social life (pp. 100-125). New York: McGraw-Hill.
Trudeau, J.V. (1986). Internalization of self-presentations: Passive and active pro-
cesses. Unpublished masters thesis, University of Florida, Gainesville, FL.
Vaillant, G. (1977). Adaptation to life. Boston: Little, Brown.
Waterman, A.S. (1981). Individualism and interdependence. American Psycho-
logist, 36, 762-773.
341-358.
Waterman, A.S. (1984). Identity formation: Discovery or creation? Journal of

Early Adolescence, 4,329-341.
Waterman, A.S. (1986). Identity formation, metaphors, and values: A rejoinder to
Berzonsky. Journal of Early Adolescence, 6, 119-121.
Wicklund, R.A., & Brehm, J.W. (1976). Perspectives on cognitive dissonance.
Wilson, T.D. (1985). Strangers to ourselves: The origins and accuracy of beliefs
about one's own mental states. In J.H. Harvey & G. Weary (Eds.), Attribution in
contemporary psychology (pp. 9-36). New York: Academic Press.
Wilson, T.D., Dunn, D.S., Bybee, J.A., Hyman, D.B., & Rotondo, J.A. (1984).
Effects of analyzing reasons on attitude-behavior consistency. Journal of Per-
sonality and Social Psychology, 47,5-16.
Wilson, W.R. (1979). Feeling more than we can know: Exposure effects without
learning. Journal of Personality and Social Psychology, 37, 811-821.
Winograd, T. (1975). Frame presentations and the declarative-procedural con-
troversy. In D. Bobrow & A. Collins (Eds.), Representations and understanding
(pp. 185-210). New York: Academic Press.
13
To Be or Not: Self and Authenticity,
Identity and Ambivalence
ANDREW J. WEIGERT
What does it profit a person to gain the whole world but suffer the loss of
one's soul? Jesus's dictum can be translated into contemporary idiom sim-
ply by changing one word: What does it profit a person to gain the whole
world but suffer the loss of one's self? Indeed, in the moral discourse of
Western thought, some sense of selfhood may be the nearest empirically
available basis for a sense of personal transcendence (Harre, 1984). Surg-
ing social scientific attention given to the concept of self after its near aban-
donment during the positivistic heyday suggests that the phenomenon to
which it refers may be a historical and cultural universal (cf. Marsella,
DeVos, & Hsu, 1985). In the modern context, however, we find the con-
struct identity working as a competing catch-all scientific and folk term to
refer to what we take to be the unique human experience of self as self-
consciously known (Weigert, Teitge, & Teitge, 1986). We briefly discuss
self and identity in order as we present a perspective on contemporary
self-understanding.
Self and Identity

Even if a kind of universality is granted to the phenomenon of self-aware-
ness, we assume that the social form and historical content of the self are
derived sociohistorical productions (cf. Wiley, 1979). The scope of this
paper limits us to a heuristic indication of what we take to be a generally
shared understanding of the historical development of self-conscious
reflexivity, personal responsibility, pluralistic relativism, and individuality
as characteristics of a Western sense of self or "selfhood." Others have
traced the selfhood of collectivism in the Judaic Old Testament, of indi-
vidual moral responsibility in Socrates, of the call of personal conversion in
the Christian New Testament, and in the vocation to individual salvation in
the Protestant Reformation (e.g., Weber, 1958). There is also an "egres-
sion" (to avoid the normative pitfalls of "progression") of self-reflexivity in
the philosophical traditions, illustrated by Socrates, Epictetus, Augustine,
264 Andrew J. Weigert
Goethe, and Sartre. Finally, we can see the emergence of pluralistic

relativism from the Medieval synthesis through the Enlightenment, and on
to the explosion of contemporary empiricism and perspectivism.
Within these formal aspects of selfhood (Le., collectivism versus indi-
vidualism, externalism versus reflexivity, certainty versus relativism), we
can discern historical types of selfhood: Old Testament creatures of God
and members of God's chosen people, inhabitants of city-states, citizens of
imperial states, individual merchant entrepreneurs and producers of eco-
nomic power, and current manipulators of financial markets and consum-
ers of products and services. Each of these historical types appears atop a
substratum of slave, serf, peasant, proletariat, or underclass that embodies
the nether side of the egression of what we may call, following Weber, the
dominant cultural "carrier" selves, (i.e., types of self based on the central
values of the society). Society has empirically available definitions awaiting
the accident of birth to impose a set of identities and locate us within its
strata. The reflexive, relativized, and individualized contemporary self
struggles with the empirical givens perhaps more fiercely than any other
self. Such is what we take to be the particular pathos underlying this
chapter.
This brief depiction of the formal and historical aspects of selfhood sug-
gests the sociohistorical context that we are assuming. The story follows
from our sociological and integrative purpose. We try to link self and
identity with concepts from philosophical and psychoanalytic origins and
from the sociology of religion, namely, authenticity, ambivalence, and
eschatology. Such an integrative intent leaves the paper relatively unde-
fended at the boundaries and relies for its effectiveness on the good will of
the reader, which we assume as well.
With friendly indulgence, we define a presumably essential aspect of the
contemporary phenomenon: self is that aspect of the actor which is aware of
self as transcending the empirical situation (cf. Weigert, 1975). The appa-
rent circularity refers to what is assumed to be an experiential datum,
namely, the concomitant awareness that persons have of self in action: "I
know what I am doing and that I am doing it." This is the self-reflexivity
that characterizes our experience, at least potentially. We assume that
concomitant awareness is available to all competent persons, and that the
frequency and content of such awareness differ historically. The variation
derives from what we assume is the necessary nesting of concomitant aware-
ness within a context of moral discourse that frees the imagination from the
determinisms of here, now, and this (a Durkheimian premise found in the
work of Bellah, Madsen, Sullivan, Swidler, & Tipton, 1985). Furthermore,
we assume that moral discourse involves a sense of the future that is
grasped symbolically. Self-awareness, then, implies a symbol of a neces-
sary, but not necessarily empirically based, human future, or what we shall
can an eschatology. Fueling our concern is the conviction that the contem-
porary world context renders "End Time" thinking particularly relevant
for sacred as well as secular minds (Mojtabai, 1986). Our view is that self is
13. Self and Authenticity, Identity and Ambivalence 265
aware of self as transcending the immediate situation by contextualizing

experience within a moral discourse that legitimates events in terms of
a symbolically implicated future. Self is historical, but derivatively and
indeterminately, since it is derived from a future that, though it is not yet,
exists as varied and indeterminate symbolizations that are here-and-now
empirical components of self-understanding.
Identity, by contrast, we see as directly historical: derived from the past
and institutionalized or enacted in the present. As such, the contents of
identity are direct products of historical processes. Identity results from
empirical referential discourse and gestures within which it is constituted.
Building on self as a more basic concept, we may define identity as a
typified self situated in a network of social relationships (Weigert et aI.,
1986, p. 53).
The first two words in the definition give us the further distinction be-
tween social and personal identity. Social identity locates self as a product
within the empirically available set of social definitions. As Berger and
Luckmann (1966) said, social identity is a human type that is totally a social
production. It is socially determined. Personal identity, on the other hand,
is a definition that takes account of self's imposed, aware, or remembered
relationship to a particular bodily experience and biographical situation.
Within the logical parameters of philosophical thought, the basis of person-
al identity remains an unresolved issue, for example, between dualists and
materialists (Shoemaker & Swinburne, 1984). In what phenomenologists
call the natural attitude of everyday experience, however, the uniqueness
of an individual's body provides a normally irrefutable and unreplicated
base for personal identity (cf. Schutz, 1962). Further linkages among per-
sonal, group, and cultural identities are provided by social institutions,
with religion as a traditional and powerful instance (Mol, 1976). Within
Western religious discourse, a conception of an ultimate destiny or "end"
of life-that is, eschatology-is central.
The paradoxical link of a public social definition and a private bodily
experience presents philosophers with an antinomy, but it is a routine,
taken-for-granted understanding for "normal" individuals. We interpret
personal identity as directly historical but partly so, since it includes bio-
graphical meanings as well. It is, as C.W. Mills characterized sociology, at
the intersection of history and biography, of sociocultural forces and
psycho-organic processes. Social identities remain relatively determined by
history, whereas personal identities remain somewhat indeterminate. His-
tory, therefore, impacts differently on the individual through each mode of
identity: directly and deterministically through social identities, directly
and partially through personal identities, and indirectly both formally and
materially through the kind of selfhood empirically available and the re-
sponses that selves make to the empirical apriorities. Self through concom-
itant awareness remains potentially substantival, though derivatively his-
torical, while identity as socially defined is directly historical (see Weigert,
1975, for this sense of "substantival").
Integrative Leads
To grasp the phenomena of self and identity, we adopt an integrative his-
torical and cultural perspective. Main-line positivistic, experimental social
psychological work in itself is not adequate to the task. Clearly, social
psychology is not identical with history, but neither is it totally separate
from history. There is an essential historical dimension to the subject mat-
ter and methodology of social psychology (see Gergen, 1973; Gergen &
Gergen 1984; Sampson, 1983).
The general approach that we take may be called "pragmatic social con-
structionism" (Weigert et al., 1986, pp. 2-3). The underlying assumption is
that meanings, personal as well as social, are produced and sustained in
the processes of symbolic interaction (see Blumer, 1969; the basic work of
G.H. Mead). The primal source of situated meanings is in the responses
that self and other make to the action, whether it be a gesture, word, or
silence. Nevertheless, once institutionalized, patterns of responses and the
symbolic representations take on the qUalities of objective facts, that is,
"facticities" that appear to be naturally occurring objects but which in fact
are precarious dramatic effects of human interaction. As facticities, these
meanings can be internalized by members of the community and shape
their experience and self-interpretation. Life, then, is conceptualized as a
historical, dialectical, and collective process producing the institutions that
function as the context of fateful facts shaping biography (cf. Featherman
& Lerner, 1985; Kegan, 1979).
History and social psychology share a concern for process, for the
intersection of structure and biography, and for the task of formulating a
more general cultural meaning relevant to the particulars of life. There is
an homology between them. We may suggest that history is to societies as
social psychology is to individuals; for example, societies may be concep-
tualized as civilizational actors and individuals as social selves. The integra-
tive intent of this chapter renders it rather broad and sketchy, but we feel
that the central direction is consistent: How are we to grasp the person with
adequate concepts and relevant historical sensitivity? Building on our ear-
lier statement that moral discourse is the context for self and identity, let us
limn a picture of communication with an eye toward the modern scene.
Self, Identity, and Communication

How do we gain historical understanding of self and identity? The general
answer is that we must access people's lives in the fundamental categories
of what they do and say. Deeds and words mediate persons' externalization
and typification of themselves (cf. Berger & Luckmann, 1966). For those
who interpret, doing and saying narrate who persons are and what they are
about (Gergen & Gergen, 1983). In every situation, there is an identity at
stake and a business at hand: who appears and what is done (Weigert,
1983). The identity and the business are embodied in the expressions that a
person intentionally gives and ambiguously gives off either knowingly or
not, and in the dramaturgical constructions, both material and mental, pro-
duced by self and other (Goffman, 1959).
We externalize ourselves in actions as processes and we objectify our
identities in the products of those processes. We are, at one and the same
time, a process and a series of products, a self and sets of multiple identi-
ties. Analysis of the process does not allow us deterministically to deduce
the products; neither does a positivistic analysis of the products allow us
mechanistically to measure the process. Access to self is along different
generalizable "paths," the etymological image of the Greek meth'od, than
those leading to identities: self may be better known through systematic
introspection or thematic cultural analyses, for example, whereas identities
may be explained as structural and dramaturgical effects measured positi-
vistically.
The analogy can be extended to the standard distinction of language as
an objectified abstract structure and an externalized concrete process, as
both language and talk. We must be careful in this linguistic turn, however,
not to reify language such that it swallows the subject. It has been argued
that language exhausts meaning, and that there is no such independent
agency as the speaker. We assume, however, that there is a self that is
ontologically prior, though not temporally prior, to the existence and use
of symbol systems, with language as the prime exemplar. We take the case
of Helen Keller as the best documented instance of this ontological priority
but temporal simultaneity or succession of the self. Language as a system
remains a social emergent irreducible to the talk of individuals, but selves
remain ontologically prior to the social and irreducible to the language
within which they are expressed (cf. Weigert, 1975).
As an experiential and reflexive reference that is irreducible to the form
and content of language, self retains a substantival status that transcends
situations. Nevertheless, self as known and communicated is derivatively
constituted by and implicated in the symbol systems available in the actor's
historical situation. We can tell only those self and other narratives that are
available for understanding who we are, what we are doing and why, and
what we ought to do (cf. Gergen & Gergen, 1984, ch. 9). At the heart of
personal awareness, therefore, is a basic dualism: self as transcendent and
self as empirically constituted. This dualism is further complicated by the
dualism of symbol systems. By analogy with language, we think of symbol
systems as bifurcated into an underlying logic of meaningful constructions
and an empirically instantiated content. Self, then, is dualistic in at least
these two senses: as transcendent and situational, and as logical form and
communicative content.
Identities, on the other hand, are eventually reducible to empirical in-
stances or series of situations. As typified selves, identities share the basic
and paradoxical dualisms of self. As historical entities, however, they are
indefinitely pluralistic or restrictedly monistic according to the empirical
potentialities of the historical context. Whereas self in principle is not

exhausted by either the form of language or the tales of talk, identity can
be captured in empirical constructions. We can have only those socially
effective identities that we can construct in our conversations with others.
Identities as well as our moral sense are realized in stories. Analysis of the
types of identities requires the study of the social organization and social
psychology of these stories and their narrators in the historical context (cf.
Habermas, 1979; Sampson, 1983).
Standard sociological analyses interpret contemporary society as special-
ized, pluralistic, and rapidly changing. Culturally, the modern world is
seen as relativistic, situational, and de- or re-mythologizing. The identity
structure of typical moderns is assumed to reflect the social and cultural
organization of contemporary society. The modern individual possesses
multiple identities relevant to the variety of primary and secondary roles to
be performed. The multiplicity of identities tends to lack a single centering
and master identity that can function as an organizing principle for all
others through an elongated lifetime. Moderns must, therefore, be skilled
at protean arts and mutable presentations of self (Goffman, 1959; Lifton,
1976; Zurcher, 1977). Identity forms, contents, and skills are historical
processes or products, and history suggests that today's context is one of
identity crisis.
Crisis in Self-Understanding
We wish to discuss two aspects of the contemporary crisis in self-under-
standing. First, focusing on self, there is a search underway for an adequate
interpretation of the meaning of existence from new religious movements
to the resurgence of fundamentalist literalism. The search eventually
comes to an analysis of everyday life and the moral discourse available for
making sense of that life. Analysts struggle with depictions of traditional
understandings versus an "emotivist" formulation of self that interprets
self-references as expressions of mere opinion or personal preference
(MacIntyre, 1984). Others see contemporary society generating dialectical
experiences of a self that is divided against itself and also set in opposition
to the other (e.g., Laing, 1965). More focused analyses find that the Amer-
ican self seeks understanding from a variety of Puritan, civic, utilitarian,
narcissistic, and pragmatic perspectives (Bellah et al., 1985; Bercovitch,
1975; Lasch, 1979; Niebuhr, 1963).
The second aspect of the crisis concerns identity (see Baumeister, 1986;
Klapp, 1969). The availability of both traditional identities and a variety of
temporary identities is one of today's defining characteristics of adulthood.
The "multiple identities" condition results at least in part from two
changes in the social psychological context. First, there are totally new
identities available, not only in the realm of secondary role identities like
occupations such as space worker or astronaut, but also in the realm of

universal and intimate primary domains like sex. For the first time in his-
tory, a person can change sex as well as gender in a near-total anatomical,
hormonal, psychological, and social sense. At least one implication of this
newfound availability of sex change is to move gender identity out of the
God- or nature-given category into the realm of yet another personal
choice. In the modern world, even my sexual identity is a question of
personal responsibility. This locates responsibility at the heart of personal
experience in a way that has never been possible before.
Second, there appears to be an increasing variety of secondary identities
available and an increasing number of persons who live long enough to try
on a greater number of identities over a lifetime. Prima facie evidence for
these changes comes from increasing longevity; increasingly specialized
occupational categories; larger numbers of persons, including the recent
combination of wife and mother in the work market; and changes in
intimate homosexual, interethnic, and interreligious-married individuals.
So-called deviant identities in the worlds of neighborhood crime, sex,
drugs, and the high-wired worlds of political and financial white-collar
crime attract more individuals who also wear traditionally respected identi-
ties or acquire others that are admired locally. With no attempt at compa-
rative historical or cross-cultural analysis, it seems reasonable to assume
that there are now more attainable or imposed identities extant in modern
society than in the past. Modernity implies identity pluralism as well as
mobility.
What the change and increase in the effective demand for available iden-
tities does is render those aspects of self-understanding that individuals
translate into personal identities more objective, relativistic, exchange-
able, transient, and, introducing a theme that we discuss below, inauthen-
tic. To use a distinction from Max Weber, modern identities are more a
matter of instrumental rationality than ideological or value rationality. The
consciousness, self-understanding, and experience of moderns is trans-
formed into the category of object by sociocultural changes in the constitu-
tion and organization of identities (cf. Simmel's classic essay on the social
psychology of the metropolis, 1964). The great transformations of mod-
ernity are social psychologically translated through personal identities. We
now examine two terms that have recently gained currency for understand-
ing the modern self.
Authenticity and Ambivalence

Interpretors develop concepts as tools for making sense out of issues that
challenge them. We have selected two rather recent concepts, authenticity
and ambivalence, for furthering our analysis. Authenticity took on its mod-
ern cast over the last 70 years or so. We use the term in a somewhat wide
sense to refer to self and the sense of historical reality that envelopes it. A
person's sense of authenticity functions at a prereflective stage in the pro-
cess of self-understanding, and it gives verisimilitude to the conviction that
we transcend our brute organic existence (Tiryakian, 1968). In the natural
attitude of our daily lives, the assumption of an authentic sense of existence
does its work unnoticed, unquestioned, and unformulated (cf. Husserl,
1970; Schutz, 1962). Existentialists use authenticity more specifically to
refer to the human condition as a project of moral freedom and to the
necessity of creating meaning through personal decisions. We see this
creative "decisional" idea of authenticity as a special critical experience of
the more pervasive sense of myself as an authentic actor in a life that is
worth living.
The second term, ambivalence, grasps an experience seemingly charac-
teristic of human life, even though it was conceptually "discovered" when
the term was coined at the turn of the century. Ambivalence was de-
veloped within the psychoanalytic movement to interpret experience typi-
cal of the then-modern family and, theoretically, of all emotional life (e.g.,
Freud, 1913/1964). The term captures the mixed emotions of the simul-
taneous push-pull of affective experience. Freud applied it psychoanalyti-
cally to the copresence of love and hate, one in the conscious and the other
in the unconscious life. Later, he used it to speculate about instinctual
affection-aggression in primitives as well as about cultural forces for life
and death in moderns. Robert Merton adapted the term for sociological
analysis of the tensions between institutional structures and individual
adjustments in roles with contradictory norms, and implicitly to the contra-
dictory demands of interaction and culture more generally (see Weigert &
Franks, 1986). It is in this wider sense that we use the term here to cap-
ture a definitive feature of contemporary culture: contradiction. Let us
take a closer look at each concept in turn, beginning with authenticity.
Within modernity, the meaning of a person's life is a problematic con-
struction. In his analysis of Western literature, for example, Trilling (1972)
argues that there is an egression from the "sincerity" of Renaissance
heroes who struggle to be true to who they are, to the "authenticity" of
moderns who struggle to show that they are indeed who they appear to be
and are true persons in the face of today's fragmented selves. This literary
analysis is paralleled by Turner's sociological speculation about the his-
torical shift in the location of a "vital sense of a real self" from social in-
stitutions to individual impulses (1976, p. 1005). Existentialists diagnose
the modern self as fragmented, searching, shipwrecked, and thrown into a
world that appears absurd because those who live in it neither made it, nor
chose it, nor understand it.
The modern condition, then, generates a personal and collective search
for the meaning that derives from being aligned with the moral forces that
make life worth living and the person matter for something, a summum
bonum (Klapp, 1969). The person seeks metaphors of meaningful self-
hood, and these metaphors are socially produced and maintained as plausi-
ble (cf. Berger, 1967; Smith, 1985). It is this cultural configuration of ideas,
values, and experience that we are trying to grasp with the notion of
authenticity.
Gecas (1986), for example, recently introduced authenticity into socio-
logical social psychology as a master motive in the domain of the self-
concept. He argues that there are three "self-motives," which he calls self-
esteem, self-efficacy, and authenticity. The first two are well developed in
social psychological literature under a number of rubrics, but he sees a
need to introduce a concept that he admits is underdeveloped and loosely
formulated (see earlier papers on contexts in which adolescents feel
authentically themselves, 1971, 1972). He describes authenticity as "an
assessment of the meaning and significance of what one is and does" and as
focusing "on the motivational implications of beliefs, especially beliefs
about self" (1986, p. 141).
We agree with Gecas's emphasis on such a self-motive and think that it is
characteristic of the modern context. Others have independently alighted
on this issue. Marshall Berman's (1970) interpretation of authenticity arose
during the heady political turmoil and idealism of the 1960s. He reflects a
rather optimistic solution to self-fragmentation: Get politically involved!
He finds the problem of, as well as a solution to, authenticity already
outlined in 18th Century France, namely, build a participatory political
system responsive to the needs of members and thus balance the tension
between individual fulfillment and community responsibility. He sounds a
note of abstracted atomism, since he conceives of authenticity as being true
to one's unique self, as if such a self exists or is known apart from social and
transcendent content.
In a more recent sociohistorical analysis that fits the conservative and
atomistic utilitarian mood of the 1980s, the psychologist Roy Baumeister
(1986) concludes that the current predicament of the self is structured
around the felt tension between an outer context that is real, labile, and
social psychologically trivial, and an inner self that is a complex mix of real
elements and illusory "constructs." He states that "the question of authen-
ticity is a question about the size and scope of the self. The emergence of
this question early in this century is consistent with the general devaluation
of the self ... " (p. 93). The struggle for identity is a particularly modern
problem and one that is largely insolvable. He ends on a note of deep
pathos concerning the split at the center of self-understanding.
Finally, in a dramaturgical analysis of service occupations, such as airline
attendants, Hochschild (1983) argues that authenticity has become a moral
concern with the rise of "emotional labor" in which workers' feelings are
part of the production and thus, like other tools and products, become
alienated from the workers. If emotions become an alienated product, the
worker is alienated from the experiential heart of self. My feelings are no
longer me! In a desparate struggle, workers, mainly women, fight to regain
their experiential heart. Authenticity arises as a personal virtue defending

self against the "corporate use of guile and the organized training of feeling
to sustain it" (p. 192). For alienated attendants, authenticity refers to
"spontaneous, natural, artless feeling" (p. 264). Only if an individual be-
lieves that personal feelings are authentic can he or she have a sense of a
true self behind the masks and roles, since it "is from the feeling that we
learn the self-relevance of what we see, remember, or imagine" (p. 196,
see also Denzin's [1984] discussion of "self feeling"). We would add that
self validates personal feelings via socially sustained narratives in which
they are integrated into a larger scheme of events. Even then, certainty
eludes moderns, who appear to remain ambivalent. Let us add a few words
to clarify this idea.
We note that ambivalence is discussed at three levels of analysis with
different empirical references. First, Freud used the term psychoanalytical-
ly to refer to the experience of mixed emotions flowing from contradictory
feelings, one conscious and the other unconscious. The experience results
from the dynamics of parent-child relations within the family drama. This
type of explanation requires the analyst to accept the construct of uncon-
scious feelings, a move that appears contradictory and difficult to render
empirical. Second, Freud spoke of what he called "instinctual" ambiva-
lence, which referred to the simultaneous existence of aggressive and
affiliative impulses in everyone, especially primitives, and the cultural
translation of these impulses into engines of love and death in contempo-
rary societies. Such a general model remains a hypothetical construction
based on a questionable appeal to universal instincts.
A third and sociocultural use, however, lends itself to analysis and
empirical investigation. Sociologists have used ambivalence to help explain
the genesis of modern social institutions (Elias, 1982), and the functional
albeit tense adaptation to contradictions in modern institutions. Merton
began codification within sociology with a definition of "core" ambivalence
as "conflicting normative expectations socially defined for a particular social
role associated with a single social status" (1976, p. 8). This definition or an
analogue illumines various social psychological issues including: the clash
of traditional ways of life threatened by technologically based change
(Marris, 1975); the crisis of traditional religion confronted with the contra-
dictions of modern thought (Kurtz, 1986); paradoxical situations generated
by contradictory norms of rationality and loyalty in the conduct of scien-
tists (Mitroff, 1974); contradictory meanings such as solidarity versus duty
within interpersonal relations (Eisenstadt & Roniger, 1984) or between
professionals and their clients (Merton, Merton, & Barber, 1983); and the
contradictory values associated with culture itself, the demands of success-
ful interaction, and the anxiety of contemporary occupational and other
identity crises (see Weigert & Franks, 1986, for further discussion). The
next step is to discuss possible responses to ambivalence. The tack we take
concerns attitudes toward the future coupled with implications of authen-
ticity and ambivalence for self-understanding.
Existential and Eschatological Responses: Implications

for Self and Identity
If we accept ambivalence as characteristic of modernity, we can ask what

are the likely implications for self-understanding. Two lines of reasoning
lead to reinforcing conclusions. Both assume that eventually ambivalence
is experienced as a painful, unsettled, anxious condition pushing toward
resolution. This is not a cognitive consistency argument, but an argument
from emotional "dis-ease." The first line of reasoning we call existential,
based on Jean Paul Sartre's argument that to live authentically in the face
of real crisis, a person must eventually recognize the freedom to create
moral good through personally responsible decisions in an absurd world.
Existentially authentic individuals realize that they live in a pluralistic uni-
verse in which moral good is not given a priori nor empirically, but is
created ever precariously by the free decisions of engaged persons. Full
flush with this realization comes the experience of existential anguish in the
face of the awesome responsibility not only to do the good, but to create
through choice the good to do in the first place.
Often torn between contradictory actions, like the young man who both
wants to stay home and protect his mother and longs to join the resistence
forces and fight the Nazis (Sartre, 1957, p. 24ff), the person feels responsi-
ble for a decision, ambivalent about competing goals, and as a result, suf-
fers deep anguish. Ambivalence and moral abulia lead the individual to ask
the seminal identity question again, "Who am I really?" in order to decide
what to do. Action follows identity. Probing the seminal identity question
leads the individual to search for an authentic moral identity within a life
narrative that is a socially constructed and mythic guide (cf. Charme, 1984,
pp. 125, 157; Zimmerman, 1981).
The second line of reasoning we call eschatological. By eschatologies,
we refer to socially constructed attitudes toward or images of a future that
provides an end-state for the individual, group, society, world, or cosmos
(Weigert, 1987). They vary in scope, mode, and content; may be religious
or secular; and are rendered plausible by inclusion in group life. Under-
stood as a religious category, eschatology helps answer the central religious
question emphasized by Max Weber, "Am I saved?" As mentioned ear-
lier, a recent observer refers to it as End Time thinking (Mojtabai, 1986).
We wish to suggest a link between ambivalence and the functioning of
contemporary versions of End Time thinking.
Consider that there is some experimental evidence to suggest that per-
sons with ambivalent attitudes toward a stigmatized population such as
minorities or handicapped persons react more negatively toward them than
persons with a more decidedly negative attitude (Katz, 1981). This sup-
ports the folk wisdom that undecided or torn individuals may overreact and
embrace one alternative with excessive zeal. Ambivalents may resolve
their anxious condition by overly loving one of the emotional objects and
overly hating the other, or adopting one emotional attitude to the seeming
exclusion of the other. Fixing on one emotional object or being dominated
by one emotional attitude, however, requires an account that legitimates
such action and allows the ambivalent to present a viable identity to self
and others. It is in the construction of an account that the eschatological
card comes into play: the individual is likely to fall back on an internalized
vocabulary of motives and world view that evokes traditional moral dis-
course and appeals to imagined futures (cf. Bellah et aI., 1985; Zurcher,
1983).
Recent American attitudes toward the "Bomb" and the nuclear arms
race with the Soviet Union, for example, are apparently ambivalent: each
evokes feelings that imply it should be simultaneously destroyed and toler-
ated, if not embraced (Yankelovich & Doble, 1984). For one aggregate in
the poll who think that the Communist and nuclear threat is very great, the
solution is to embrace a traditional religious answer that gives superior
moral worth to a Christian identity and to American policy and attitudes.
Compared to 56% of the sample, this group has 78% agreeing with the
statement, "The Soviet Union is like Hitler's Germany-an evil empire
trying to take over the world." Compared with 40% of the sample, 77% of
this aggregate agree that "When all is said and done, someday the U.S. will
have to fight the Soviet Union to stop communism." Indeed, 57% of these
respondents believe that the likelihood of a nuclear war is (very or fairly)
great in the next 10 years, compared to 38% of the sample. Many of these
respondents have reasons to be ambivalent toward the Bomb.
This aggregate also have an eschatological solution to their tom feelings.
Their solution takes the form of fundamentalist religious beliefs stated in
traditional evangelical discourse that takes the Bible literally. Fully 67% of
these respondents, compared to 39% of the sample, agree that "When the
Bible predicts that the earth will be destroyed by fire, it's telling as about a
nuclear war." Such literal acceptance of Biblical prophecy indicates a
transcendent supernatural frame for interpreting events that locates causal
reality outside the workings of human agency (ct. Johnson & Weigert,
1980).
The supernatural frame is applied in response to another item. Although
only 18% of this aggregate believe that there is "at least a 50-50 chance of
personally living through a nuclear war," fully 55% of them believe that
"In a nuclear war with communists, our faith in God would ensure our
survival," compared to only 26% of the sample. This solution places
American or Fundamentalist identities, motives, and actions in a priv-
ileged moral position; embraces a transcendent supernatural definition of
the situation; and thus is able to find a meaning that overcomes ambiva-
lence. An extended version of this kind of thinking is found in the best-
selling book of the 1970s, Hal Lindsey's, The Late Great Planet Earth
(1977, p. 156), in which he states that "As the battle of Armageddon
reaches its awful climax and it appears that all life will be destroyed on
Existentialist Authenticity
Pluralist
~~~~r;- Ambivalence- Anxiety a \ n 9uish
" " Authenticity Free Moral Action Encompassing

with • Both Emotional Objects
Ambivalence
Eschatological Authenticity
Pluralist
Empirical . . . Ambivalence ...... Anxiety
\
Universe
Obedient Moral Authenticity Simplified

Action Embracing_ Without _Moral
One Emotional Ambivalence Universe
Object
FIGURE 13.1. Types of authenticity: self-understanding in response to ambivalence.
earth-in this very moment Jesus Christ will return and save man from
self-extinction .... " As long as the Lord is coming to save believers
through Rapture, there is no need to fear thermonuclear war (see Halsell,
1986; Mojtabai, 1986).
We can summarize the existentialist and eschatological responses to
ambivalence and the link to self-understanding by including self and au-
thenticity in the discussion. It seems that an historically sensitive social
psychology must recognize that the issues raised by the concepts of ambiva-
lence and authenticity are characteristic of modern society. We assume
that the modern self must learn to live with and through the media of
ambivalent identities in the larger historical quest for authenticity. The
thesis that emerges from our discussion is that the modern search for
alternative forms of authenticity is fueled by the experience and recogni-
tion of ambivalence. The reasoning is summarized in Figure 13.1.
The existentialist solution follows a classic stoic line of reasoning based
on the sharp distinction between those things "within our power" and
those "beyond our power." The wise person, then, changes SUbjective atti-
tudes toward the object while taking the objective world as it comes (Epic-
tetus, 1955). In this way, the person attains a kind of authenticity cum
ambivalence by transcending the situation through the agency of free mor-
al action encompassing the world as it is and trying to create a new and
unknown future. The self attains the identity of free, responsible moral
agent.
The eschatological resolution, on the other hand, comes in the anomaly
of a resurgent Biblical fundamentalism in the midst of a secular, utilitarian,
and pluralistic culture that is modern America (cf. Bellah et aI., 1985). It
reenacts a past solution and thus provides an ethnocentric mythic frame for
interpreting the current situation. This is a kind of collectivist romantic
resolution based on personal conversion to a privileged vision from the
past that becomes normative for the future. The person achieves an au-
thenticity sans ambivalence by transcending empirical referents through
obedience to an authoritative version of events that embraces a past defini-
tion that presumably predicts a determined future in a simplified moral
universe. The stoic existentialist and romantic eschatological kinds of
authenticity both transcend the present situation but in contrary ways: the
former in terms of creative moral agency struggling to create unknown
futures; the latter in terms of obedient moral action fatalistically accepting
a past version of a known future.
The existentialist must live pragmatically in a pluralistic world with the
results of personal action constrained within an individualistic moral
scheme that leaves the person's identities fundamentally normless, anomic,
and ambivalent now as before. The eschatologist, on the other hand, lives
in a simplified mythic world that enforces community norms and beliefs
demanding obedience to a moral scheme that encysts the members in a
collectivistic identity that is normative, alienated from the constructive
dialectic of producing meaning, and authenticated once and for all. The
conjunction of existentialist anomie and eschatological alienation is
meant to place the former outside the social solidarity characteristic of a
Durkheimian world view, and save the latter inside a taken-for-granted
world view imposed by membership in a community. The latter tends to
prevent the self from realizing the anthropological dialectic of the social
and self-construction of the very myths that bind us, whereas the former
attracts self to an overly creationist view of human agency (see Berger &
Luckmann, 1966).
Conclusion and Summary

We would like to summarize the argument. First, we conceptualize the
person as constituted by two levels of experience: the self with its aware-
ness of the unity and continuity of the person; and multiple identities com-
posing the pluralistic experience of modern persons. The self seeks a sense
of authenticity, that is, a conviction that self is in line with the real forces
underlying experience and history and transcends the empirical limits of
the situation. Through its identities, self realizes the moral pluralism and
confluence of ambivalent emotions diagnosed with the birth of modernity.
Modernity generates ambivalence. Eventually, ambivalence becomes a
condition of painful emotional "dis-ease" that intitiates a search for a
remedy in the form of a sense of authenticity. We suggest two types of
authenticity, one derived from the cultural past and the other generated
in the personal present. The former is eschatological and the latter is exis-
tential. Eschatological authenticity transforms the empirical world into my-
thic simplicity within which the individual acquires transcendent meaning
through a simple act of intentionality: I believe. Existential authenticity
engages the empirical world in its complexity such that the individual
struggles to transcend the present through a moral decision to create value
through behavior: I act. Eschatological authenticity alienates the indi-
vidual from the sources of personal creativity and integrates self into a
collectivist community identity through a myth that gives security to life, a
"biomyth" (cf. Hankiss, 1981). Existential authenticity leads the person to
take responsibility for the consequences involved in action and leaves him
or her in an anguished state of anomie.
At this point, we would like to note the formal similarity between the
earlier "egressions" of selfhood from collectivism, externalism, and
certainty to individualism, reflexivity, and relativism and the existential-
eschatological symbolic futures. 1 At first glance, there is an apparent com-
monality between the historical kinds of selfhood and eschatological and
existential authenticity. One could apply a developmental model and see
eschatological responses as reactionary pseudo authenticity and existential
responses as progressive genuine authenticity. From our sociological per-
spective, such a parallelism is an illicit normative move. There may be a
neutral developmental model applicable to individual life with its psycho-
organic imperatives, but we do not accept any easy translation of such
models to sociohistorical realities. Nevertheless, we do see a parallelism
between existential authenticity and individualism broadly conceived, and
eschatological authenticity and concern for community solidarity.
Solidarity is a function of moral discourse defining what a community
takes as finally right or wrong, and some kind of symbolic future is, as we
assumed for this chapter, an essential aspect of moral discourse. There
does appear to be a historical sequence that suggests parallels with indi-
vidual development, though we do not see any sociocultural form of self
understanding as absolute or exhaustive. Different forms may be more or
less functional for social organization or personal development, but our
view is that no one of them is functional for all of the goods that life re-
quires. The issue for the modern self is whether any individual or society
can balance the goods necessary for supporting individual development
and the institutional arrangements that are functional for sustaining human
life. The contemporary self faces, for the first time, responsibility for the
viability of the life-support systems themselves: a brand new set of deci-
sions for existentialist freedom and eschatological End Time thinking. In a
word, we are left with an ambivalent situation-the hallmark of modernity
from our point of view.
1 We would like to thank Dan Lapsley for suggesting this issue.

Eschatology is the production and legitimation of such futures, and thus,

a necessary part of a community's self-understanding. Within the general
category, however, we do find different kinds of eschatology such as liberal
and fundamentalist, basically the difference between working toward
unknown futures and literal belief in a prophecied chronology of future
events (Weigert, 1987). The fundamentalist version is an apocalypticism,
that is, a prophecied future that details present events. The symbolic is
interpreted as referring to a nonempirical, transcendental series of events
that have a knowable one-to-one application to the contemporary world.
The liberal version sees the symbolic as perhaps serving a nonempirical,
transcendent plan for the world, but with no empirical application to cur-
rent events. For the apocalypticist, the future is closed, determined by di-
vine causality and literally known through prophetic literature. The future
is grasped through present portents. For the liberal, the future is empirical-
ly open even though the moral purpose of events must serve some divine
plan that may be known in its broadest outlines. In contrast to both these
stances, the existentialist version sees the symbolic in a more pragmatic
mode as that which is incipient, promissory, projecting, and imaginative.
In this framework, the future is that which is constantly being created by
moral choice in the face of continual ignorance and ultimate absurdity. The
existentialist, like the eschatologist, seeks a moral self in terms of a wished-
for future, but the sense of life and the search itself is experienced in dif-
ferent symbolic orders. We see potential contradictions in efforts of any
moral self to decide and act authentically.
The implication of this discussion is that the modern self is faced with
finding social identities and forming personal identities that can coexist
with the inevitable ambivalence that accompanies attempts at self-under-
standing. At the same time, the modern self confronts the task of balancing
existentialist and eschatological authenticity. The modern must struggle to
formulate biomyths and narratives that allow self to transcend the empiri-
cal limits of the present situation in the context of community without
alienating self from the creative processes of constructing social reality. On
the other hand, the modern self must formulate creative courses of moral
action that do not leave him or her in an unreclaimed state of anomie, but
that allow for membership in a community as an integral part of selfhood.
As Jose Ortega y Gasset has written, human life results from a mix of
fate, chance, and character. From the self-identity perspective that we are
trying to develop, social identities are the historical fate we are born into;
personal identities are the chance that results from the accidents of organic
birth joined to the contingencies of historical context; and self is a reflexive
process that tries to shape the character that we are constantly becoming
but never finally are. The self must confront the primal ambivalence that is
part of the human condition as well as cope with the cultural ambivalence
that accompanies modernity.
The effort to resolve felt ambivalence, however, must steer a careful
course among the types of authentic existence that persons seek. There are
cheap biomyths carried by closed groups that provide a sense of selfhood
that functions authentically but that cannot generate a self authentically
engaging today's issues. There are, for example, the bravado "tough guy"
kinds of functional authenticity like that of the American cowboy, who
lives a biomyth with no community beyond his horse. Need we mention the
absurd biomyth emblazoned in Rambo? From historical perspective, we
could speculate that the founding Puritan self has been vulgarized into a
narcissistic self in contemporary America (cf. Bercovitch, 1975; Lasch,
1979). Meanwhile, the pluralism and strata of American society challenge
us to work through the identities that are thrust upon us or that we
ourselves seek. That predicament is highlighted by the dialectics of
ambivalence and authenticity. The modern, perhaps more than any other
self, lives a sense of "the end" and may end a sense of the living.
REFERENCES
Baumeister, R.F. (1986). Identity. New York: Oxford University Press.
Bellah, R.N., Madsen, R., Sullivan, W.M., Swidler, A., & Tipton, S.M. (1985).
Habits of the heart. Berkeley: University of California Press.
Bercovitch, S. (1975). The Puritan origins of the American self New Haven, CT:
Yale University Press.
Berger, P.L. (1967). The sacred canopy. Garden City, NY: Doubleday.
Berger, P.L., & Luckmann, T. (1966). The social construction of reality. Garden
City, NY: Doubleday.
Berman, M. (1970). The politics of authenticity. New York: Atheneum.
Blumer, H. (1969). Symbolic interaction ism: Perspective and method. Englewood
Cliffs, NJ: Prentice-Hall.
Charme, S.L. (1984). Meaning and myth in the study of lives. Philadelphia: Uni-
versity of Pennsyvania Press.
Denzin, N.K. (1984). On understanding emotion. San Francisco: Jossey-Bass.
Eisenstadt, S.N., & Roniger, L. (1984). Patrons, clients, and friends. Cambridge,
England: Cambridge University Press.
Elias, N. (1982). Power and civility. New York: Pantheon.
Epictetus (1955). The enchiridion. Indianapolis: Bobbs-Merrill.
Featherman, D.J., & Lerner, R.M. (1985). Ontogenesis and sociogenisis: Prob-
lematics for theory and research about development and socialization across the
lifespan. American Sociological Review, 50, 659-676.
Freud, S. (1964). Totem and taboo: Complete psychoLogical works. London:
Hogarth (originally published 1913).
Gecas, V. (1971). Parental behavior and dimensions of adolescent self-evaluation.
Sociometry, 34, 466-482.
Gecas, V. (1972). Parental behavior and contextual variations in adolescent self
esteem. Sociometry, 35, 332-345.
Gecas, V. (1986). The motivational significance of self-concept for socialization
theory. In E.J. Lawler (Ed.), Advances in group processes: Theory and research,
(Vol. 3). Greenwich, CT: JAI Press.
Gergen, K.J. (1973). Social psychology as history. Journal of Personality and Social
Gergen, K.J., & Gergen, M.M. (1983). Narratives of the self. In T.R. Sarbin &
K.E. Scheibe (Eds.), Studies in social identity (pp. 254-273). New York:
Praeger.
Gergen, K.J., & Gergen, M.M. (Eds.). (1984). Historical social psychology.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY:
Doubleday Anchor.
Habermas, J. (1979). Communication and the evolution of society. Boston: Beacon.
Halsell, G. (1986). Prophecy and politics. Westport: Lawrence Hill.
Hankiss, A. (1981). Ontologies of the self: On the mythological rearranging of
one's life history. In Daniel Bertaux (Ed.), Biography and society (pp. 203-209).
Beverly Hills: Sage.
Harre, R (1984). Personal being. Cambridge, MA: Harvard Univesity Press.
Hochschild, A.R (1983). The managed heart. Berkeley: University of California
Press.
Husserl, E. (1970). The crisis of European sciences and transcendental phenomenol-
ogy: An introduction to phenomenological philosophy. Evanston: Northwestern.
Johnson, C.L., & Weigert, A.J. (1980). Frames in confession: The social construc-
tion of sexual sin. Journal for the Scientific Study of Religion, 19, 368-381.
Katz, I. (1981). Stigma: A social psychological analysis. Hillsdale, NJ: Erlbaum.
Kegan, R.G. (1979). The evolving self: A process conception for ego psychology.
Counseling Psychologist, 8, 5-34.
Klapp, O.E. (1969). Collective search for identity. New York: Holt, Rinehart &
Winston.
Kurtz, L.R (1986). The politics of heresy. Berkeley: University of California Press.
Laing, RD. (1965). The divided self. Baltimore: Penguin.
Lasch, C. (1979). The culture of narcissism. New York: Norton.
Lifton, R.J. (1976). The life of the self. New York: Simon & Schuster.
Lindsey, H. (1977). The late great planet earth. Grand Rapids, MI: Zondervan.
MacIntyre, A. (1984). After virtue. Notre Dame: University of Notre Dame Press.
Marris, P. (1975). Loss and change. Garden City, NY: Doubleday.
Marsella, A.J., DeVos, G., & Hsu, F. (Eds.). (1985). Culture and self. New York:
Methuen Tavistock.
Merton, R.K. (1976). Sociological ambivalence. New York: Free Press.
Merton, V., Merton, RK., & Barber, E. (1983). Client ambivalence in profession-
al relationships: The problem of seeking help from strangers. In B.P. DePaul et
aI., (Eds.), New directions in helping (pp. 13-44). New York: Academic Press.
Mitroff, 1.1. (1974). Norms and counternorms in a select group ofthe Apollo moon
scientists: A case study of the ambivalence of scientists. American Sociological
Review, 39, 579-595.
Mojtabai, A.G. (1986). Blessed assurance. Boston: Houghton Mifflin.
Mol, H.J. (1976). Identity and the sacred. New York: Free Press.
Niebuhr, H.R (1963). The responsible self. New York: Harper & Row.
Sampson, E.E. (1983). Justice and the critique of pure psychology. New York:
Plenum.
Sartre, J.P. (1957). Existentialism and human emotions. New York: Philosophical
Library.
Schutz, A. (1962). The problem of social reality. The Hague: Martinus Nijhoff.
Shoemaker, S., & Swinburne, R. (1984). Personal identity. London: Blackwell.
Simmel, G. (1964). The sociology of Georg Simmel. (Translated, edited, and with
an introduction by K.H. Wolff). New York: Free Press.
Smith, M.B. (1985). The metaphorical basis of selfhood. In A.l. Marsella, G.
DeVos, & F. Hsu (Eds.), Culture and self (pp. 56-88). New York: Methuen
Tavistock.
Tiryakian, E.A. (1968). The existential self and the person. In C. Gordon & K.l.
Gergen (Eds.), The self in social interaction (pp. 75-86). New York: Wiley.
Trilling, L. (1972). Sincerity and authenticity. Cambridge, MA: Harvard University
Press.
Turner, R.H. (1976), The real self: From institution to impulse. American Journal
of Sociology, 81 (March), 989-1016.
Weber, M. (1958). The protestant ethic and the spirit of capitalism. New York:
Scribner.
Weigert, A.l. (1975). Substantival self: A primitive term for a sociological psychol-
ogy. Philosophy of the Social Sciences, 5, 43-62.
Weigert, A.l. (1983). Social psychology. Notre Dame: University of Notre Dame
Press.
Weigert, A.l. (1987). Types of Christian eschatology. Unpublished paper, Uni-
versity of Notre Dame.
Weigert, A.l., & Franks, D. (1988). Sociological ambivalence. In D. Franks &
E.D. McCarthy (Eds.), Works in the sociology of emotions. Greenwich, CT: lAI
Press (in press).
Weigert, A.l., Teitge, 1.S., and Teitge, D. (1986). Society and identity. New York:
Cambridge University Press.
Wiley, N. (1979). Notes on self genesis: From me to we to I. In N.K. Denzin (Ed.),
Studies in symbolic interaction, (Vol. II, pp. 87-105). Greenwich: lAI Press.
Yankelovich, D., & Doble, 1. (1984). The public mood: Nuclear weapons and the
U.S.S.R. Foreign Affairs, 63, 33-46.
Zimmerman, M.E. (1981). Eclipses of the self" The development of Heidegger's
concept of authenticity. Athens: Ohio University Press.
Zurcher, L.A. (1977). The mutable self. Beverly Hills: Sage.
Zurcher, L.A. (1983). Social roles. Beverly Hills: Sage.
Author Index
Abelson, R.P., 31, 40, 42, 246, 261 Beilin, H., 179,201
Abelson, J., 76, 89, 92, 98, 105, 106, Belenky, M., 187,201
128,224 Bellah, R., 264, 274, 276, 279
Adams, G.P., 68 Bern, S., 213, 223, 253, 257
Adams, G.R., 214, 223, 248, 251-252, Benjamin, J., 184, 201
257,259,261 Benjamin, L.S., 126, 127
Adorno, T., 184, 186, 198, 201 Bengston, V., 130
Ainsworth, M., 95, 104,214,223 Bercovitch, S., 268, 279
Ajzen, I., 248, 257 Berger, P., 265, 271, 279
Allen, Woody, 75 Berglas, S., 253, 259
Allport, G., 55, 67, 199 Bergman, A., 94, 105, 112, 124, 128
Armon, C., 186,205 Berkowitz, L., 41, 70, 90, 129
Arnoff, J., 205 Berkowitz, M., 28
Aronson, E., 38, 40 Bertanthal, B., 115, 127
Augustine, 263 Berzonsky, M., 243-244, 246-251, 254,
Ausubel, D., 179 256-258
Asch, S.E., 38, 40 Birch, H.G., 4, 27
Berndt, T., 76, 89
Blanshard, B., 194,201
Baillet, S.D., 31, 41 Blank, A., 246, 259
Baillargeon, R., 33, 40 Blasi, A., 9, 27, 192,201,231,234-235,
Bakan, D., 91, 104 241,243,258
Baldwin, J., 9, 10,27,53,67,76,78, Blehar, M.e., 95, 104,214,223
89, 188,201 Bleiberg, E., 124, 127
Balint, M., 123, 127 Block, J., 4, 27, 247, 258
Baltes, P., 3, 27, 29 BIos, P., 86, 89, 92, 93, 104, 110,
Bandura, A., 51, 52,67 112-113, 118-121, 123, 127
Barclay, c., 246-248, 250, 256, 258 Boggiano, A., 68
Bargh, J.A., 30, 40 Bourne, E., 212, 224, 226-227, 241,
Baumeister, R., 268, 271, 279 243,248,258
Baumrind, D., 215, 223 Bowen, R., 86, 89
Bearison, D., 241 Bower, G.H., 40
Beattie, O.V., 16,27 Bowlby, J., 101, 104,213-214,218,221,
Beedy, J., 9, 28 224
284 Author Index
Breckler, SJ., 34, 37, 39, 40, 41 Cullen, J.H., 105

Brehm, J., 253, 258 Cummings, E., 130, 145, 149
Bresnick, S., 59, 63, 69 Czikszentmihalyi, M., 16, 27
Brim, 0.0., Jr., 29, 55, 67
Bromley, 85,90
Brooks-Ounn, J., 44, 69, 114 Damon, VV.,9,27,44,51,68, 72, 77,
Broughton, J., 6, 27, 115-116, 122, 127, 82,89, 114-115, 127, 190,203
181, 184-187, 190-192, 194, 198, Day, J., 41
202 DeBoeck, 82, 89
Bruner, J.S., 32, 40 DeVos, 0., 263
Brunswik, E., 185,202 Dickstein, 44, 68
Borgatta, E., 185-201 Dobert,R., 180, 184, 199,203
Buber, M., 93, 104 Doble, J., 274, 281
Buck-Morss, S., 184,203 Dobson, vv., 214, 223, 252, 261
Buhler, C., 3, 27 Doehrman,M., 76, 89
Busch-Rossnagel, N., 5, 28 Douvan, E., 92, 98, 105
Buss, A.H., 40 Dryer, R., 124, 127
Bybee, J., 249, 262 Dullaert, J., 95, 105
Dunn, 249
Dykstra, C., 202
Calkins, M., 62, 68
Cacioppo, J., 249-251, 258, 260
Campbell, B., 42 Edelson, M., 196,203
Cameron, N., 185,201,203 Edelstein, VV., 9, 27, 28
Candee, D., 176, 184, 203 Eisen, M., 72, 90
Cantor, N., 41, 76, 90, 125, 129 Eissler, R.S., 89
Carson, J., 59, 68 Elias, N., 272, 279
Carver, C.S., 42 Eisenstadt, S.N., 272, 279
Chandler, M., 9, 22, 27, 222, 224 Elkind, D., 86, 89, 110, 120-121,
Chanowitz, B., 246, 250-251, 258-259 126-127
Charme, S., 273, 279 Epictetus, 263, 275, 279
Chasseguet-Smirgel, J., 193,203 Epstein, S., 55, 59, 68, 75, 230,
Chess, S., 4, 27 242-243,258
Chodorow, 100, 104 Erickson, L., 151, 155, 158, 161, 176
Ciompi, S., 12, 27 Erikson, E., 3, 8,14,15,27,58,92,97,
Clinchy, B., 187, 201 99, 105, 106, 130-132, 136-137,
Colby, A., 161, 166, 176, 177, 198 142, 145, 149, 172, 175, 176, 179,
Comer, D., 186,203 211-212,224,226-227,228,233,
Condin, S.M., 95, 105,214, 224 240-241,243-244,258
Connell, J., 51, 52, 69, 201 Erikson, J., 132, 149
Conway, M., 249, 258
Cook, T.D., 38, 42
Cooley, C.H., 43, 44, 45, 50, 51, 52, 53, Fairbairn, W.R.D., 100, 103, 105
65,68 Fast, I., 180, 203
Cooper, A., 117, 118, 124, 127 Fazio, R., 254, 258
Cooper, c., 95, 105, 214, 224 Featherman, D., 29, 266, 279
Cooper, J., 254, 258 Felix, S., 7, 28
Cottrell, N.D., 38, 40 Fenigstein, A., 37, 40
Cox, R.D., 99, 105 Festinger, L., 32, 40, 253, 258
Author Index 285
Fischer, K., 60, 68, 115 Goslin, D., 204

Fischer, L.R., 92, 105 Gould, R., 258
Fishbein, M., 248, 257 Green, c., 126, 128
Fisher, S., 196,203 Green, R.D., 186,203
Fiske, S., 258 Green, M., 203
Fitch, S., 248, 257 Greenberg, J., 110, 127,218
Flavell, J.H., 9,27,40,203 Greenberger, E., 95, 105
Flamer, G., 186, 203 Greenspan, S., 109, 127
Follansbee, D., 176 Greenwald, A.G., 30, 31, 33, 34, 37, 38,
Ford, M., 186,203 39, 40, 41, 42, 63, 68, 189, 197,
Fowler, J.w., 6,7,17,27,138-139,143, 203-204,230-231,242,244,250,
148-149, 199 259
Frankel, J., 95, 105 Grinker, R.R., 124, 127
Franks, D., 272, 281 Grotevant, H., 95, 105,214,224,251,
Frenkel-Brunswik, E., 184,201 259
Freud, A., 92,105,112,127,179 Grubb, P.D., SO, 90
Freud, S., 100, 105, 180-181, 196 Gutman, H., 179,204
Friedman, M., 252, 260 Greenberg, P., 196,203
Fritzsche, V., 7, 29
Fromm, E., 103, 105
Frondizi, R., 230, 242 Haan, N., 151, 176
Habermas, 1., 180, 184, 187-188, 194,
197, 199,204,268,280
Gabel, J., 202, 203 Hansberg, H., 212, 224
Galanter, E., 246, 260 Hall, C., 126, 129
Gallup, G.G., 30, 40 Halsell, G., 275, 280
Gangestad, S., 37, 40, 125 Haltiwanger, 57, 69
Gara, 79, 86, 90 Hamid, P.N., 95, 105
Garrison, w., 178 Handel, G., 185,204
Garrod, A., 161, 176 Hankiss, A., 277, 280
Gasset, J.O., 278 Hanson, R., 246, 257, 259
Gecas, V., 77, 87, 89, 197,271,279 Harkins, S., 38, 41
Gelman, R., 33, 40 Harlow, H., 100, 101
Gergen, K., 55, 56, 68, 89, 186,203, Harre, R., 263, 280
266-267,280 Hart, D., 9, 27, 44, 68, 72, 77, 82, 89,
Gibbs, J., 176, 195 114-115, 127
Giddens, A., 188,203 Harter, S., 43, 44, 45, 46, 48, 49, 51,
Gilligan, c., 3, 28, 98, 105, 175-177, 52,54,56,59,63,64,68,69,72,
187,218,222,224 90, 114, 127
Giovacchini, P., 123, 127 Hartmann, H., 179, 186,204
Gmeunder, P., 7, 29 Hartup, w., 76, 90
Goffman, E., 267-268, 280 Harvey, 0.1., 190,204,253,259
Goethe, 264 Hastorf, A.H., 40
Goldberg, A., 106, 126, 127 Hauser, S., 151, 160, 176, 180, 181,
Goldberger, N., 187,201 184, 204
Goldstein, A.P., 69 Havighurst, R., 130, 150
Gordon, c., 68 Hays, w., 177
Goodman, R., 7, 28 Healy, J., 241
Goldman, R., 249, 260 Hegel, GWF., 192, 204
286 Author Index
Heider, F., 32, 33, 41, 247, 259 Jourard, S.M., 55, 69
Heiler, E., 23, 28 Jung, C., 3, 28, 55, 69
Henry, W.E., 130, 145, 149
Hepburn, A., 186,203
Hess, R., 185,204 Kafka, F., 4, 18-26,28
Hetherington, M., 90, 127 Kagan, J., 4, 28
Hewer, A., 176 Kahneman, D., 247, 259
Higgins, E.T., 40 Kaiser, M., 84, 90
Hobhouse, L., 179, 204 Kanfer, F.H., 51, 52
Hochschild, A., 271, 280 Katz, I., 273, 280
Hoffer, E., 36, 41 Kao, C., 249, 258
Hoffman, J., 214, 223 Kaplan, B., 191,204
Hoffman, M., 251, 259 Keenan, J.M., 30, 41
Hogan, A., 45, 52, 69 Kegan, R., 5, 6, 7, 9,17,28,103,105,
Holstein, 151, 176 266,280
Holt, R., 181, 188, 198,204 Keller, H., 267
Honess, T., 89 Kelly, G., 220, 230, 242, 246-247
Horkheimer, M., 194,204 Kelman, H., 251, 259
Homey, K., 55, 68 Keniston, K., 199,204
Horowitz, G., 196,204 Kernberg, 0., 124, 128, 224
Hovland, G., 253, 259 Kihlstrom, J., 31, 41, 76, 90, 246, 257,
Howard, K., 92, 106 259
Hsu, F., 263, 280 Kilkenny, R., 9, 29
Hult, R., 212, 224 Kirker, W.S., 31, 42
Hunt, D., 190,204 Kivnik, H., 132, 149
Husser!, E., 191-192,204,270,280 Klapp, 0., 268, 280
Hyman, D., 249, 262 Kohlberg, L., 3, 4, 6, 8, 28, 35, 41, 109,
131,135,139,146,149,150-154,
156-159,163-164,166,173,
Ingleby, D" 196, 204 175-178,180,186,191,195,217,
Inhelder, B., 32,41 222
Irwin, T., 149 Kohut, H., 14, 15,28,97, 102, 103,
Isen, A.M., 40 105,117-118,123-124
Koopman, R., 212, 225, 252
Kroger, J., 212, 224
Jackson, J.D., 38, 42 Kubie, L., 196, 205
Jacobson, A., 176, 178 Kuhn, M., 72, 90
Jacoly, R., 196,204 Kuhn, T., 190,205,244,259
James, W., 9, 28, 31, 41, 43-46, 48, 50, Kurtz, L., 272, 280
52-53,58,69,71-73,75-76,89, Kramer, R., 155, 175, 177
114,128,181,192,230,242,246 Kuiper, N.A., 31, 42
Jensen, A., 92, 105
Jesus, 263, 275
Johnson, J., 223, 225 Labov, w., 187,205
Johnson-Laird, P.N., 31, 32, 41 Lasch, C., 268, 279, 280
Jones, E., 253, 256, 259 Lahey, L., 7, 28
Joseph, C., 79, 90 Laing, R.D., 5, 28, 268, 280
Josselson, R., 93,95,97,98,99, 105, Lakatos, I., 190,205,244-245,257,259
110,113-114,118-119,128 Lambert, H., 154, 156, 158, 160, 177
Author Index 287
Langer, E., 246, 250-251, 259 Markman, E., 40

Lapsley, D.K., 9, 28, 33, 41, 110, 118, Markus, R., 30, 41, 230, 242, 246, 260
120, 122-123, 126, 128, 277 Marris, P., 272, 280
Latane, B., 38, 41 Marsella, A., 263, 280
Leahy, R., 68 Martindale, e., 76, 90
Lecky, P., 32, 33, 41, 55, 69, 75 Martindale, D., 179-206
Lee, B., 29 Maslach, C., 36-41
Leiper, R., 212, 224 Maslow, A., 55, 69, 103, 105
Leites, N., 188,205 Masterson, 1, 102, 105
Lemon, B., 130, 150 Matteson, D., 212, 224
Lepper, M., 250, 259, 260 Mayer, R., 157, 177, 191,205
Lerner, R., 266, 279 McAdams, D., 180, 184, 188, 192,206,
Lester, E., 109, 128 244,260
Lerner, R.M., 5, 28 McClelland, D., 179,206
Levine, 1, 126, 128 McClintock, e.G., 40
Levinson, D., 184,201,257 McCollough, G., 151,178
Levinson, D.l, 3, 28, 100, 105 McConochie, D., 95, 105
Lewis, M., 30, 41, 44, 69, 114 McGillicuddy-Delisi, A., 223, 225
Lewittes, R.S., 99, 105 McGuire, 71, 90
Lickona, T., 41 McKillop, K., 254, 258
Lieberman, M., 176, 177 McPartland, 72, 90
Lifton, R.J., 268, 280 McPhail, e., 78, 90
Light, R., 159, 177 McPherson, C., 194,206
Lindsey, R., 274, 280 Meacham, 1, 90
Livesly, W., 85, 90 Mead, G.R., 9, 28, 43, 69, 76, 78, 90,
Loevinger, 1, 4, 7,17,28,151-154, 189, 191,206
157-159, 163-164, 166, 173, Meissner, W., 125, 128
177-178,179-200,206,228, Mentkowski, M., 166, 178
234-235,238,240,242 Messer, B., 49, 69
Looft, W., 181,206,243,259 Merton,R., 179,206,270,272,280
Lord, C., 250, 259, 260 Milgram, S., 38, 41
Lorge, I., 185, 206 Mill, lS., 130, 150
Lucca-Irizarry, N., 72, 89 Miller, A., 118, 128
Luckmann, T., 265-266, 279 Miller, G., 246, 260
Lukes, S., 191,206 Miller, lB., 99, 100, 105
Lynch, M.D., 89 Millon, T., 126, 128
Mills, e.w., 265
Minsky, M., 32, 41
MacIntyre, A., 268, 280 Mischel, T., 56, 69, 90
Madsen, R., 264, 279 Mitchell, S., 110, 127, 218
Magana, R., 156, 178 Mitroff, I., 272, 280
Mahler, M., 93, 94, 103, 105, 112, 124, Modell, A., 117, 118, 128
127,213,218,221 Modgill, C., 177, 178
Maloney, 1, 72, 89 Modgill, S., 177, 178
Marcia, 1, 212, 213, 215, 224-225, 227, Mojtabai, A., 264, 273, 275, 280
242,243,247-248,252-253, Montague, w.P., 68
256-257,260 Monsour, A., 59, 69
Marcus, R., 197,206 Moore, 1, 192,206
Marcuse, R., 200, 206 Morgan, E., 99, 105
288 Author Index
Morrison, A., 124, 127, 128, 129 Peterfreund, E., 180,206

Moss, H.A., 4, 28 Peterson, 1., 1972
Montmayor, R., 72, 90 Petty, R., 249-251, 256, 260
Much, N., 161, 178 Piaget, 1., 3, 7, 8, 9, 11, 12,29, 32, 33,
Murphy, M., 110, 118, 120-123, 126, 41, 53, 109, 157, 185,214, 261
128 Pillemer, D., 159, 177
Murray, A., 179,206 Pine, E, 94, 105, 112, 124, 128, 214
Mussen, P., 127 Pittman, T., 69
Myers, D.G., 38, 41 Podd, M., 212-225
Poppen, P., 212, 225
Popper, K., 244, 261
Neeman, 49, 69 Power, A., 150
Nelson, K., 156, 178 Power, Ee., 7, 29, 139, 146, 150
Nesselroade, 1.R., 3, 27 Powers, S., 9, 29, 178
Neugarten, B., 130-131, 150 Pratkanis, A.R., 30, 31, 33, 39, 41, 230,
Newberger, C., 216, 223, 224 242
Nickerson, R.S., 41 Preston, E., 250, 259
Nielsen, E., 214, 223 Pribram, K., 246, 260
Neisser, U., 246, 260 Pryor, 1., 41
Niebuhr, H., 268, 280 Pulver, S., 128
Nisbett, R., 249, 260
Noam, G., 4, 6, 8, 9, 28, 29,109,111,
128,131,135,150,176,178,185, Quintana, S., 33, 41
202
Norem-Hebeisen, A.A., 89
Nowakowski, M., 51, 69 Rabin, A., 205
Nowicki, 72, 90 Rappaport, D., 179, 185,206
Nunberg, H., 188, 193,206 Raskin, R., 126, 129
Nunner-Winkler, G., 180, 199,203 Read, D., 252, 261
Nurius, P., 197,206 Reese, H., 3, 27
Redmore, e., 166, 178,206-207,
234-235,242
O'Brien, R., 84, 90 Rhodewalt, E, 253, 259, 261
Offer, D., 92, 105, 106 Rice, K.G., 9, 28
Offer, 1., 92, 106 Ridl, 1., 38, 41
Oresick, R., 231, 241 Rieff, P., 196,207
Oser, E, 7, 29, 138, 150 Riegel, K., 192,202
Ostrov, E., 92, 106 Reisman, K., 187, 207
Overton, W.E, 68 Ricoeur, P., 197,207
Ozer, D., 241 Robinson, 64, 69
Rogers, T.B., 30, 42, 261
Roniger, L., 279
Padawer-Singer, 71, 90 Ronis, D.L., 38, 41
Parks, S., 202 Rose, S., 69
Parsons, T., 179, 206 Rosenberg, M., 45, 51, 56, 61, 62, 70,
Partol, S., 216, 225 72,78-79,86,197
Payne, A., 185,206 Ross, L., 249, 260
Pepper, S., 179, 206 Rothstein, A., 118, 119, 120, 129
Peevers, 72, 90 Rotter, 1., 185, 207
Author Index 289
Rowe, I., 212, 225 Sokel, W.H., 18,29

Rubin, D.C., 30, 42 Sorrentino, R., 40
Rubinsky, E.C., 37,40 Souvaine, E., 7, 28
Ryan, J., 214, 223 Speicher-Dubin, B., 176-177
Sru1, T.K., 41, 242
Stager, M., 151, 178
Sanford, R., 184, 185,207 Steinberg, L., 113, 129
Santilli, N.R., 90 Stern, J.P., 23, 29
Santostefano, S., 178 Stern, P., 106
Sarbin, T., 230, 242 Stewart, A., 241
Sartoe, J.P., 264, 273, 280 Strait, M., 161, 178
Schachter, S., 249, 261 Strickland, B., 72, 90
Schank, R.C., 246, 261 Stroud, J., 151, 176
Scheier, M.P., 37, 40, 42 Suarez, S.D., 30, 40
Schiedel, D., 213, 215, 225 Sullivan, E., 151, 156, 158, 178, 179
Schlenker, B., 41,245,252,259 Sullivan, H.S., 9, 29, 100, 106, 179,
Schneider, w., 246, 261 185, 193, 196
Schroder, H., 190,204 Sullivan, w., 264, 279
Schutz, A., 270, 281 Suls, J., 40, 41, 42, 68, 292
Sears, R., 197 Suppe, P., 244, 262
Secord, P., 72, 90 Surrey, J., 100, 106
Selman, R., 5, 7, 9, 29, 109, 115-116, Swensen, c., 180, 184, 188,207
121-123, 129, 157 Swidler, A., 264, 279
Sentis, K., 230, 242 Swinturne, R., 265, 281
Shapiro, R., 109, 129 Symonds, P.M., 92, 106
Shaver, P., 90, 109, 129
Sherif, M., 253-254, 261
Sherman, S.,249, 261 Tarule, J., 187, 201
Shiffrin, R., 246, 261 Taylor, C., 188,207
Shils, E., 179,206 Thylor, S., 258
Shirk, S., 28 Thao Van Trinh, V., 202
Shoemaker, S., 265, 281 Thomae, H., 29
Showers, C., 247, 261 Thomas, A., 4, 27
Shulik, R., 139, 142, 150, 160-161, 178 Thorndike, E., 185,206
Sigel, I., 215-216, 222-223, 225 Tillich, P., 137, 150
Silverberg, S., 113, 129 Tipton, S.M., 279
Singer, J., 249, 261 Tiryaleian, E., 270, 281
Skelton, J., 253, 259 Tobin, S., 130, 150
Skoe, E., 212, 225 Tolstoy, L., 96, 106
Sloan, D., 202 Tracy, D., 138, 150
Slugoski, B., 212, 225 Trilling, L., 237, 239, 242, 270, 281
Smelser, N., 106, 179,207 Trungpa, C., 65, 70
Smith, M., 271, 281 Thcker, c., 78, 90
Smollar, J., 78, 90 Tulku, T., 65, 70
Snarey, J., 4,28, 109, 128, 131, 136, Turiel, E., 187
150, 151, 155-156, 158, 161, 178,
186
Snyder, M., 37, 40, 42, 67, 70, 125 Unger, R., 191,207
Socrates, 263 Uzgiris, I., 69
290 Author Index
Vaillant, G., 3, 29, 100, 106 Winnicott, D.W., 5, 29, 102, 103, 106,
Vallacher, R.R., 56, 70 117, 123
Volpe, J., 77, 90 Winograd, T., 247, 262
Vygotsky, L., 9, 10, 29 Winston, P.H., 41
Wolf, E.S., 102, 106
Wolff, P., 109, 124, 129
Wright, D., 180,207
Wall, S., 95, 104 Wyer, R.D., 41
Waters, E., 95, 104,214,223 Wyler, J., 242
Waterman, A., 212, 228, 242 Wylie, A.J., 95, 105
Weber, M., 263-264, 269, 273, 281 Wylie, R., 44, 70
Weber, S.1., 38,42
Wegner, D.M., 70
Weigert, A., 264-265, 272, 278, 281 \hlom, I.D., 103, 106
Weiss, B., 176 Yankelovich, D., 274, 281
Wessler, R., 158, 163, 166, 178 Yardley, K., 89
Werble, B., 124, 127 Youniss, 1.,77,78,90
Werner, H., 111, 129
Wheeler, L., 90
White, M.T., 117, 118, 129 Zahaykevich, M., 181, 184, 185-186,
Whiteley, 1., 154, 156, 161, 178 198-199,207-208
Wicklund, R., 51, 52, 70, 253, 261 Zajonc, R.B., 38, 42
Widtner, K., 7, 29 Zanna, M.,254,258
Wiley, N., 263, 281 Zavalloni, M., 186, 208
Williams, K., 38, 41 Zimiles, H., 241
Williams, K.D., 38, 42 Zimmerman, M., 273, 281
Wilson, 249, 262 Zucker, R., 205
Wilson, W., 249, 259 Zurcher, L., 268, 274, 281
Subject Index
Acceptance oflimits, 142, 144, 148 Dimensions of Childhood Depression

Accommodation, 11, 12, 17,221 Profile, 51
Adolescence Discounting, 49, 50, 62, 63
development, 91, 94, 110 Disengagement, 130, 145
as second individuation in phase, 93, Distancing, 215-216, 222-223
112-114, 119-122 Drive gratification, 100
self, 58, 62
Ambivalence, 269-270, 272
eschatological response to, 273-276 Ego
existential response to, 273-276 cognitive style, 190
Androgyny, 213 conscious preoccupations, 190
Anxiety, 19, 75 cybernetic approaches to, 189, 192
Aretaic review, 142-143, 146-147, 148 development, 181, 184, 186, 189,235
Assimilation, 11, 12,221 Erikson's lifespan model of, 132,
Attachment, 92, 95, 96, 218 174-175
Authenticity, 136, 269-272 as executive of the personality, 188
Autonomy, 23, 91, 92, 93, 94, 97, 98, functions, 188
103,104,180,191,193-194 general, 154
impoverished,92
integrity, 130-131, 136, 137, 146,
Beneffectance, 63 148-149
Bhagavad Gita, 145 interpersonal style, 189
Borderline personality, Ill, 118-119, involvement, 33, 34
124 themes of, see Aretaic review, Accep-
tance of limits, Detachment, Holis-
tic perfection
Character, 189 moral qualities of, 189
Cognitive dissonance, 38 organization, 92
Confirmation bias, 247-257 psychology, 179-180
Construct validity, 185, 212 strength, 14, 136, 152
Core themes, 15,16,17,18,19,25 synthetic function, 188-189, 193
tasks, 33, 34, 36, 37
transcendent, 145
Despair, 136, 148 weakness, 92
Detachment, 142, 144-145, 148 Egoism, 189
292 Subject Index
Empathy, 15 status paradigm, 212-213, 219

Encapsulation, 6, 12, 14, 15, 16, 18,21, criterion of, 227
25, 111 process view, 243, 256
horizontal, 13, 15 social cognitive characteristics, 252
vertical, 12, 13, 15 as theorist, 243, 246-247
Epigenesis, 93, 132, 211 in women, 92-93, 97-100
Equilibration, 12, 17, 186 Imaginary audience, 110, 116, 120-126
Equilibrium, 8 Impression management, 61
Eschatology, 265, see also Ambivalence Individualism, 194-195
Existential isolation, 103 Individuation, 191, 214,216-219
Intentional action, 228-229
aspects of, 230
Faith development, 138 Intentionality, 232
Faith Development Interview, 139, 146, Interiorization, 217, 222
148 Internal perspective, 10
Fidelity, 227, 233 Intimacy, 93, 212
Field dependence, 250-251
Formal operations, 219
Latency, 93
Latitude of rejection, 253
Generalized other, 45 Levels of representation, 31-33
Generativity, 176 Liberalism, 195
Good-enough parenting, 111, 117 Life themes, 16
Locus of control, 72
Hermeneutics, 197
Holistic reflection, 142, 148 Mirroring, 15, 117, 119, 123-124
Moral Judgment Interview, 139, 146,
151, 157, 164-165
Idealogy, 93, 190, 194-196,200,216, Moral reasoning, 212, 217
226,234,269 Mourning, 22, 113-114, 119
Identity, 91, 92, 96, 99, 104, 188, 193, Multiple selves, 75
211-212,226-241
achieved, 99, 175,212,217,249,
252 Narcissism, 111, 118, 121, 123-126
as authenticity, 238-239 Narcissistic object choice, 119
commitment, 213,227,248-250 Narcissistic personality, 111, 112-113,
components of, 234 117-118, 123-125
crisis, 136, 227-228 Narcissistic rage, 118
diffusion, 97, 212-213, 251-252 Narrative, 15
exploration, 213 New Ego Paradigm, 198
foreclosure, 212-213, 217, 248-252
formation, 98, 132, 173,243
management of, 237 Object relations, 100, 102, 109, 114,
moratorium, 96, 202, 212, 248 116, 123, 211-212
observed, 235 Observing ego, 115, 119-122
orientation to, 239, 248 Oedipal period, 211, 213, 215
psychosocial,214 Ogive rules, 158, 159, 163
as self-theory, 220, 244-246 Old Testament, 263-264
social,265 Ontological anxiety, 137-138
Subject Index 293
Orthogenetic principle, 111 existential, 114, 116

Overassimilation, 12, 21 false, 5,117-118,124-126
fragmentation, 14
grandiose, 15, 110, 111, 116-118, 124
Parenting styles, 216 I-self, 44, 50, 52, 53
Partial transformation, 13 as integrated, 55
Personal agency, 115, 116, 122-123 as known, 214, 230
Personal construct theory, 220 looking glass, 45, 53
Personal fable, 110, 116, 120-126 me-self, 50, 52, 53, 214
Positivism, 179, 181, 188 monitoring, 67,110,125-126,192,
Pragmatic social constructivism, 266 251,256
Psychoanalytic theory, 91, 109, 111, narcissistic, 15, 97
112-114 as object, 14, 72, 230
Psychological birth, 221 ontological, 116, 123
organization, 18, 230
and personality traits, 85-88
Rapprochement, 94, 95, 214, 218, 221 presentation, 34, 61
Recapitulation, 5, 17, 110-116, 123 private, 34, 36, 39
Refueling, 94, 218, 221 as producer of development, 5
Relatedness, 94, 99, 103, 104 public, 34, 36, 39
Religious development, 138 schemata, 30,230
Research programs as scientific-theory, 55
scientific, 245 strength, 8, 14, 15
personal, 245 as subject, 72, 114, 228-230
Rites of passage, 219 as trait labels, 56-58, 63
Role-taking, 9, 19 transcendent, 267
as organized structure of knowledge,
30, 31
Schema, 11,31,33 understanding, 71, 72, 87, 88, 110,
Self 115,121,126,268
as agent, 44 as unified, 55
attributions, 71 worth, 30,35, 37, 39, 43-45, 46-54,
autonomous, 218 148
Buddhist perspective, 65-66 Self-awareness, 189, 191, 192,263
categorical, 115-116, 123 Self-concept, 39, 44, 46, 61, 71, 74, 88,
cohesion, 14 190, 230
collective, 34, 36, 39 core, 89
complexity, 8 confederacy of, 75
conservation of, 52-58 social, 76, 78, 82-87
consistency of, 55 Self-consciousness
as constructed, 44 public, 37
continuity, 72, 73, 97 private, 37
diffuse, 34, 35, 39 Self-Consciousness Scale, 37
divided, 5 Selthood, 263-264
as empirically constituted, 267 Self-Monitoring Scale, 37
enhancement, 63 Self-objects, 102
epistemological, 116, 123 Self-Perception Profile, 48
esteem, 38, 39, 48, 50, 52, 72, 77, Self-Perception Profile for Children, 46
113-114,117,227,253 Self-Perception Profile for College Stu-
evaluation, 72, 77, 87, 149 dents, 49
294 Subject Index
Self-reflexivity, 263-264 TOrE,246

Sentence Completion lest, 139, 151, Transitory narcissism stage, 119, 123
157, 162-163, 164-165, 198 Transmutating internalization, 15, 111,
Separation anxiety, 119 117-118, 123, 126
Separation-individuation, 93, 94, 95, 98,
99, 101, 110-114, 116, 118, 120,
123, 126
Mahler's phases of, 112-113 Volition, 72, 116
Set theory model, 78
Social embeddedness, 91, 94
Sturm and Drang, 92
Symbolic interactionism, 9, 10, 192 Wisdom, 136, 137

Self, Ego, and Identity - Integrative Approaches

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Self, Ego, and Identity - Integrative Approaches

Hochgeladen von

Copyright:

Verfügbare Formate

Daniel K. Lapsley F. Clark Power

Self, Ego, and Identity

Library of Congress Cataloging-in-Publication Data

© 1988 by Springer-Verlag New York Inc.

Typeset by Asco Trade Typesetting Ltd., Hong Kong.

ISBN 978-1-4615-7836-9 ISBN 978-1-4615-7834-5 (eBook)

definitional preference. This is in keeping with the integrative spirit of the

Harter argues that a comprehensive model of self-worth must incorporate

theory of the self must be interwoven with a theory of relationship." This

tive on self-understanding that is motivated by "pragmatic social construc-

1. The Self, Adult Development, and the Theory of Biography

2. A Social-Cognitive Account of the Self's Development

3. The Construction and Conservation of the Self: James and

4. The Adolescent Self-Concept in Social Context

5. The Embedded Self: I and Thou Revisited

Part II: Ego

6. The "New Look" at the Imaginary Audience

7. Integrity and Aging: Ethical, Religious, and Psychosocial

8. The Relationship Between Ego and Moral Development: A

9. Ego and Ideology: A Critical Review of Loevinger's Theory

Part III: Identity

10. Common Processes Underlying Ego Identity,

11. Identity and the Development ofthe Self

12. Self-Theorists, Identity Status, and Social Cognition

13. To Be or Not: Self and Authenticity, Identity, and

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Subject Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

Michael D. Berzonsky, Department of Psychology, State University of

Augusto Blasi, Department of Psychology, University of Massachusetts-

John M. Broughton, Box 33, Teacher's College, Columbia University,

Anthony G. Greenwald, Department of Psychology, NI-25, University of

Daniel Hart, Department of Psychology, Rutgers University, Camden, NJ

Susan Harter, Department of Psychology, University of Denver, Denver,

Ruthellen Josselson, Department of Psychology, Towson State University,

Daniel K. Lapsley, Department of Psychology, University of Notre Dame,

Loren Lee, School of Education, Harvard University, Appian Way, Cam-

James E. Marcia, Department of Psychology, Simon Fraser University,

Ann R. Power, South Bend Community School Corporation, 635 South

F. Clark Power, Program of Liberal Studies, 343 Decio Hall, University of

Kenneth Rice, Department of Psychology, University of Notre Dame,

John Snarey, Candler School of Theology, Bishops Hall, Emory Universi-

Andrew J. Weigert, Department of Sociology, University of Notre Dame,

Marta K. Zahaykevich, Center for the Study of Psychological Develop-

have uncovered personality continuities over time (e.g., Block, 1971;

status of structural integration. Each developmental transformation pro-

Expanding Structural Perspectives for a Theory

adequate and more comprehensive. In Kohlberg's scheme, for example,

The Self in Transformation

1. The Subjective-Physical Self Stage

2. The Reciprocal-Instrumental Self Stage

At the reciprocal-instrumental stage, there is the possibility of understanding self-interests

3. The Mutual-Inclusive Self Stage

At the mutual-inclusive self-other perspective, the person understands others in a relationship

TABLE 1.1. (Continued).

4. The Systemic-Organizational Self State

Developmental Processes of the Self: Assimilation,

velopment. Equilibrium is the state of balance between assimilation and

for their parents' divorce. What begins as one area of developmentally

Strength and Development of the Self:

recent Piagetian research, which show considerable discrepancies in de-

narcissistic) self. As is widely known, he views disturbed self-object rela-

Core Themes: A Bridge Between Equilibration