(Perspectives On Individual Differences) Kenneth H. Craik (Auth.), Kenneth H. Craik, Robert Hogan, Raymond N. Wolfe (Eds.) - Fifty Years of Personality Psychology-Springer US (1993) PDF

Fifty Years of
Personality Psychology
PERSPECTIVES ON INDIVIDUAL DIFFERENCES
CECIL R. REYNOLDS, Thxas A&M University, College Station
ROBERT T. BROWN, University of Nurth Carolina, Wilmington
Current volumes in the series

EXPLORATIONS IN TEMPERAMENT
International Perspectives on Theory and Measurement
Edited by Jan Strelau and Alois Angleitner
FIFTY YEARS OF PERSONALITY PSYCHOLOGY

Edited by Kenneth H. Craik, Robert Hogan, and Raymond N. Wolfe
HANDBOOK OF CREATIVITY
Assessment, Research, and Theory
Edited by John A. Glover, Royce R. Ronning, and Cecil R. Reynolds
HANDBOOK OF MULTIVARIATE EXPERIMENTAL

PSYCHOLOGY, Second Edition
Edited by John R. Nesselroade and Raymond B. Cattell
HISTORICAL FOUNDATIONS OF EDUCATIONAL PSYCHOLOGY

Edited by John A. Glover and Royce R. Ronning
INDIVIDUAL DIFFERENCES IN CARDIOVASCULAR

RESPONSE TO STRESS
Edited by J. Rick 'fumer, Andrew Sherwood and Kathleen C. Light
LEARNING STRATEGIES AND LEARNING STYLES

Edited by Ronald R. Schmeck
THE NEUROPSYCHOLOGY OF INDIVIDUAL DIFFERENCES

A Developmental Perspective
Edited by Lawrence C. Hartlage and Cathy F. Telzrow
PERSONALITY, SOCIAL SKILLS, AND PSYCHOPATHOLOGY

An Individual Differences Approach
Edited ~y David G. Gilbert and James J. Connolly
SCHIZOPHRENIC DISORDERS
Sense and Nonsense, in Conceptualization, Assessment, and Treatment
Leighton C. Whitaker
THEORETICAL FOUNDATIONS OF BEHAVIOR THERAPY

Edited by Hans J. Eysenck and Irene Martin
A Continuation Order Plan is available for this series. A continuation order will bring delivery
of each new volume immediately upon publication. Volumes are billed only upon actual
shipment. For further information please contact the publisher.
Fifty Years of
Personality Psychology
Edited by
Kenneth H. Craik
University of California
Berkeley, California
Robert Hogan
University of Tulsa
Tulsa, Oklahoma
and
Raymond N. Wolfe
State University of New York
Geneseo, New York
SPRINGER SCIENCE+BUSINESS MEDIA, LLC

Llbrary of Congress Cataloglng-ln-Publlcatlon Data
Flfty years of personallty psychology / edlted by Kenneth H. Cralk,

Robert Hogan, and Ray.ond N. Holfe.
p. c •. -- (Perspectlves an IndIvIdual dlfferences)
A co.paratlve analysls of the 1937 textbooks Personallty by Gordon
H. Allport and Psychology of personallty by Ross Stagner.
Includes blbllographlcal references and Index.
ISBN 978-1-4899-2313-4 ISBN 978-1-4899-2311-0 (eBook)
DOI 10.1007/978-1-4899-2311-0
1. Personallty--Hlstory. 2. Allport, Gordon H. (Gordon Hl11ard),
1897-1967. Personal1ty. 3. Stagner, Ross, 1909- Psychology of
personal1ty. 1. Cralk, Kenneth H. II. Hogan, Robert, 1937-
III. Holfe, Ray.ond N. IV. Allport, Gardon H. (Gordon Hl11ard),
1897-1967. Personal1ty. V. V. Stagner, Ross, 1909- Psychology of
personal1ty. VI. Serles.
[DNLM: 1. AII port, Gordon H. (Gordon Hl11ard), 1897-1967.
2. Stagner, Ross, 1909- 3. Personal1ty--congresses.
4. Personal1ty Assess.ent--congresses. 5. Psycholog1cal Theory-
-congresses. BF 698 F4691
BF698.F525 1993
155.2'09'04--dc20
DNLM/DLC
for Llbrary of Congress 92-48903
CIP
ISBN 978-1-4899-2313-4
10 1993 Springer Science+Business Media New York

Origina11y published by Plenum Press, New York in 1993
Softcover reprint ofthe hardcover lst edition 1993
All rights reserved
N o part of this book may be reproduced, stored in a retrieval system,

or transmitted in any form or by any means, electronic, mechanical,
photocopying, microfilming, recording, or otherwise, without written
permission from the Publisher
To GoRDON W. ALLPORT AND Ross STAGNER
in celebration of their contributions
to personality psychology
Contributors
Irving E. Alexander, Department of Psychology, Duke University,

Durham, North Carolina 27706
Roy F. Baumeister, Department of Psychology, Case Western
Reserve University, Cleveland, Ohio 44106
Peter Borkenau, Department of Psychology, University of Bielefeld,
N-4800 Bielefeld 1, Germany
Bertram J. Cobler, Committee on Human Development, University
of Chicago, Chicago, Illinois 60637
Kenneth H. Craik, Institute of Personality and Social Research, Uni-
versity of California, Berkeley, California 94720
Bella M. DePaulo, Department of Psychology, University of Virginia,
Charlottesville, Virginia 22903
Alan C. Elms, Department of Psychology, University of California,
Davis, California 95616
Robert A. Emmons, Department of Psychology, UDiversity of Cali-
fornia, Davis, California 95616
Garth J. 0. Fletcher, Department of Psychology, University of
Canterbury, Christchurch 1, New Zealand
David C. Funder, Department of Psychology, University of Califor-
nia, Riverside, California 92521
Robert Hogan, Department of Psychology, University of Tulsa, Tulsa,
Oklahoma 74104
vii
viii CONTRIBUTORS
Oliver P. John, Department of Psychology, University of California,

Berkeley, California 94720
Salvatore R. Maddi, School of Social Ecology, University of Cali-
fornia, Irvine, California 92717
Gerald A. Mendelsohn, Institute of Personality and Social Research,
Berkeley, California 94720
Lawrence A. Pervin, Department of Psychology, Rutgers Univer-
sity, New Brunswick, New Jersey 08903
Richard W. Robins, Department of Psychology, University of Cali-
fornia, Berkeley, California 94720
M. Brewster Smith, Board of Studies in Psychology, University of
California, Santa Cruz, California 95064
Ross Stagner, Department of Psychology, Wayne State University,
Detroit, Michigan 48202
David G. Winter, Department of Psychology, University of Michigan,
Ann Arbor, Michigan 48109
Raymond N. Wolfe, Department of Psychology, State University of
New York, Geneseo, New York, 14454
Lawrence S. Wrightsman, Department of Psychology, University of
Kansas, Lawrence, Kansas 66045
Preface
This volume celebrates the textbooks Personality: A Psychological In-

terpretation by Gordon W. Allport and Psychology of Personality by
Ross Stagner, both first published in 1937. In 1987, several occasions
were held to mark the fiftieth anniversary of the publication of the two
volumes and to acknowledge their role in defining and establishing the
identity of personality psychology as a distinctive field of scientific in-
quiry within the United States.
At any given time, the textbooks of a science offer revealing infor-
mation about its intellectual structure and research program. Even
more so, its ''founding'' textbooks provide a temporal anchor for gain-
ing a historical perspective on the field and a source of insights con-
cerning its subsequent development and current situation.
In spring 1987, at the University of California's Institute of Per-
sonality Assessment and Research (IPAR) (now the Institute of Person-
ality and Social Research) in Berkeley, Ross Stagner was invited to
offer his personal perspective on these two textbooks and the sub-
sequent development of personality psychology. In addition, a special
symposium was devoted to an appreciation of the life and works of
Gordon W. Allport. In a third series of talks, invited speakers offered
commentaries on issues that were central to personality psychology in
1937 and that continue to warrant our attention today. A£, part of its
annual meetings in August 1987, the American Psychological
Association's Division of Personality and Social Psychology held a spe-
cial marathon four-hour symposium that was an expanded version of
!PAR's celebrations.
In this volume, we offer highlights of these occasions as well as
additional contributions developed especially for this publication. The
ix
X PREFACE
book is organized into four sections. First, the introductory chapter

presents a comparative analysis of the 1937 textbooks by Allport and
Stagner and then uses this context to describe the origins and concep-
tualization of our volume and to give a detailed account of its organiza-
tion and contents. The second section of the volume includes three
chapters dealing with the historical and personal background of the two
textbooks. The third section groups three chapters concerning the cur-
rent state of personality psychology and its contemporary textbooks. In
the fourth section, we devote twelve chapters to gaining a present-day
perspective on such abiding issues in personality psychology as the
individual and the single case, motives and the self, judging persons,
and personality assessment and prediction. Finally, the epilogue offers
an optimistic view of the future of our field.
This volume has three aims. First, we see it as a contribution to
our ongoing task of gaining a historical perspective on the development
of personality psychology as a scientific endeavor. Beyond its intellec-
tual importance, such collective remembering holds the promise of
serving valuable social functions for a community of researchers. For
example, narrative accounts of a community's origins can generate
vivid reminders of its members' shared aspirations. Second, we hope
that the volume will bring about more explicit and concerted discussion
of the possible forms and substance of contemporary textbooks in per-
sonality psychology. Many pertinent issues have been raised in rather
fragmented fashion over the years in textbook reviews published in
Contemporary Psychology and elsewhere. This volume constitutes a
beginning forum for reflecting on and joining opposing views on these
matters. The third aim of the volume is to employ a historical vantage
point as one means of gaining a comprehensive overview of our current
research agenda in personality psychology. Has a particular research
topic now perceived as "trendy'' in fact been an enduring concern of
our field from the outset? Have some topics and methods fallen by the
wayside that should not have been abandoned after all? The compara-
tive analysis of textbooks over time can afford us a broad picture of
such continuities and discontinuities in our field's research directions.
Given these purposes, we have been remarkably fortunate in gain-
ing the cooperation of an array of contributors who are authorities in
the fields of their individual assignments. We greatly appreciate their
willingness to take time away from their own research programs to aid
in this effort at historical perspective taking. The logistics of assem-
bling papers delivered at several different gatherings and arranging for
additional new works for the volume have combined with our own geo-
graphical separation to cause some delays in publication. At times we
PREFACE xi
feared that the materials for this book would themselves have become
''historic" by the time of their publication. We appreciate the patience
and goodwill of our contributors in this regard. We also want to acknowl-
edge the dispatch with which publication has been facilitated once
the final manuscripts became available-thanks are due here to Eliot
Werner and Patrick Connolly and their colleagues at Plenum and to
Robert Brown and Cecil Reynolds, coeditors of Plenum's series Per-
spectives on Individual Differences. Finally, we salute the sustaining
interest and support of Ross Stagner throughout this project.
Editing a volume can sometimes turn into a bit of a chore, but not
so in this case. We have been happy to serve as agents of a scientific
community delighted in sharing a worthy and inspiring heritage. That
community encompasses our contributors, our symposia audiences, and
now-we trust-the readers of this volume as well.
KENNETH H. CRAIK
ROBERT HOGAN
RAYMOND N. WOLFE
Contents
PART I. INTRODUCTION
Chapter 1 The 1937 Allport and Stagner Texts in

Personality Psychology . . .. . . . . . . .. . . . .. .. . . . . . 3
Kenneth H. Craik
Allport and Stagner: Personal Backgrounds and Intentions . . . . . . . 4
Allport and Stagner: Convergences and Divergences . . . . . . . . . . . . . 5
Cross-Generational Differences in Vantage Points . . . . . . . . . . . . . . . 7
A Major Contrast: An Analysis of the Individual per se versus
the Individual in Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Allport: The Individual Transcendent . . . . . . . . . . .. . . . . . . . . . . . . . . . 11
Stagner: The Individual in Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Planning and Organization of This Volume . . . . . . . . . . . . . . . . . . . . . . 14
Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
PART II. HISTORICAL AND PERSONAL

BACKGROUND OF THE 1937 TEXTS
Chapter 2 Fifty Years of the Psychology of Personality:

Reminiscences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Ross Stagner
Some Personal Backgrounds . .. . .. .. . . . .. . . . . . . . . . . . .. . . . . . .. . . 26
Personality and Psychoanalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
xiii
xiv CONTENTS
Views on Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Retrospections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Chapter 3 Allport's Personality and Allport's Personality . . 39

Alan C. Elms
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Allport's Encounter with Freud .. . . . .. .. . . . . . . . . . . . .. . . . .. . .. . . 40
Allport's Theory of the Clean Personality . . . . . . . . . . . . . . . . . . . . . . . 42
The Evolution of a Classic Textbook .. . . . .. . . . .. . . . .. . . . . . . . . .. . 46
Not a Little Boy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Allport as Unique Individual................................... 50
Allport as an Adult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Chapter 4 Allport and Murray on Allport's Personality:

A Confrontation in 1946-1947.................. 57
M. Brewster Smith
Allport on Allport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
The Neglect of Culture . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . .. . . . 58
Uniqueness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Issues of Motivational Theory . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . 59
Murray on Allport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Allport in Rejoinder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Commentary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
PART III. CURRENT STATE OF PERSONALITY

PSYCHOLOGY AND ITS TEXTS
Chapter 5 Pattern and Organization: Current Trends

and Prospects for the Future . . . . . . . . . . . . . . . . . . 69
Lawrence A. Pervin
The Units of Personality and Their Organization . . . . . . . . . . . . . . . . 70
Concepts of Trait and Motive in Allport and Stagner . . . . . . . . . . 70
The Goal Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Affect.............................................. .......... 76
The Organization of the Self .. .. . .. . . . . . . . . .. . .. . . . .. . .. .. . . . 79
CONTENTS XV
CUITent Status and Prospects for the Future . . . . . . . . . . . . . . . . . . . 80

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Chapter 6 The Continuing Relevance of Personality

Theory .......................................... 85
Salvatore R. Maddi
Psychology, the Fragmented Discipline . . . . . . . . . . . . . . . . . . . . . . . . . 86
The Prevalence of Warring Factions . . . . . . . . . . . . . . . . . . . . . . . . . 86
The Prevalence of Middle-Level Theorizing . . . . . . . . . . . . . . . . . . . 89
Overspecialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Personality Theorizing as One Antidote . . . . . . . . . . . . . . . . . . . . . . . . . 91
Tasks and Components of Personality Theorizing . . . . . . . . . . . . . 92
Integrating Potential of Personality Theorizing . . . . . . . . . . . . . . . 94
The Empirical Spirit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Teaching Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
A Final Word . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Chapter 7 It's Time to Put Theories of Personality in

Their Place, or, Allport and Stagner Got It
Right, Why Can't We? . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Gerald A. Mendelsohn
The Distinction between Theorizing about Personality and
Theories of Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
The Genres of Textbooks in Personality . . . . . . . . . . . . . . . . . . . . . . . . 105
The Critique of Theories of Personality . . . . . . . . . . . . . . . . . . . . . . . . . 107
Questionable Scientific Standing of the Theories . . . . . . . . . . . . . . 107
Questionable Contemporary Relevance of the Theories . . . . . . . . 109
Questionable Heuristic Value of the Theories . . . . . . . . . . . . . . . . . 111
What Is To Be Done? ......................................... 113
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
PART IV. PRESENT-DAY PERSPECTIVES

ON BASIC ISSUES
Chapter 8 Science and the Single Case . . . . . . . . . . . . . . . . . . . 119
Irving E. Alexander
Study of the Individual: Inhibiting Forces . . . . . . . . . . . . . . . . . . . . . . 121
Study of the Individual: Sustaining Forces ...................... 123
xvi CONTENTS
The Current Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Chapter 9 Describing Lives: Gordon Allport and the

"Science" of Personality ....................... 131
Bertram J. Cohler
Allport's Approach to the Study of the Person . . . . . . . . . . . . . . . . . . 133
The Interpretive Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
The Interpretive Thrn and the Nomothetic-Idiographic
Debate ................................................ 136
Personal Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Allport's Construal of Personality Structure . . . . . . . . . . . . . . . . . . . . . 139
Traits ..................................................... 139
Realization of Personal Integrity: The Proprium . . . . . . . . . . . . . . 140
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
References ................................................... 143
Chapter 10 Gordon Allport and "Letters from Jenny" ..... 147

David G. Winter
Gordon Allport and "Glenn" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Understanding the Case of Jenny .............................. 150
"Jenny" and Allport's Theoretical Ideas ......................... 152
Conflict and Mediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
The Idiographic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Therapeutic Skepticism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Normal and Abnormal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Functional Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Jenny and Ross in Gordon Allport's Life ........................ 157
Jenny as Mother ........................................... 158
Ross as Double . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Chapter 11 Allport's Personal Documents:

Then and Now . ................................. 165
Lawrence S. Wrightsman
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
The Purposes of Personal Documents in Psychological Research . . 166
Theoretical Perspectives-Allport's and the Present . . . . . . . . . . . . . 170
Methodologies-Allport's versus the Present . . . . . . . . . . . . . . . . . . . . 171
CONTENTS xvii
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
References ................................................... 174
Chapter 12 Conceptions of Self and Identity: A Modern

Retrospective on Allport's View . . . . . . . . . . . . . . . 177
Roy F. Baumeister
Allport's Views in Retrospect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Allport's Integrative Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
The Natural Self . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
The Conceptual Self . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
The Action Self . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Chapter 13 Current Status of the Motive Concept . . . . . . . . 187

Robert A. Emmons
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Other Current Approaches .................................... 190
Recommendations ............................................. 192
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Chapter 14 The Ability to Judge Others from Their

Expressive Behaviors
Bella M. DePaulo
The Consistency of Expressive Behaviors . . . . . . . . . . . . . . . . . . . . . . . 197
The Deliberate Regulation of Expressive Behaviors . . . . . . . . . . . . . 198
Allport's Six Questions about the Face . . . . . . . . . . . . . . . . . . . . . . . . . 199
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Chapter 15 Judgments of Personality and Personality

Itself ............................................ 207
David C. Funder
Allport on Personality Judgment ............................... 207
Research on the Process of Judgment . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Research on lnte:rjudge Agreement ............................ 210
Criteria for Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
The Accuracy Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
xviii CONTENTS
Chapter 16 Gordon Allport: Father and Critic of the

Five-Factor Model .............................. 215
Oliver P. John and Richard W. Robins
Allport-Father of the Five-Factor Model ...................... 216
Allport and Odbert's "Psycholexical Study" ................... 216
Replicating Allport and Odbert in German:
A Prototype Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Reducing the Semantic Nightmare: Five Broad Dimensions
Underlying Trait Terms ................................... 219
The Empirical Basis of the Five Factors ........................ 221
Research in the Lexical Tradition ............................ 221
Research in the Questionnaire Tradition . . . . . . . . . . . . . . . . . . . . . . 222
The Emerging Consensus and Heuristic Potential of the
Big Five . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Conceptual Status of the Factors: From the Big Five to the
Five-Factor Model ........................................ 223
Allport-Critic of the Five-Factor Model ....................... 225
Trait Theory and Personality Theory . . . . . . . . . . . . . . . . . . . . . . . . 225
Description and Explanation ................................ 225
Nomothetic and Idiographic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Molar Dimensions and Individual Traits ...................... 230
Conclusions .................................................. 231
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
Chapter 17 To Predict Some of the People More of

the Time: Individual Traits and the
Prediction of Behavior ......................... 237
Peter Borkenau
Allport's Critique of Hartshorne and May . . . . . . . . . . . . . . . . . . . . . . . 238
Bern and Allen's Elaboration of Allport's Ideas .................. 240
Was Bern and Allen's Study Truly Idiographic? .................. 242
The Replicability of Bern and Allen's Findings . . . . . . . . . . . . . . . . . . 244
Trait Relevance or Trait Extremity? . . . . . . . . . . . . . . . . . . . . . . . . . 244
Who Are the Trait-Consistent Subjects? ...................... 245
Was the Search for Moderators Successful? ................... 246
Conclusions .................................................. 247
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
CONTENTS xix
Chapter 18 The Scientific Credibility of Commonsense

Psychology ..................................... 251
Garth J. 0. Fletcher
A Realist Theory of Psychological Science . . . . . . . . . . . . . . . . . . . . . . 252
Truth as a Scientific Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
The Relation between Theory and Data . . . . . . . . . . . . . . . . . . . . . . 253
A Generative Concept of Causality . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Evaluating the Layperson's Social Cognition: Scientist
or Simpleton? ............................................ 255
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Folk Personality-Social Psychology Theories: Touchstone
or Crock? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Dispositions as Causal Explanations . . . . . . . . . . . . . . . . . . . . . . . . . 262
Caveats and Conclusions ...................................... 265
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
Chapter 19 A Commonsense Approach to Personality

Measurement ................................... 269
Raymond N. Wolfe
Veridicality ................................................... 272
Devising Items ............................................... 275
Preliminary Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
From the Theory to the Item . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
From Itemmetrics to Psychometrics . . . . . . . . . . . . . . . . . . . . . . . . . 278
From the Item to the Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Scale Breadth versus Scale Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Constraints on the Kinds of Constructs That Can
Be Measured . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
PART V. EPILOGUE
Chapter 20 An Optimistic Forecast . . . . . . . . . . . . . . . . . . . . . . . . 293

Robert Hogan
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
XX CONTENTS
Author Index ............................................... 299
Subject Index .............................................. 307

PART ONE
INTRODUCTION
CHAPTER ONE
The 1937 Allport and Stagner

Texts in Personality Psychology
KENNETH H. CRAIK
In August 1937, Gordon W. Allport and Ross Stagner completed and

dated the prefaces to their distinctive new textbooks. Allport's text-
book was entitled Personality: A Psychological Interpretation and was
published by Henry Holt in New York, with a later English edition by
Constable & Company issued in London in 1949. Stagner's textbook
was entitled Psychology of Personality and was published in New York
and London by McGraw-Hill. In 1961, Allport prepared a revision that
preserved the overall structure of his 1937 volume but with major re-
writing and a new title: Pattern and Growth in Personality. Stagner's
textbook continued steadily through four editions (1937, 1948, 1961,
1974) with substantial revisions, especially between the first and sec-
ond editions.
The identity of personality psychology as an empirically based sci-
entific field was heralded by the appearance of these two textbooks.
Each in its own way brought the considerable body of pertinent re-
search from the pre-1937 era together into the formulation of a coher-
ent field of inquiry (Craik, 1986). The present volume derives from
celebrations of the golden anniversary of the publication of these two
books and offers a set of commentaries on their origins, distinctive
themes, and topical contents. Before describing the development and
KENNETH H. CRAIK • Institute of Personality and Social Research, University of Cali-

fornia at Berkeley, Berkeley, California 94720.
Fifty Years of Personality Psychology, edited by Kenneth H. Craik et al. Plenum Press,
New York, 1993.
3
4 KENNETH H. CRAIK
organization of this volume, it is helpful to reintroduce briefly these

original textbooks of personality psychology to modern readers.
ALLPORT AND STAGNER:

PERSONAL BACKGROUNDS AND INTENTIONS
In the 1920s and 1930s, an increasingly vigorous, empirically ori-

ented research program addressing issues of personality was emerging
in the United States and Europe. Allport (1968) suggests that his 1924
course at Harvard University, "Personality: Its Psychological and So-
cial Aspects," may have been the first undergraduate personality
course in the United States. In any case, offerings of personality
courses were becoming widespread during the subsequent decade and
the need for adequate textbooks was evident by the mid-1930s.
Allport later recalled that from the time he completed his Ph.D.
dissertation at Harvard in 1922 on traits of personality he had been
"haunted with the idea that I should write a general book on personal-
ity" (Allport, 1968, p. 392). The project began to take form during his
four years on the faculty at Dartmouth College (1926-1930). Support
from Richard Cabot, a Boston Brahmin philanthropist and professor of
social ethics at Harvard, provided Allport with a semester of free time
in 1936 to complete Personality: A Psychological Interpretation. All-
port recalls that "I did not write the book for any particular audience.
I wrote it simply because I felt I had to define the new field of the
psychology of personality as I saw it" (1968, p. 394). When he com-
pleted his textbook, Allport was almost 40 years old (born November
11, 1897), 15 years beyond his Ph.D. degree, and an associate professor
at Harvard University.
Ross Stagner received his Ph.D. from the University of Wisconsin
in 1932, with a dissertation in which he demonstrated that the relation
between academic aptitude and achievement is moderated by personal-
ity traits (Stagner, 1933). As he reports in his contribution to this vol-
ume, Stagner's final commitment to write a textbook came when he
was teaching the course on personality psychology during his first full-
time appointment in 1935 at the University of Akron: "Seeking a possi-
ble text, I found that none was available, and decided to write my own"
(Stagner, Chapter 2, this volume). The project enjoyed the encourage-
ment of a McGraw-Hill publisher's representative and drew upon mate-
rials Stagner had already gathered in 1932-1933 when he had been
awarded a Social Science Research Council fellowship at the University
of Wisconsin (Stagner, 1937, p. ix). When he completed his textbook,
THE 1937 ALLPORT AND STAGNER TEXTS 5
Stagner was 28 years old (born June 15, 1909), five years beyond his
Ph.D. degree, and an assistant professor at the University of Akron.
Allport maintained his association with Harvard University until
his death in 1967. Stagner's academic career also took him to Dart-
mouth College, then to the University of Illinois, and finally to Wayne
State University, where he is professor emeritus. As we well know,
both Allport and Stagner subsequently enjoyed eminent scientific ca-
reers: Allport in personality psychology and social psychology, Stagner
in personality psychology and industrial-social psychology. For exam-
ple, Allport was elected president of the American Psychological Asso-
ciation (1939-1940) and Stagner was elected president of APA's
Division of Personality and Social Psychology (1960-1961) as well as its
Division of Industrial (now Industrial-Organizational) Psychology
(1965-1966).
ALLPORT AND STAGNER:

CONVERGENCES AND DIVERGENCES
Although generated independently, these two pioneering textbooks

in personality psychology display a number of important similarities, as
well as certain noteworthy differences. The most important shared
characteristic of the two volumes is their authors' self-assured formula-
tion of personality psychology as a coherent and distinctive field of
inquiry. Allport sets out "to gather into a single comprehensive survey
the most important fruits of the psychological study of personality" and
"to supply new coordinating concepts and theories" (1937, p. ix) for this
purpose. Stagner presents "a systematic set of concepts on which to
hang the framework of trait psychology, measurement procedures, etc."
(1937, p. viii).
Just five years earlier, Gardner Murphy and Friedrich Jensen
(1932) had published a volume entitled Approaches to Personality:
Some Contemporary Conceptions Used in Psychology and Psychiatry.
In it, they offered an eclectic assortment of points of view, drawing
upon Gestalt and behaviorist psychology, psychoanalysis, analytic
(Jungian) and individual (Adlerian) psychology, and genetic-develop-
mental notions from the field of child guidance. Murphy and Jensen
make explicit their orienting conclusion that "The confusion in the con-
temporary psychology of personality is considerable" (p. vi) and assert
that ''We do not believe that anyone today can seriously undertake to
say that he knows what personality is" (p. x).
In contrast, Allport (1937) moves decisively through an array of 49
6 KENNETH H. CRAIK
definitions of personality and confidently introduces his own, now well-

known version. And Stagner (1937) opens his preface by claiming that
"The material which may legitimately be included in a treatment of the
psychology of personality has grown too large to be brought within the
compass of a single volume of reasonable size" (p. vii) and then pro-
ceeds briskly to list the principles of exclusion and inclusion for his
textbook. Given these aims, Allport and Stagner generated organiza-
tions for their textbooks that embody their 1937 identifications of the
inherent structure and fundamental issues of the field of personality
psychology.
Allport's volume consists of five major parts. Part I deals with his
approach to personality and includes chapters on (1) psychology and
the study of individuality, (2) defining personality, and (3) a brief his-
tory of characterology. Part II covers the development of personality,
with chapters on (1) foundations of personality, (2) basic aspects of
growth, (3) the self and its constraints, (4) the transformation of mo-
tives, and (5) the mature personality. Part III focuses on the structure
of personality, with chapters on (1) the search for elements, (2) the the-
ory of identical elements, (3) the theory of traits, (4) the nature of
traits, and (5) the unity of personality. Part IV, on the analysis of per-
sonality, brings together chapters on (1) a survey of methods, (2) com-
mon traits: psychography, (3) analysis by ratings, tests, experiments,
and (4) expressive behavior. The concluding Part V, on understanding
personality, consists of chapters on (1) the ability to judge people,
(2) inference and intuition, and (3) the person in psychology.
Stagner's book consists of four sections. Section I, the introduction,
also includes two chapters on methods, one dealing with analytic ap-
proaches, including behavior observations, ratings, free associations
and scales, and one dealing with the study of the total personality,
including clinical formulations, life histories, and sociocultural contexts.
Section II addresses the descriptive psychology of personality and in-
cludes chapters on (1) the nature of personality structure, (2) basic re-
actions: feelings and emotions, (3) personality acquisition: simpler
forms of learning, (4) personality acquisition: complex forms of learn-
ing, (5) implicit traits of personality, (6) character, (7) attitudes and val-
ues, (8) the overt level of personality structure, and (9) type theories of
personality. Section III treats the dynamics of personality, with chap-
ters on (1) appetites and aversions, (2) theories of dynamics: Freud,
Adler, Lewin, and (3) a cultural interpretation of motivation. The final
section contains eight chapters that review factors that shape personal-
ity: (1) biological determinants of personality, (2) social determinants of
personality: the family (i), (3) social determinants of personality; the
family (ii), (4) play and recreation relationships, (5) personality and the
school system, (6) economic conditions affecting personality, (7) person-
ality and patterns of culture (a chapter contributed by Abraham H.
Maslow), and (8) the personality and social values.
The basic organization of these two textbooks was continued in
their subsequent editions. In Pattern and Growth in Personality, All-
port (1961) retains five major parts: I. An approach to personality;
II. Development of personality; III. Structure of personality; IV. As-
sessment of personality; and V. Understanding personality. Some minor
changes occur at the chapter level; for example, the addition of a chap-
ter on culture, situation, and role. In his third edition, Stagner followed
Allport by adding a fifth section on development, through some reas-
signment of chapters: I. Introduction; II. Development; Ill. Descrip-
tion; IV. Dynamics; and V. Determinants.
Thus, the 1937 Allport and Stagner textbooks both embody an as-
surance regarding what the field of personality is all about. Further-
more, both authors organized their textbooks according to their own
personal and comprehensive formulation of the intellectual structure of
the field and its array of fundamental issues. Finally, in light of pres-
ent-day trends in the merchandising of textbooks, it is worth noting
that both authors risked a high level of difficulty in the expositions
presented in their books. Allport recalled his "desire to avoid jargon
and try to express my thoughts in proper and felicitous English. The
result was that some readers regarded the book as difficult and preten-
tious," but he also notes that others labeled it as "classic" (Allport,
1968, p. 394). In his preface, Stagner (1937) warns the reader that he
had set out to offer a coherent point of view on the field and "I ac-
cepted, in doing so, the danger of making the book too difficult" (p. viii).
CROSS-GENERATIONAL
DIFFERENCES IN VANTAGE POINTS
Perhaps it is indicative of the then early stage of personality

psychology's development that both Allport and Stagner began to con-
sider writing a textbook so soon after receiving their doctoral degrees.
Allport was thinking about the textbook project in 1922, at age 24;
Stagner set about the task in 1935, at age 26, and worked more rapidly
on it.
Hints of a generational difference in their perspectives and aims
can be discerned in the volumes they produced in 1937. Allport ap-
proached the task as a first-generation pioneer with the goal of justify-
8 KENNETH H. CRAIK
ing the new field's existence. As a second-generation personality psy-

chologist, Stagner could take the field's existence as a given and set
about fashioning a textbook in more standard form.
The tone of Allport's preface is that of a founding member of a
new field. After all, he had quite likely completed the first empiri-
cal American dissertation on the component traits of personality in
1922 and probably had offered the first undergraduate course in
the field in 1924. Within the context of academic psychology of that
time, Allport considered his person-centered formulation of a new
field as "thoroughly radical" (1937, p. 549). As a pioneer, he had to ask
himself: "But did I have enough courage and ability to develop my
deviant interests?" (1968, p. 385). Thus, here was Allport, with over
a decade of experience observing the field as it emerged in the 1920s,
setting out to write, as an early explorer, "a guide book that will define
the new field of study-one that will articulate its objectives, formulate
its standards, and test the progress made thus far" (p. vii; italics in the
original).
In doing so, Allport was particularly dedicated to establishing the
distinctiveness of the field within a broad historical, interdisciplinary,
and scholarly context. One problem to which Allport appears to have
been responding was highlighted by A. A. Roback's A Bibliography of
Character and Personality (1927). In this comprehensive, 340-page
compendium, Roback emphasizes the breadth and diversity of then
available approaches to the study of personality, ranging from anthro-
pology, biography, history, and literature to philosophy, psychoanalysis,
psychiatry, religion, and sociology. Thus, it is no accident that Allport
added the subtitle, A Psychological Interpretation, to qualify his pri-
mary title, Personality. To a significant degree, he saw his mission as
one of making a persuasive case for the intellectual and scientific dis-
tinctiveness of the psychology of personality.
Eleven years younger and coming on the scene a decade later,
Stagner appears to have found the field of personality psychology as
much more of a given, even though proper textbooks were still needed
for the courses being widely offered by the mid-1930s. In his preface,
Stagner (1937) explicitly takes on the task of differentiating a proper
textbook in the psychology of personality from an array of textbooklike
competitors. In stating his basis for selecting material for his volume,
he announces that he had "in almost all cases eliminated consideration
of therapeutic techniques, suggestions for self-improvement, hints to
parents and other aspects of what might be considered 'the applied
psychology of personality' " (p. vii).
Early examples of this self-help tradition include William M.
Thayer's Tact, Push, and Principle (1885), whose organization is fore-

cast in its title. Under "tact," Thayer presents separate chapters on
thoroughness, singleness of purpose, and observation (a combination of
social insight and practical intelligence); under "push," separate chap-
ters on decision and energy, perseverance, industry, economy of time
and effort, punctuality, and order; and under "principle," chapters on
character, conscience, honesty, benevolence, the Bible, and religion in
business. The volume is affectionately dedicated to its intended reader-
ship: "To the young men of the United States, facing difficulties, sub-
ject to reverses, unassisted by influence or capital, the brave and
hopeful of success." A later, assessment-oriented volume along similar
lines is John T. Miller's (1922) Applied Character Analysis in Human
Conservation, dealing with vocational types, guided primarily by phre-
nological theory, and dedicated to "All who use the true science of mind
in human improvement, physically, socially, intellectually, morally and
spiritually."
Volumes published closer in time to the appearance of the 1937
Allport and Stagner textbooks often stressed philosophical or spiritual
issues (Brown, 1927), ways to deal with or avoid the development of
abnormalities in personality (Bagby, 1928; Gordon, 1928), and issues
concerning character and personality in such fields as child guidance,
education, and mental hygiene (Betts, 1937; Thorpe, 1938; Valentine,
1927). Some of the latter genre, such as Thorpe's Psychological Foun-
dations of Personality: A Guide for Students and Teachers (1938), drew
on much of the same research literature as did Allport and Stagner, but
possessed a practical rather than scientific orientation and were pri-
marily addressed to educators of our youth and students of education
rather than to students of psychology itself.
In summary, Allport's 1937 volume can be seen retrospectively as
a textbook and at the same time as more than a textbook. Allport
sought to articulate and justify the identity of a new field of scientific
inquiry and to establish the nature of its basic concepts and issues
(Pettigrew, 1990). This latter characteristic may primarily account for
the volume's continuing high frequency of citation. Stagner's more
classroom-oriented volume also considered the person within a social
framework, in contrast to Allport's treatment of the individual tran-
scendent. A perusal of most present-day personality textbooks as well
as the research literature of our field strongly suggests that, in seeking
to place the person within a cultural, institutional, and societal context,
Stagner was going against the grain of personality psychology, at least
as it has emerged in this country.
10 KENNETH H. CRAIK
A MAJOR CONTRAST: AN ANALYSIS OF THE INDIVIDUAL

PER SE VERSUS THE INDIVIDUAL IN SOCIETY
A basic divergence in the formulations guiding the 1937 Allport

and Stagner textbooks occurs in their treatment of the individual in
context. Allport was strongly committed to establishing personality
psychology as an intellectually and epistemologically distinct field of
inquiry. Perhaps for this reason, he concentrated on making a case for
the study of the individual as a primary scientific unit of analysis and
indeed, for personality psychology, as the fundamental unit of analysis.
Taking a different approach, Stagner devoted over a third of his text-
book to the task of understanding the individual within a societal con-
text. This emphasis distinguishes his volume not only from Allport's
but also from those of his contemporary "pre-textbook" competitors.
While some volumes may have highlighted such social factors as the
family, the school, or the general culture, none appear to have offered
the broad-based social institutional framework for the study of individ-
ual personality that is advanced by Stagner.
This contrast between the two textbooks is clearly illustrated by
their treatment of the upheavals associated with the Great Depression,
at its worst depths when these volumes were being completed. Reading
Stagner today, one gets a vivid sense of this catastrophe's impact on
individuals in the United States and Europe. A chapter reviewing eco-
nomic conditions affecting personality deals explicitly with such topics
as the effects of unemployment on workers' morale, the influences of
parents' economic uncertainty on children and child rearing, and the
role of such "economic hazards" in the incidence of juvenile delin-
quency. In his preface, Allport acknowledges at the outset that he an-
ticipates being criticized for neglecting the important relationships
between personality and culture. Furthermore, in the one reference to
these wider events of his own times, Allport puts an individualistic
twist on the issue. He asks: "Why is it that in our times, when Western
culture is sadly disorganized, our personalities are not correspondingly
disorganized?" Allport concludes that "Cultural determinism is one of the
monosymptomatic approaches; it has a blind spot for the internal balanc-
ing factors and structural tenacity within personality'' (1937, pp. viii-ix).
Allport makes a well-substantiated claim that he provides a valu-
able historical orientation for the varied formulations concerning per-
sonality that have been advanced over the course of Western culture.
However, because of their differing approaches toward analyzing the
person within society, it is the Stagner volume that provides the
present-day reader with a sense of the historical context of the vol-
umes themselves. In contrast, Allport's textbook is written with a gen-

erality of tone that affords it a timeless, or at least ahistorical, aura.
Conceptually, Allport and Stagner take a similar stance in distin-
guishing between psychological and sociological levels of analysis. For
Allport, culture is important only to the extent that "it has become
interiorized within the person as a set of personal ideals, attitudes, and
traits. Likewise, culture conflict must become inner conflict before it
can have any significance for personality" (1937, p. viii). On this point,
Stagner concurs, asserting that "The object of our study is a single
human being.... Social systems, cultures, etc., have psychological real-
ity as, and only as, habits and beliefs of specific human beings" (1937,
pp. viii-ix). Nevertheless, they differ markedly in how central a place
they assign to the analysis of interactions between persons and their
society.
ALLPORT: THE INDIVIDUAL TRANSCENDENT
Allport, while not denying their role, narrowly circumscribed his

own treatment of societal factors influencing personality, for "the inter-
est of psychology is not in the factors shaping personality, rather in
personality itself as a developing structure" (1937, p. viii). Allport's
conceptual contributions were aimed at appreciating this relatively
enduring and unique organization and bringing it under appropriate
scientific analysis. First, his neuropsychic formulation of traits empha-
sized their status as "internally generalized, flexible, interdependent
dispositions" (pp. 559-560). Second, he distinguished between the con-
cept of individual traits and that of common traits; a common trait
provides a basis for comparison among persons but never "corresponds
exactly to the neuropsychic dispositions of individuals" (p. 300). Third,
he dealt thoroughly with the multiplicity of traits and motives of any
individual, in part as a means of highlighting the complex organization
specifically entailed by the structure of any individual's personality
(p. 560). Fourth, he developed the concept of the functional autonomy
of motives as a device for analyzing the high degree of differentiation
and specific multiplicity to be found in the adult personality structure
of an individual (p. 207). Fifth, he isolated the psychological study of an
individual's personality from any analysis of the person's social stimulus
value (May, 1932) and reputation in society at large. Sixth, he granted
primary emphasis to the task of understanding the unity of personality.
In his revised textbook, Allport (1961) still viewed the problem of the
organization of personality as our field's greatest conceptual challenge
12 KENNETH H. CRAIK
and continued to address it, notably through his notions of the

morphogenetic analysis of personal dispositions (p. 358) and the propri-
ate functions of personality.
Allport was explicit and almost provocative in eschewing system-
atic treatment of sociocultural factors (Allport, 1937, pp. viii-ix; Smith,
Chapter 4, this volume). He appears to have made a strategic decision
that his efforts to focus the spotlight on the study of the individual and
the analysis of personality organization could be best realized by a
deliberate neglect of the relations between persons and their society. A
secondary factor may have been Allport's initial tendency to view any
attention to this issue as entailing a position of cultural determinism
versus individualism; he gives little attention to the possibility of an
interactional model. The result of his strategy is a ringing declaration
of the integrity of each individual, at the expense of presenting a se-
verely decontextualized formulation of personality. This depiction is at-
tenuated somewhat in his 1961 volume, where he adds a chapter taking
note of cultural context, social role, and situation and the distinction
between individual and collective structures.
Given Allport's important contributions to both personality psy-
chology and social psychology, one might hypothesize that perhaps
Allport conceptually split his scientific identities-attending to the indi-
vidual as a personality psychologist and to the social environment as a
social psychologist. However, Allport's contributions to social psychol-
ogy have been identified from some theoretical perspectives as exces-
sively individualistic. For example, his conception of attitudes in
traitlike fashion can be contrasted to their formulation as more collec-
tive, social representations (Farr, 1981; Moscovici, 1984). In this sense,
Allport's basic credo pervaded his scientific orientation:
Thus, there are many ways to study man psychologically. Yet to
study him most fully is to take him as an individual. He is more
than a bundle of habits; more than a nexus of abstract dimensions;
more too than a representative of his species. He is more than a
citizen of the state, and more than a mere incident in the gigantic
movements of mankind. He transcends them all. (1937, pp. 566-567)
STAGNER: THE INDIVIDUAL IN SOCIETY
Stagner's commitment to a contextual perspective on personality is

evident in the organization and context of his textbook. Eight of its
twenty-three chapters are devoted to sociocultural factors. One of the
three chapters of the third section, on the dynamics of personality,
deals with a cultural interpretation of motivation. In this exposition,

Stagner rejects a generalized energy conception of drive in favor of a more
particularistic formulation of individual motivational patterns as sensi-
tive to cultural and other environmental influences. He cites cross-cultural
and historical variations in the manifestations of sexuality, will-to-power,
acquisitiveness, and social approval in support of this framework.
The fourth and longest section of the volume examines the determi-
nants of personality. In this way, Stagner incorporates questions regard-
ing the shaping of personality that Allport had explicitly rejected from
consideration. One of its eight chapters reviews biological determinants
and the remaining seven address an array of sociocultural determinants.
The family as an institution is covered in two chapters. The first
focuses on child rearing as influenced by parental affection, quarrels
and jealousy, problems of discipline, sibling relationships, parent substi-
tutes, and adolescent concerns regarding dependence-independence.
The second deals with the adults of the family, examining the role of
personality in marital satisfaction and conflict, mate similarity, illness
and drinking as escapes from marital conflict, and the relation of per-
sonality to parenting styles. The social context of play and recreation
are highlighted as important extrafamilial settings for competition in
the realms of physical prowess, sociability, and sexual attractiveness as
well as settings for cooperation through team play and membership in
gangs and cliques. The role of reading and the movies in identity for-
mation is also discussed as well as the increasing commercialization of
recreation. A chapter on the school system covers the topics of person-
ality and academic achievement, the role of teachers' personalities, and
the influence of the school on economic and social attitudes through the
curriculum, and more indirectly through power relations and the way
in which controversial issues of the day are handled. We have already
noted the extent to which a chapter on the impact of economic condi-
tions on personality reflected the urgent effects of the Great Depres-
sion. Maslow's chapter on culture and personality stresses the point of
cultural relativism in part as a counter to ethnocentric biases in concep-
tions of personality and also offers a review of cultural implications for
the concepts of abnormal personality and adjustment and a brief inter-
pretation of psychotherapy as social philosophy.
Stagner's final chapter on personality and social values is of partic-
ular interest for a present-day reader. He addresses the issue of levels
of analysis and offers an interactional model of the individual and so-
ciety. In this framework, he argues that individual personalities are
shaped more or less equally by sociocultural factors, on the one hand,
and individual factors (e.g., biological, hereditary, specific biographical
14 KENNETH H. CRAIK
events) on the other. Sociocultural systems are shaped predominantly

by other sociocultural forces (e.g., the effect of political change oneco-
nomic functioning) but also to some important extent by actions of
individual personalities (Stagner, 1937, pp. 431-434 and Fig. 24). This
model permits Stagner to incorporate systematic treatment of sociocul-
tural influences on personality in his textbook without being committed
to the unidirectional cultural determinism that so concerned Allport. At
the same time, it permits Stagner to incorporate a consideration of the
impact of individual actions on the sociocultural system. He cites the
"Russian experiment," with its combination of political dictatorship and
rejection of capitalism, as an instance of dramatic change brought about
in a social system.
Stagner is explicit about his interactional formulation:
Our purpose throughout has been to depict the development of the
human organism with its inherent tendencies and equipment, strug-
gling through an environment of things and people, trying, failing,
learning, viewing situations from successively different angles; the
description of a personality and its determination by cultural influ-
ences has been sketched. We should now ... see the beginning of a
new cycle in which the personality tries various ways of modifying
the environment. (1937, pp. 443-444)
Thus, the model encompasses processes by which the individual
emerges as an agent within the societal context. Specifically, when an
individual becomes conscious of his system of values, then
it becomes possible for him to achieve a certain psychological dis-
tance from his social setting and attempt to evaluate it without
confusion from his own set of pre-existing biases and prejudices.
He may then be able to see clearly its own virtues and vices, and
then decide calmly whether or not it should be preserved, reformed
or remade. (p. 443)
This interaction model is formulated primarily at the institutional
level and thus remains largely within the tradition of personality system-
social system analysis (Smelser & Smelser, 1964) rather than within the
more recent framework of efforts to analyze everyday person-environ-
ment interactions (Walsh et al., 1992). Nevertheless, Stagner's textbook
had a modern interactive and contextual sound.
PLANNING AND ORGANIZATION OF THIS VOLUME
In 1987, several celebrations of the golden anniversary of the 1937

Allport and Stagner textbooks took place. At the University of Califor-
nia in Berkeley, the Institute of Personality Assessment and Research

(IPAR)* organized a special colloquium series commemorating the oc-
casion. Ross Stagner took part in this series, delivering an address on
"Fifty Years of Development in the Psychology of Personality." Other
speakers in this series included Irving E. Alexander, Bertram J.
Cobler, Lewis R. Goldberg, Ravenna Helson, Robert Hogan, Salvatore
R. Maddi, Lawrence A. Pervin, and Paul Wink. In addition, a special
session was devoted to Allport and his textbook. In this context, Allan
C. Elms presented a talk on "Allport's Personality and Allport's Per-
sonality," which was followed by a roundtable discussion by five former
graduate students at Harvard during Allport's tenure there. The dis-
cussants (with the year of their Ph.D.) included the late Sheldon J.
Korchin (1946), Gardner Lindzey (1949), Pavel Machotka (1962), R.
Nevitt Sanford (1934), and M. Brewster Smith (1947), with Ernest R.
Hilgard (Yale, 1930) moderating.
Subsequently, Robert Hogan and I decided to take a modified
version of this successful "show" to Broadway, as the annual meet-
ings of the American Psychological Association were scheduled to
take place in New York City in August. Division 1 (Division of Gen-
eral Psychology) generously afforded us the opportunity to organize
two consecutive two-hour sessions, which were heavily attended.
For this occasion, Alan Elms presented a modified version of his talk
on "Allport's Personality and Allport's Personality," Salvatore Maddi
spoke on "The Continuing Relevance of Personality Theory," and
Lawrence A. Pervin reviewed "Current Trends in Personality The-
ory." The remainder of the two sessions was organized around five
topics suggested by the contents of Allport's 1937 textbook. On the
topic of the self, Stephen R. Briggs spoke on "Constructing the Self''
and Anthony G. Greenwald on "How Shall the Self Be Conceived?"
On the topic of the single case, Irv Alexander dealt with "Science
and the Single Case," and Lawrence S. Wrightman with "Allport's
Personal Documents: Then and Now." On the issue of psychological
maturity, Roy F. Baumeister addressed "Modern Conceptions of
Identity" and Ravenna Helson addressed "Allport's Conception of
Maturity." On motivation, Robert A. Emmons reviewed the "Current
Status of the Motive Concept" while Christopher Langston and
Nancy Cantor analyzed "Life Tasks and Motivation." Finally, on the
issue of judging other persons, David C. Funder spoke on "Judg-
* In conjunction with the expansion of its research program to encompass the study of
persons within their cultural, institutional, organizational, and societal contexts, this
institute was renamed in 1992 as the Institute of Personality and Social Research (IPSR).
16 KENNETH H. CRAIK
ments of Personality" and Bella M. DePaulo on "The Ability to Judge

Others."
In developing this book, we were able to enlist the editorial collab-
oration of Raymond N. Wolfe and contributions from most of the partic-
ipants in the APA symposia. Due to prior commitments, we lost the
involvement of Cantor and Langston, Greenwald and Helson. However,
relevant contributions from Peter Borkenau, Bertram J. Cohler, Garth
J. 0. Fletcher, Oliver P. John, Richard W. Robins, Gerald A. Mendel-
sohn, M. Brewster Smith, David G. Winter, and Ray Wolfe have served
wonderfully to augment the materials from the APA sessions.
CONTENTS
The volume is organized around three major themes. First, the

personal and historical contexts of the 1937 Allport and Stagner text-
books are addressed in contributions by Ross Stagner, Alan C. Elms,
and M. Brewster Smith. Stagner provides his own perspective on his
and Allport's textbooks gained from over 50 years of personal experi-
ence in the field. Elms relates the structure and emphases of Allport's
textbook to characteristics of his background and personality. Through
notes taken from meetings in 1946-1947 at Harvard, Brewster Smith
offers a glimpse of Allport's personal and theoretical relations with a
colleague who also contributed an important volume to personality psy-
chology in the late 1930s: Henry A. Murray, who, with his research
team, had published Explorations in Personality in 1938.
Second, analysis of the current state of personality psychology and
its textbooks is offered by Lawrence A. Pervin, Salvatore R. Maddi,
and Gerald A. Mendelsohn. In his stocktaking essay, Pervin selects the
central issues of the pattern and organization of personality as a frame-
work for examining current trends and prospects for our field. He also
notes how Stagner's conceptual orientation in the subsequent editions
of his textbook was influenced by his close reading of Kurt Lewin's
Dynamic Theory of Personality, which had been published in 1935.
Maddi presents a spirited argument in favor of comprehensive theoriz-
ing in personality psychology. Specifically, he continues the textbook
theme of our volume by advocating a model of the personality textbook
that emphasizes the integrative function of comparative analysis of al-
ternative grand theories. His position appears to be: Preach what per-
sonality psychology might ideally become, not the fragmented manner
in which it is now practiced. In contrast, Mendelsohn discerns a serious
gap between these grand theories of personality and the nature of con-
temporary personality research. He opts instead for a textbook format

exemplified by the 1937 Allport and Stagner textbooks, which is organ-
ized around basic topics and fundamental issues in personality psychol-
ogy, such as units of analysis, methods, development, determinants, and
dynamics. In opposition to Maddi, Mendelsohn's position is explicitly
asserted: Preach what you practice.
The third theme of our volume constitutes a present-day perspec-
tive on some of the basic issues in personality psychology, as suggested
by the themes and organization of Allport's 1937 textbook:
1. The individual and the single case. In this section, Alexander
deals with the scientific issues of single-case analysis and offers a suc-
cinct historical perspective on the fate of this issue over the past five
decades of personality research. Cohler presents a review of current
work on the description and understanding of individual lives as it
bears on Allport's formulations. Drawing on biographical information
regarding the relations between Allport and "Jenny," Winter suggests
new interpretations of his major single-case analysis, the Letters from
Jenny (Allport, 1965; Anonymous, 1946). Finally, Wrightsman provides
a general summary of Allport's work on the analysis of personal docu-
ments and reviews current research activities using this method.
2. Motives and the self. Baumeister places Allport's views on the
self in the perspective of current formulations regarding the natural
self (the knower and the body), the conceptual self (self-representations
and identity), and the action self (self-referenced motives). Emmons
historically traces analyses of the concept of motive from its treatment
in the 1937 Allport and Stagner textbooks to its present revival in
personality psychology.
3. Judging persons. DePaulo and Funder revisit Allport's analy-
sis of expressive behavior and the ability to judge others. DePaulo
discusses recent work on on-self-conscious expressive behavior and
the issue of deception. During the 1960s and 1970s, psychologists
seemed to be discouraged about the prospects of examining the abil-
ity to judge others because of the difficulties encountered in establish-
ing criteria regarding the target persons. Funder offers encouragement
that convergently valid criteria can indeed serve as a basis for study-
ing the accuracy of individuals' judgments of the personalities of other
persons.
4. Personality assessment and prediction. The 1937 Allport and
Stagner textbooks both highlighted the importance of trait concepts
and their measurement. In this section, John and Robins place the 1936
Allport-Odbert psycho-lexical study of trait names (Allport & Odbert,
18 KENNETH H. CRAIK
1936) in the context of subsequent research on trait taxonomies derived

from the ordinary language of personality description. In so doing, they
advance a clarifying distinction between the empirically derived "big
five" factor structure of personality descriptions, on the one hand, and
a theoretical "Five-Factor Model" of personality traits and personality
structure that is increasingly being used as a conceptual guide for or-
ganizing research in personality, on the other.
Borkenau relates Allport's 1937 treatment of the consistency-
specificity controversy, raised by Hartshorne and May (1928, 1929) and
Hartshorne, May, and Shuttleworth (1930) to the ways in which it has
been addressed more recently in the form of the person-situation con-
troversy. Fletcher examines Allport's stance on naive ''heuristic real-
ism" regarding the nature of traits. He locates the issue within more
recent and broader discussion concerning the relevance that studying
laypersons as naive scientists may hold for the more formal process of
personality and cognitive social theorizing. In formulating the field of
personality psychology, Allport noted that "I have tried to make a spe-
cial ally of common sense" (Allport, 1937, p. viii); and the same spirit
imbues his approach to trait concepts and methods of assessment.
Wolfe identifies an emerging common-sense orientation in the construc-
tion and interpretation of self-report scales for assessing trait concepts
and discusses the extent and limits of this current trend in personality
measurement.
Finally, in the epilogue, Hogan offers a closing perspective on the
current state of personality psychology as a scientific enterprise.
In hindsight, it now strikes us that by planning the third section of
this volume around the themes and organization of Allport's 1937 text-
book, we have continued an unduly narrow orientation, through placing
greater emphasis on the individual as transcendent than on the individ-
ual in societal context. But our goal has not been to review the past 50
years of personality research; comprehensive handbooks in personality
psychology are more appropriate to that task (Briggs et al., forthcom-
ing; Pervin, 1990). Instead, our intention is to foster a greater appreci-
ation of our field's past and trends in its development. We will consider
this volume a success if it enhances recognition of the impact of these
pioneering personality textbooks and sustains the respect and grati-
tude that our field has already bestowed on their authors, highlights
the usefulness of systematic discussion concerning the nature and influ-
ence of contemporary textbooks in personality psychology, and encour-
ages greater historical perspective on the development of personality
psychology as a field of inquiry.
ACKNOWLEDGMENTS. I wish to thank R. T. Hogan, 0. P. John, G. A.

Mendelsohn, R. W. Robins, W. M. Runyan, and R.N. Wolfe for their
helpful comments on earlier drafts of this chapter.
REFERENCES
Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt.

Allport, G. W. (1961). Pattern and growth in personality. New York: Holt, Rinehart and
Winston.
Allport, G. W. (1965). Letters from Jenny. New York: Harcourt, Brace and World.
Allport, G. W. (1968). An autobiography. In G. W. Allport, The person in psychology:
Selected essays by Gordon W. Allport (pp. 376-409). Boston: Beacon Press.
Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study. Psychological
Monographs, 47 (Whole No. 211).
Anonymous (1946). Letters from Jenny. Journal of Abnormal and Social Psychology, 41,
315-350, 449-480.
Bagby, E. (1928). The psychology of personality: An analysis of common emotional
disorders. New York: Holt.
Betts, G. H. (1937). Foundations of character and personality: An introduction to the
psychology of social adjustment. New York: Bobbs-Merrill.
Briggs, S. R., Hogan, R. T., & Jones, W. (forthcoming). Handbook of personality psychol-
ogy. New York: Academic Press.
Brown, W. (1927). Mind and personality: An essay in psychology and philosophy. New
York: Putman's.
Craik, K. H. (1986). Personality research methods: An historical perspective. Journal of
Personality, 54, 18-51.
Farr, R. M. (1981). On the nature of human nature and the science of behavior. In P.
Heelas & A. Lock (Eds). Indigenous psychologies: The anthropology of the self
(pp. 303-317). New York: Academic Press.
Gordon, R. G. (1928). Personality. London: Kegan Paul, Trench, Trubner.
Hartshorne, H., & May, M. A. (1928). Studies in the nature of character: Volume 1.
Studies in deceit. New York: Macmillan.
Hartshorne, H., & May, M. A. (1929). Studies in the nature of character: Volume 2.
Studies in service and self-control. New York: Macmillan.
Hartshorne, H., May, M. A., & Shuttleworth, F. K. (1930). Studies in the nature of
character: Studies in the organization of character. New York: Macmillan.
Lewin, K. (1935). A dynamic theory of personality. New York: McGraw-Hill.
May, M. A. (1932). The foundations of personality. In P. S. Achilles (Ed.), Psychology at
work (pp. 81-101). New York: McGraw-Hill.
Miller, J. T. (1922). Applied character analysis in human conservation. Boston: Gorham
Press.
Moscovici, S. (1984). The phenomenon of social representation. In R. M. Farr & S.
Moscovici (Eds.), Social representation (pp. 3-69). Cambridge: Cambridge Univer-
sity Press.
Murphy, G., & Jensen, F. (1932). Approaches to personality: Some contemporary concep-
tions used in psychology and psychiatry. New York: Coward-McCann.
Murray, H. A. (1938). Explorations of personality. New York: Oxford University Press.
20 KENNETH H. CRAIK
Pervin, L.A. (1990). Handbook of personality: Theory and research. New York: Guilford
Press.
Pettigrew, T. F. (1990). A bold stroke for personality a half-century ago. Contemporary
Psychology, 35, 533-536.
Roback, A. A. (1927). A bibliography of character and personality. Cambridge, MA:
Sci-Art.
Smelser, N.J., & Smelser, W. T. (1964). Analyzing personality and social systems. InN.
J. Smelser & W. T. Smelser (Eds.), Personality and social systems (pp. 1-18). New
York: Wiley.
Stagner, R. (1933). Relation of personality to academic aptitude and achievement. Jour-
nal of Educational Research, 26, 648-660.
Stagner, R. (1937). Psychology of personality. New York: McGraw-Hill.
Thayer, W. M. (1885). Thct, push, and principle. Boston: Earle.
Thorpe, L. P. (1938). Psychological foundations of personality: A guide for students and
teachers. New York: McGraw-Hill.
Valentine, P. F. (1927). The psychology of personality. New York: Appleton.
Walsh, W. B., Craik, K. H., & Price, R. H. (Eds.) (1992). Person-environment psychology.
Hillsdale, NJ: Erlbaum.
PART TWO
HISTORICAL AND
PERSONAL BACKGROUND
OF THE 1937 TEXTS
CHAPTER TWO
Fifty Years of the

Psychology of Personality
Reminiscences
ROSS STAGNER
As we are fond of saying about psychology in general, the study of

personality has a long past but only a short history. The proposal to
specify 1937 as the date of origin for modern personality psychology
has a certain plausibility: Gordon Allport and I both published formal
textbooks on personality in that year, giving shape and academic
respectability to the field; and Henry Murray's Explorations in Per-
sonality (1938) came out less than 12 months later. It seems proper,
therefore, to designate 1987 as the 50th year of the psychology of
personality.
We have a recent precedent for doing this. The American Psycho-
logical Association declared the year 1879 as the birth date of scientific
psychology. At once, of course, critics noted the importance of various
contributions prior to that date; and the same is possible with regard to
the study of personality. The prehistory, or archaeology, of personality
study goes back to Galen and Plutarch. In 1927, A. A. Roback assem-
bled a bibliography of over 2200 titles dealing with personality; how-
ever, most of these were literary or speculative. Few could be called
empirical investigations (although case studies were common), and
Ross STAGNER • Department of Psychology, Wayne State University, Detroit, Michigan

48202.
New York, 1993.
23
24 ROSS STAGNER
even fewer had any finn anchoring in the mainstream of psychological

science. Of the sources Roback catalogued, writings by Freud, Jung,
and Adler stand out as the works that are still viewed as important.
Another set of important ancestral influences is found in the writ-
ings of Charcot and Janet, psychiatrists who sought to explain suggest-
ibility, dissociation, and symptom formation. By contrast, the Gennan
experimentalists we celebrated in 1979 (Wundt, Stumpf, Kiilpe, Eb-
binghaus, and others) ignored the phenomena nowadays labeled as per-
sonality. We note, of course, the disagreement between Wundt and J.
M. Cattell over the place of individual differences in a scientific psy-
chology; and it is appropriate to mention Hennann Rorschach, whose
inkblot test was designed to exclude meaning from the measurement of
personality-an effort stimulated by Wundt's distinction between
experience and meaning. And it seems only fair to mention the feeble
gesture of E. B. Titchener, that advocate of research on the "mind-in-
general," who included in one edition of his introductory textbook a
page dealing with the topic of temperament, using Galen's scheme of
the four humors. Titchener suggested a relationship between these and
such phenomena as speed of association and intensity of affect; how-
ever, he quickly dropped the subject and never returned to it.
It is also proper to mention the speculation of William James, who
could hardly ignore a phenomenon as ubiquitous as personality. Adopt-
ing an environmentalist stance, he speculated about the importance of
the social context in shaping emotions and social interactions. James
wrote
... at the age of 25 you see the professional mannerism settling
down on the young commercial traveler, on the young doctor, on
the young minister ... you see the little lines of cleavage running
through the character, the tricks of thought, the prejudices . . . by
the age of 30, the character has set like plaster, and will never
soften again. (1890, pp. 121-122)
I regret to have to add that James considered this molding of
personality a fine idea because ''it keeps different social strata from
mixing." It is surprisingly difficult to write about personality without
revealing your personal prejudices.
James acknowledged the influence of Charcot and Janet on his
thinking, but his emphasis on social detenninants links him to sociolog-
ical theorists such as George Herbert Mead and Charles Horton
Cooley. On the subject of the self, James seems to have embraced the
concept of "social stimulus value" while oscillating between emphasis
on unity of the self and dissociation of role-related selves.
FIFTY YEARS OF THE PSYCHOLOGY OF PERSONALITY 25
Properly speaking, a man has as many social selves as there are
individuals who recognize him and carry an image of him in their
mind. But ... we may practically say that he has as many different
social selves as there are distinct groups of persons about whose
opinions he cares (1890, p. 294)
In this statement James anticipates the views of Walter Mischel
and others to the effect that the self is no more than a response to a
social context (and its associated reinforcements), and Kurt Lewin's
treatment of the subjective personality as a kind of internalization of
group memberships.
I want to mention one other pre-1937 book because of its relevance
to a contemporary theoretical debate. English Bagby's Psychology of
Personality, published in 1928, reflected some psychoanalytic and some
behavioristic leanings. While its focus was on counseling, it raises
issues of general importance. The following case summary illustrates
some theoretical problems:
The patient, a university sophomore, sought assistance in connec-
tion with a strong impulse to gnaw the back of his right hand. The
tendency had existed for a period of two months and already a
large callous area had developed. The patient appeared to be quite
ashamed of his inability to secure control of this habit, and said that
he had been wearing a glove to conceal the scar, although the
weather had not been cold. (Acid was applied to his hand so that a
bitter taste would result when it was placed in his mouth.)
On the third day the young man reported that the inclination
to bite his hand was no longer troubling him. However, he called
attention to a new symptom. He found himself almost constantly
beset by moral worries. (These were illustrated by trifling points
such as choice of neckties, walking with friends on the campus, etc.,
over which he had worried to excess.)
The condition of moral uncertainty persisted for several days,
Finally, however, the patient came to report that his distressing
condition had completely disappeared and that he was once again
serene. But, in the course of the conversation, it was noted that he
repeatedly bit his left hand. Thus, a modified form of the original
impulsive habit had developed, and, when the fact was called to his
attention, he gave unmistakable evidence of surprise and chagrin.
(1928, p. 7)
I cite this early instance of an attempt at behavior modification to
point up some recurring themes in the psychology of personality.
(1) There has been a tendency to focus on distinctive or unusual behav-
ior, or on unique personality characteristics. (2) The behavior of inter-
est is often performed unconsciously. (3) There is a presumed
26 ROSS STAGNER
underlying phenomenon-a trait perhaps, or some specifiable personal-

ity dynamics-that can trigger alternative behaviors. One task for the
personality theorist is to identify this underlying process.
The foregoing illustration suggests that the psychology of person-
ality has its roots in clinical psychology and of course there is some
truth in this statement. However, the outstanding aspect of the publica-
tions of 1937 and 1938, cited as the emergent manifestations of the new
psychological specialty, is that the clinical approach is a minor feature,
at least as regards the contribution of Allport and Stagner; and even of
Murray's work it can be said that it is closer to "pure" scientific psy-
chology than to clinical practicalities.
Another differentiating feature of the new approaches to person-
ality was the concern for individual differences. Allport had already
collaborated with his brother Floyd on a scale for measuring the traits
of ascendance and submission; and with Philip Vernon on the Study of
Values, each devised to obtain quantitative indices of hypothesized per-
sonal attributes. Murray and his students also attempted to derive
quantitative indices of various behavioral tendencies. My own experi-
ence included extensive work with personality inventories, the develop-
ment of a Thurstone-type scale for attitude to parents, and the first
"fascist attitude" scale for American populations. Thus McKeen
Cattell's (1886) concern with individual differences was deeply embed-
ded in the formation of the new specialization within psychology.
SOME PERSONAL BACKGROUNDS
It is inevitable that one's position regarding the psychology of per-

sonality will be affected by one's personality. This need not lead to
esoteric psychoanalytic speculations; in many instances the relevant in-
formation is empirical and indeed obvious.
Let me illustrate this generalization with the case of Gordon All-
port. In his autobiography (Allport, 1967), he conceded that he had
experienced some sibling rivalry with his older brother, Floyd. Floyd
received his Ph.D. from Harvard in 1919, the same year in which Gor-
don completed his B.A. Although Gordon did not try to conceal their
rivalry, it was most clearly expressed in an anecdote by Floyd in his
autobiography (Allport, 1974). His description of a family incident is
quite revealing:
Not long ago my brother Gordon was visiting us in California. At
the breakfast table my wife took the occasion to recount to him
what she considered to be certain of my "fine qualities." After her
lengthy eulogistic recital my brother looked up and without a

moment's hesitation added: "And is he still stubborn, lazy, and pro-
crastinating?" Aside from the fact that they were delivered by a
master of the science of personality traits, what startled and dis-
mayed me most about these words was the glibness with which he
uttered them, not needing to pause for a moment's thought or rec-
ollection. (1974, p. 5)
Floyd did not raise the possible issue of sibling rivalry, but this would
seem to be the appropriate rubric under which to file this incident.
The father of Floyd and Gordon was a physician who was finan-
cially able to send both of his sons to Harvard. Henry Murray also
came from a well-to-do family and got his education at preparatory
schools, followed by a B.A. at Harvard and a medical degree at Colum-
bia. Like Sigmund Freud, he spent several years in what would be
considered "pure" research (on embryonic development in chickens);
this ultimately led to his receipt of a Ph.D. degree in biochemistry from
Cambridge University in England.
Murray's transition from biochemistry to psychology almost had
the quality of a conversion. In 1923, he read C. G. Jung's Psychological
Types, which fascinated him; and in 1925 he spent three weeks with
Jung in what seems to have been a therapeutic relationship. He aban-
doned his work in biochemistry and decided on a career that would
combine medical and academic psychology. His earliest studies in per-
sonality were experimental tests of Jungian concepts.
Murray's career at Harvard throws light on some of the more ob-
scure features of academia, particularly in the early part of this cen-
tury. The psychiatrist Morton Prince offered Harvard a gift of $80,000
(an impressive sum in 1927) to establish a psychological clinic. Over the
bitter opposition of the psychology staff, the gift was accepted and the
clinic established. Prince was appointed director of the clinic, with Mur-
ray as his assistant; when Prince died in 1928, Murray was promoted to
the position of director, but with only a three-year term as an assistant
professor. In 1931, his reappointment was opposed by Harvard's exper-
imental psychologists. At this point, Gordon Allport rallied support
from other departments and saved Murray from being dropped from
the faculty.
It is interesting to note that while Murray freely acknowledged his
indebtedness to Allport for this, the two nevertheless continued a low-
temperature feud over the relevance of psychoanalytic theory to re-
search on personality. Of the three of us (Allport, Murray, and myself),
Murray looked most favorably upon the potential usefulness of psycho-
analytic concepts.
28 ROSS STAGNER
As I was much younger than either Allport or Murray, there is

less to be said about my training and career up to 1937. Unlike them, I
was born into a poor Texas family with few books or other stimuli
toward intellectual development. Neither of my parents attended high
school. My early behavior must have puzzled them. My mother used to
tell with amusement of her surprise when, after my first day in the
first grade, I could read the first two pages in my primer. She was even
more astonished when she found that I could read the text just as well
upside down. I had simply memorized the words and recited them in
perfect sequence without using any visual cues. (This quasi-eidetic
memory served me well in graduate school, where it was a real advan-
tage to be able to recall many details of procedure and results of the
experiments being discussed.)
My father died in 1923, when I was 14. My mother, no doubt puz-
zled as to what to do with this "duckling," took me to a newly estab-
lished child guidance clinic in Dallas. A social worker there, Mrs. F. T.
Buss, was an alumna of Washington University in St. Louis, and she
contrived to obtain a tuition scholarship for me at that school. I worked
for my expenses-as a printer in downtown St. Louis and later on
campus. My first major was English literature, and I looked forward to
a career of teaching and writing. However, by what was probably good
fortune, Marion Bunch wanted to offer a course in the psychology of
learning and needed one more enrollee to have the course offered. He
saw me in the corridor and urged me to elect it (he had been impressed
by my performance in his introductory class). The learning course in-
volved laboratory research with human subjects. This fascinated me,
and I promptly changed my major from English to psychology.
Bunch not only gave me a good introduction to research methods;
he also arranged for me to enter graduate training at Wisconsin and to
work there as an assistant to Clark Hull. Regrettably (from my point of
view), Hull moved to Yale before I moved to Madison. Hull's aura,
however, continued to dominate the graduate program. Of those who
influenced me, Norman Cameron is the first to come to mind; a brilliant
lecturer, he taught the course on theories of psychology. W. H. Sheldon
was another dynamic teacher, deeply committed to behaviorism but a
bit more systematic than John B. Watson. Richard Husband and Harry
Harlow came to Madison directly from Stanford and espoused gener-
ally a moderate, functionalist-behaviorist view. Kimball Young, another
significant influence, adopted what might be called a functionalist posi-
tion when he dealt with purely psychological issues as opposed to soci-
ological matters. Hulsey Cason, who joined the Wisconsin faculty just
before I finished, could be characterized as a functionalist in the Thorn-
dike-Woodworth tradition. I have often wondered how Abe Maslow

(who studied with these same teachers) developed his strong humanis-
tic streak. Perhaps most graduate students are like B. F. Skinner, who
once wrote of his graduate work at Harvard that "in graduate school I
had the advantage of scarcely being taught at all" (1983, p. 25). At any
rate, I cannot blame my graduate professors for any more than the
strong behavioristic bias in the first edition of my book. Later editions
made more use of Gestalt and phenomenological frameworks.
My resistance to indoctrination by the faculty was shown in my
selection of a master's thesis topic. Despite the behavioristic atmos-
phere, I proposed a Gestalt-based study on the role of absolute versus
relative differences in a visual discrimination task, designed to test
Kenneth Spence's theory regarding relational learning. This was pro-
gressing well until an epidemic wiped out my cat colony.
Feeling financial pressure, I wanted to get my M.A. and earn some
money right away. I hastily devised a project using human subjects and
a theoretical issue derived from psychoanalysis. Essentially, what I
demonstrated was that high scorers on the Woodworth Personal Data
Sheet ("neurotic" personalities) remembered unpleasant words in a lab-
oratory task better than they recalled pleasant words while the "non-
neurotic" subjects remembered more of the pleasant words.
I taught one semester at Gustavus Adolphus College. While there
I followed up my master's project with one on the forgetting of details
of pleasant and unpleasant experiences from real life. Back in Madison,
I next chose to study whether personality test scores could enhance
the prediction of college grades from aptitude test scores.
Nominally, my doctoral advisor was V. A. C. Henmon, who was
well known for his work in intelligence testing in World War I. Regret-
tably, he was chronically ill during my years at Wisconsin, and while he
approved my plans for a dissertation, he was not in a position to pro-
vide regular guidance as it progressed. It focused on an interaction
hypothesis; that a student's use of his or her aptitude was moderated
by personality traits. This was one of the earliest moderator variable
studies (although unfortunately I did not label it as such). My popula-
tion was the entire freshman class enrolling at the University of Wis-
consin in September 1931. They took the Bernreuter Personality
Inventory during registration. In June, their aptitude scores were cor-
related with their grade-point averages for the first year of college
work. The results were most gratifying: Subgroups scoring high on a
given trait-measure differed from corresponding subgroups of low
scorers on the same trait-measure in terms of the ability-achievement
correlation. For example, in a group of high-dominance males, the ability-
30 ROSS STAGNER
achievement correlation was .71, whereas among low-dominance males

this correlation was only .44 (Stagner, 1933).
When I received my doctorate in 1932, the nation was in the grip
of a severe economic recession. Luckily I received a Social Science
Research Council (SSRC) fellowship grant to continue my work on per-
sonality. With this support, I investigated some of the relationships
between family characteristics and children's personalities; the findings
showed clearly that poverty was detrimental to desirable personality
development (Stagner, 1935).
The rise of fascism was another prominent feature of the 1930s. It
seemed to me that psychology should be capable of providing some
insights into that phenomenon. I culled from the German and Italian
reports some policy statements that could be recast in the American
context and presented to subjects in the form of propositions or
attitude statements to be endorsed or not. The source of the statement
was not given and the word "fascism" was never used. The resulting
scale was thus able to tap the susceptibility of Americans to fascist
attitudes. Articles on the use of this scale (Stagner, 1936a,b) started a
series of studies that culminated in the "authoritarian personality''
project at Berkeley (Adorno et al., 1950).
Another aspect of the personality-politics relationship that ap-
pealed to me was the extent to which young people from fairly secure,
well-to-do homes espoused radical doctrines. In collaboration with my
friend Maurice Krout, I collected some autobiographical data from hun-
dreds of Chicago youths and separated them into conservative, Stalin-
ist, and Trotskyist groups. In general, our findings supported the thesis
that rebellious adolescents were often expressing their hostility against
their parents in their political activities.
In 1935, I got my first full-time teaching appointment, at the Uni-
versity of Akron. There I was asked to offer a course in the psychology
of personality. Seeking a possible text, I found that none was available,
and decided to write my own. Allport was engaged in a parallel project
at the time, but I knew nothing of this. I was, of course, familiar with
his work on the ascendance-submission scale and on expressive move-
ments (Allport & Vernon, 1933). As he often remarked, his course at
Harvard was the first on personality in the United States; I am sure
this is correct, but my own offering developed quite independently.
My book was finished in 1936, and I began talking to publishers'
representatives about it. I was lucky to mention it to the McGraw-Hill
representative, Ed Booher, and he negotiated a contract for publica-
tion. An interesting consequence of this occurred in 1974, on the occa-
sion of the fourth edition. Booher had by now attained the presidency
of McGraw-Hill and was about to retire. As a memento of his first

acquisition for the company, he had two copies of the fourth edition
bound handsomely in leather, one for himself, the other a gift for me.
As I look back at 1937-1938, I am pleased to note that the three
books under discussion (Allport's, Murray's, and mine) represented the
three major currents in systematic psychology of that period. Allport
wrote in a broadly humanistic tradition with attention to phenome-
nology; Murray emphasized psychoanalytic hypotheses; and I built an
interpretation of major phenomena involved in "personality'' as behav-
ioristic. I believe it is fair to say that these are still the major perspec-
tives characterizing European and American theorizing about
personality. Parenthetically, I regret the ethnocentrically American ap-
proach of most recent writings in this field. But this is no place to
introduce a detailed discussion of European viewpoints.
Allport, of course, fought against this "isolationist" trend in Amer-
ican psychology. He praised Wilhelm Stern's "Verstehen" psychology,
and included in his text brief discussions of traditional philosophical
issues, many of which were, in my opinion, beside the point and/or
potentially confusing to the student. Gordon's approach was one of em-
phasizing what he considered to be the important issues and deleting
those that did not seem worth special attention. This extended to the
American tradition of reviewing all kinds of historical material that
might be relevant to a theoretical or empirical issue. Once I casually
mentioned to him that he had completely ignored a certain empirical
study; cheerfully, he asserted of his book that "There isn't a fact in it."
This was not true, but it does illustrate Gordon's preference for a co-
herent treatment of an issue as against piling up mountains of empirical
studies that defy integration.
I often thought of this dictum and decided that what Gordon pre-
ferred was fitting "facts" into a theoretical framework, not simply cat-
aloging the literature. Of course, this approach was used in my book,
too, and in Murray's (1938) volume. No one can review all the directly
or tangentially relevant literature on a significant topic. Murray, for
example, always considered-and usually found-a psychoanalytic in-
terpretation for his empirical data. Similarly, in my 1937 book, I always
looked for, and usually offered, a Hullian or Watsonian model for my
findings. A point that does intrigue me in this connection is that the
three dominant currents in psychology-phenomenology, behaviorism,
and psychoanalysis-were so visible in the foundation of a "psychology
of personality."
Of the three textbooks that we have identified as proof of the
viability of this new specialty within psychology, Allport's was the first
32 ROSS STAGNER
to appear. It was, deservedly, a tremendous success, and it no doubt

contributed importantly to his election as president of the American
Psychological Association in 1939. I got no such ''immediate" reinforce-
ment, but after the second edition of my book appeared in 1948, I was
elected president of Division 8 (Personality and Social Psychology) of
APA. Murray's academic honors, though delayed even longer, included
APA's Distinguished Scientific Contribution Award in 1961 and awards
from the Society for Projective Techniques, the American Psychoan-
alytic Association, and the American Psychosomatic Society. He was
chosen president of Division 8 in 1962.
In my personal contacts with Allport, he impressed me with his
easygoing, confident manner. I think he must have resembled William
James, whom I never met. I recall with pleasure a conversation I had
with Allport shortly after the 1948 revision of my book appeared.
Gordon congratulated me, and then suggested-no doubt with tongue
firmly in cheek-that I should now revise his 1937 volume. "But Gor-
don," I protested, "I've already revised your book-right into mine!"
There was a lot of truth in my statement, and he accepted it as a
compliment rather than an admission of piracy.
My theoretical orientation evolved over time. Starting in 1937 as a
Hullian behaviorist, by 1948 I had shifted to a cognitive emphasis influ-
enced by Allport and Lewin. At this time I was also collaborating with
T. F. Karwoski on some work on homeostasis and applied this approach
to personality (Stagner, 1951, 1977).
In 1939, I moved to Dartmouth, where I taught introductory psy-
chology and some specialized courses in world politics and in industrial
conflict. I was pleased to fancy that I was a replacement for Allport,
who had taught at Dartmouth from 1926 to 1930. I was also astonished
to learn (from faculty gossip) why he had returned to Harvard in 1930.
It seems that in 1929, the Dartmouth administration informed the psy-
chology staff that one faculty member could be awarded tenure; the
department voted to grant tenure to Edwin M. Bailor rather than to
Allport. Not surprisingly, Gordon wrote some letters and got an invita-
tion to rejoin Harvard, where he remained for the rest of his life.
This experience at Dartmouth may well have been an important
factor in Allport's considerable efforts, soon after he joined the Har-
vard faculty, to improve Murray's prospects for receiving tenure. Many
faculty members, including such influential figures as Edward G. Bor-
ing and Karl S. Lashley, were opposed to tenure for Murray. Gordon
rather uncharacteristically mounted a vigorous campaign on Murray's
behalf, collecting endorsements of Murray from distinguished psycholo-
gists around the country. The campaign succeeded, but only to a limited
extent: Murray's appointment was renewed, but he was not granted

permanent tenure. In fact, he did not achieve tenure until 1947, and
was not promoted to the rank of professor until1951.
Psychologists who believe in the validity of human judgments of
the abilities of others will perhaps be baffled by the treatment of All-
port at Dartmouth and of Murray at Harvard. Certainly, if we are
going to use psychological assessments of persons to predict their fu-
ture performance, we ought to be able to achieve higher validity than
these two cases indicate. And the judges in both instances were profes-
sional psychologists!
I do not know enough about the Dartmouth matter to offer an
explanation. However, in the Harvard situation, strong professional
biases were operative. Boring and Lashley led a faction seeking to
assert the "scientific" status of "pure" psychology. Because Murray
often espoused a psychoanalytic approach, they rejected his claim to
belong in a scientific department that aspired to parity with, or at least
acceptance by, such disciplines as physics and chemistry. Many of
today's psychologists are unaware of the schism within the American
Psychological Association in 1935, when clinical and industrial psychol-
ogists broke away from the "scientific" establishment and set up the
American Association of Applied Psychology. This was followed, in
1936, by the organization of the Society for the Psychological Study of
Social Issues when the officers of APA refused to take stands on fas-
cism and other burning public issues. (We now witness a reversal of the
1935 fission; the "applied" psychologists seem to have taken over APA
and "scientific" psychologists transfer into the American Psychological
Society.)
From the perspective of 1993, it seems strange that Boring and
Lashley, two devotees of objectivity in psychology, fought so vigorously
against Murray's promotion. He was, after all, a physician and a bio-
chemist, eminently qualified by the criteria of the "established" sci-
ences. The explanation lies in academic politics. Boring and Lashley
wanted psychology to be accepted as a science; they feared loss of
status if they were identified with the ambiguities of psychoanalysis.
PERSONALITY AND PSYCHOANALYSIS
I have referred to the hostility expressed by my instructors at

Wisconsin to psychoanalytic theories. Much of this was, as I see now,
due to a confusion of some highly imaginative writings with the solid
clinical observations that underlie most of Freud's generalizations. As I
34 ROSS STAGNER
read Murray, he was critical of psychoanalytic fantasies but not of the

better-founded components of the doctrine.
Not having been psychoanalyzed, I refrain from offering any psycho-
diagnosis of my inner dynamics. As I have noted, in terms of instruc-
tional influence, I should have been biased against Freudian theory. At
least two of the professors I admired (Cameron and Sheldon) were
vigorous critics of psychoanalysis. Yet I remained friendly, or at least
tolerant, with regard to this approach to personality. Maybe the rela-
tionship between student and teacher is as often dialectic as it is imita-
tive. Perhaps B. F. Skinner spoke for many graduate students when he
wrote of his Harvard Ph.D., "I had the advantage of scarcely being
taught at all" (1983, p. 25). Perhaps our future theoretical predilections
are laid down in early childhood, as Freud speculated. At any rate, I
tried to integrate Freud into my behavioristic frame of reference, and I
believe that I succeeded, at least in later years. One relevant bit of
evidence is a study I did with another graduate student, Neal Drought
(Stagner & Drought, 1935), utilizing a Thurstone scaling technique to
measure students' attitudes toward their parents. This study was men-
tioned in the Time magazine roundup of items dealing with psychology
from 1923 to 1988 (Gerow, 1988, pp. 21-22).
VIEWS ON METHODOLOGY
It is not likely that Gordon Allport's most enthusiastic admirers

would describe him as an expert on methodology. His approach would
have to be characterized as thoughtful and occasionally inspired, but
not emphasizing precision or elaborate experimental controls. On the
other hand, he liked to push at the boundaries of accepted methodolo-
gies-as can be seen in H. S. Odbert's dissertation, Trait-names: A
Psycholexical Study (Allport & Odbert, 1936). To Gordon, language and
culture were instrumental in shaping individuals' personalities.
Allport also advocated the use of autobiographies and other per-
sonal documents as ways of studying personality. A maJor problem here,
as he probably would have been the first to admit, is the tendency to
distort reporting in such a way as to improve the self-image. B. F.
Skinner's writings (1976, 1979) provide a case in point. In the first
volume of his autobiography, Skinner described himself, during gradu-
ate school at least, as a creature of strict routine with tight time sched-
ules governing his daily activities:
I would rise at six, study until breakfast, go to classes, laboratories,
and libraries with no more than fifteen minutes unscheduled during
the day, study until exactly nine o'clock at night and go to bed. I
saw no movies or plays, seldom went to concerts, had scarcely any
dates and read nothing but psychology and physiology.
(1976, p. 398).
This passage delighted me because it seemed to confirm Allport's
theory about the unity of personality, tying in so well with Skinner's
concern with uniformities of behavior, schedules of reinforcement, and
stimulus control of behavior. My enthusiasm was dashed when, in a
second volume of autobiography that appeared three years later, Skin-
ner dismissed the earlier description of his graduate school years as
pure fiction: "I was recalling a pose rather than the life I actually led"
(1979, p. 5).
Allport might have defended his faith in autobiographies by noting
that scholars can search for just such contradictions or revisions of the
writer's past. To me, the excerpts from Skinner's accounts suggest that
we ought to defer to the historians and political scientists, who have for
centuries warned us about the weaknesses of autobiographies as
sources of reliable data about past events.
As a matter of fact (fact as I see it), Allport was not seriously
interested in methodological questions. He sympathized with Wilhelm
Stern's psychology of understanding (Verstehen) and subordinated
"objectivity" to a highly subjective empathy. Even so, Allport deserves
credit for an important methodological formulation: the idiographic
hypothesis. He proposed that instead of studying frequencies of re-
sponses across a universe of persons (the traditional research prac-
tice), we could study patterns within the universe of actions by a single
individual.
Murray's interest in methodology was much stronger than
Allport's. At least two of Murray's innovations met with wide accep-
tance: the Thematic Apperception Test (TAT) (now a basic clinical tool)
and team assessment. During World War II, he headed a group of
psychologists (OSS Assessment Staff, 1948) that administered perfor-
mance tasks under stressful conditions. The OSS staff were seeking to
predict behavior in actual high-stress circumstances, such as those en-
countered by parachutists on a sabotage mission when dropped behind
enemy lines. Variations on OSS procedures for assessing complex per-
sonal attributes became quite popular in industrial psychology during
the 1960s and 1970s; the selection decisions made at these assessment
centers tended to be valid. The relative success of team assessment (in
contrast to the errors noted above in the Allport and Murray tenure
decisions) may have been partly due to the fact that observers were
trained specifically to avoid biases in judgment.
36 ROSS STAGNER
As for myself, I cannot claim any innovative contributions to method-

ology. The "fascist attitude" studies did set in motion a train of re-
searches on the "authoritarian personality" and later on dogmatism and
rigidity. My interests in these topics, and the organization of the Soci-
ety for the Psychological Study of Social Issues in 1936, in which I was
actively involved, led me to deal with linkages between personality and
political behavior in my 1937 book. Allport and Murray were both
drawn into applied aspects of psychology during the war, Allport mak-
ing contributions to rumor analysis and Murray, of course, in combat-
related selection procedures. I took a defense-related job in industrial
personnel work, and this played a large part in my subsequent focus on
industrial psychology. However, I found plenty of personality problems
among industrial personnel, and often obtained insights from my work
in industry that were useful in the study of personality. So I feel that I
have had not two separate careers in psychology, but one-it merely
moved from academic to industrial settings.
I would like to add one note with respect to methods in the study
of personality. In a recent review, Craik (1986) analyzed this field thor-
oughly, and what struck me about his paper was that, of the seven
classes of procedures he identified, every one is mentioned in the three
books that we have taken as our starting point in this review of the
psychology of personality.
RETROSPECTIONS
As I look back at 50 years of psychological research on and specu-

lation about personality, I find myself uncertain as to how much prog-
ress we have made. We have certainly increased the reliability of our
methods, but a parallel increase in validity is hard to find. Murray's
introduction of the team assessment format has been widely imitated
and seems to have been of some utility in industrial work. On the other
hand, I have noted the low validity of judgments made by professional
psychologists (with Allport at Dartmouth and Murray at Harvard),
which casts doubt on the validity of the team approach to assess-
ment. Similarly, Allport's reliance on autobiographical data is under-
mined by the frequency with which individuals publish reports about
themselves that are full of distortions and inaccuracies. Personality in-
ventories are susceptible to faking and even indirect measures such as
the Rorschach inkblots and the TAT pictures can be faked without
much trouble.
Although these last 50 years have certainly been eventful, we have
not yet arrived at a consensus on just what we mean by personality nor

on the optimal procedures for studying it. Nevertheless, the search has
been fun.
ACKNOWLEDGMENTS: I wish to acknowledge the valuable assistance of

Dr. Sheldon J. Lachman in verifying some items of information cited
here. My thanks also go to Dr. M. Brewster Smith for valuable com-
ments on an earlier draft of this chapter.
REFERENCES
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The
authoritarian personality. New York: Harper & Row.
Allport, F. H. (1974). Autobiography. In G. Lindzey (Ed.), A history of psychology in
autobiography (Vol. VI, pp. 3-29). Englewood Cliffs, NJ: Prentice-Hall.
Allport, G. W. (1967). Gordon W. Allport. In E. G. Boring (Ed.), A history of psychology
in autobiography (Vol. V, pp. 3-25). New York: Appleton-Century-Crofts.
Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycholexical study. Psychological
Monographs, 47 (1, whole No. 211).
Allport, G. W., & Vernon, P. E. (1933). Studies in expressive movement. New York:
Macmillan.
Bagby, E. (1928). Psychology of personality. New York: Holt.
Cattell, J. M. (1886). Psychometrische Untersuchungen. Philosophische Studien, 3, 30&-
355.
Craik, K. H. (1986). Personality research methods in historical perspective. Journal of
Personality, 54, 18--51.
Gerow, J. R. (Ed.) (1988). Psychology 1923-1988 (Special Issue of Time). New York:
Time.
James, W. (1890). Principles of psychology (Vol. 1). New York: Holt.
Murray, H. A. (1938). Explorations in personality: A clinical and experimental study of
fifty men of college age. New York: Oxford University Press.
OSS Assessment Staff (1948). Assessment of men. New York: Rinehart.
Roback, A. A. (1927). Personality, character, and temperament: A bibliography. Cam-
bridge, MA: Sci-Art.
Skinner, B. F. (1976). Particulars of my life. New York: Knopf.
Skinner, B. F. (1979). The shaping of a behaviorist. New York: Knopf.
Skinner, B. F. (1983). Interview. Psychology Today, 17, 25ff
Stagner, R. (1933). Relation of personality to academic aptitude and achievement. Jour-
nal of Educational Research, 26, 648-660.
Stagner, R. (1935). Economic status and personality. School and society, 42, 551-552.
Stagner, R. (1936a). Fascist attitudes: An exploratory study. Journal of Social Psychol-
ogy, 7, 309-319.
Stagner, R. (1936b). Fascist attitudes: Their determining conditions. Journal of Social Psy-
chology, 7, 438-454.
38 ROSS STAGNER
Stagner, R. (1951). Homeostasis as a unifying concept in personality theory. Psychologi-

cal Review, 58, 5-17.
Stagner, R. (1977). Homeostasis, discrepancy, dissonance: A theory of motives and moti-
vation. Motivation and Emotion, 1, 103-138.
Stagner, R., & Drought, N. (1935). Measuring children's attitudes toward their parents.
Journal of Educational Psychology, 26, 169-176.
CHAPTER THREE
Allport's Personality and

Allport's Personality
ALAN C. ELMS
INTRODUCTION
In Gordon Allport's (1967) brief autobiography, written a year before

he died, he asked the question repeatedly: "How shall a psychological
life history be written?" It was a question he had asked recently about
another life, in preparing for book publication the Letters from Jenny
(1965). It was a question he had asked more broadly a quarter-century
earlier, in writing his monograph on The Use of Personal Documents in
Psychological Science (1942). It was a question he had asked quite
early in his career, in a paper on a confessional work titled The Loco-
motive God (1929). Indeed, Allport said that even his doctoral disserta-
tion, An Experimental Study of the Traits of Personality (1922a), "was
an early formulation of the riddle: How shall a psychological life history
be written?" (1967, p. 7).
I did not begin as early and I have not persisted for nearly as long
as Allport, but for over 15 years I have been asking the same question,
"How shall a psychological life history be written?" Among a variety of
psychobiographical subjects, I have asked the question most often about
personality theorists, ranging from Freud and Jung to Skinner and
Murray. My first published attempt to answer the question was a five-
ALAN C. ELMS • Department of Psychology, University of California, Davis, California

95616.
New York, 1993.
39
40 ALAN C. ELMS
page paper on Gordon Allport, which appeared in 1972. I felt rather

more nervous than usual about the publication of that paper. It sug-
gested an interpretation of Allport's personality and a set of connec-
tions between his life and his work for which I had little hard evidence.
I had never met Allport; I had not discussed him in detail with anyone
who had known him well; and I was advancing a psychoanalytically
tinged account of the core theoretical concepts of a man who had been
a career-long critic of psychoanalysis.
Fortunately for my own budding career as a psychobiographer,
reactions to my paper on Allport-or at any rate the reactions I heard
about-were generally favorable. Several friends and colleagues of his
told me I had done a pretty good job, at least in so brief a space and for
someone who didn't know him. Most offered additional information to
support my hypotheses. I felt encouraged enough by these reactions to
move on to psychobiographical studies of other theorists. I also began
to think about writing a book that would combine my studies of a
number of theorists, including Allport. So I have continued to read his
work and to collect information about him, rather sporadically, through
interviews with additional colleagues of his and through occasional
examination of archival materials. The larger part of my attention in
recent years has been directed toward more provocative subjects, espe-
cially Sigmund Freud and Henry Murray, among psychologists, plus
such bizarre personalities as Elvis Presley and Alexander Haig in other
areas of endeavor. However, I keep returning, now and then, to the
quiet but firm person of Gordon Willard Allport.
ALLPORT'S ENCOUNTER WITH FREUD
My first paper on Allport focused on his one and only meeting with
Sigmund Freud. It was a story that Allport told often-so often that
the several· surviving versions of it in his own words are almost identi-
cal, sentence for sentence. It was the single personal anecdote included
in his New York Times obituary (October 10, 1967). The version below
has not been previously published, but it will sound familiar to the
many people who heard the story from Allport himself:
So, in my callow youth-this story is very much at my own ex-
pense-in my callow youth, I wrote Freud a note, announcing that
I was in Vienna and no doubt he would be glad to see me. I re-
ceived a reply in his own handwriting inviting me to come to his
office, at a certain time and place. I went there, to the famous red
burlap room with pictures of dreams on the wall. He opened the
ALLPORT'S PERSONALITY AND ALLPORT'S PERSONALITY 41
door and invited me into the inner sanctum, and sat there silent.
You may not believe this possible, but I was not prepared for si-
lence. It occurred to me it was up to me to say something, so I
fished around in my mind and I told him about an episode on the
tram car coming out. I had seen a little boy about four years old
and the little boy obviously was developing a real dirt phobia. His
mother was with him. She was a starched Hausfrau, terribly clean
and purposive looking. And the little boy was saying: "I don't want
to sit there! Don't let him sit near me, he's dirty." He kept this up
all the time, and I thought it might interest Freud how early a
phobia of dirt can get set. He listened, and fixed his therapeutic
eye upon me and said "Was that little boy you?" Honestly, it
wasn't, but I felt guilty. (Allport, 1962, p. 2)
Allport's adjectives describing the boy's mother vary slightly from
version to version: "excessively clean and well starched ... domineering
in manner" (Allport, 1958, p. 2); "a well-starched Hausfrau, so domi-
nant and purposive looking that I thought the cause and effect appar-
ent" (Allport, 1967, p. 8); "well starched and very prim" (quoted by
Evans, 1970, p. 4). The little boy is described in 1958 as looking "exces-
sively clean and well starched" like his mother; in the 1967 account,
Allport says that to the little boy "everything was schmutzig" (p. 8);
otherwise the little boy's description remains the same in each telling.
The story's punchline ("Was that little boy you?" or "And was that
little boy you?") is always followed by Allport's denial: "It wasn't. (But
I felt almost guilty!)" (Allport, 1958, p. 2); "It was not, honestly, but I
felt guilty" (1964 filmed interview, slightly reworded in Evans, 1970,
p. 4); "Flabbergasted and feeling a bit guilty, I contrived to change the
subject. While Freud's misunderstanding of my motivation was amus-
ing.... " (Allport, 1967, p. 8).
As Allport indicates here, he told his story both as a kind of joke
on Freud and as a joke on the young Gordon Allport. However, he also
made clear that the incident was ultimately no joke for him. He re-
ferred to it variously as "profoundly revealing" about Freud in ways he
had thought back on "many times since becoming a professional psy-
chologist myself' (1958, p. 2), as "a very important experience to me"
(1962, p. 2), and as "an event of pungent significance ... [having] the
character of a traumatic developmental episode" (1967, p. 7). Allport
described his reaction following the meeting in this way:
But thinking that little episode over, it grew in importance to me,
because the fact is that Freud, with his tendency to see pathologi-
cal trains, and most of his people who came to see him were pa-
tients, of course, it was natural he should break through my
42 ALAN C. ELMS
defenses and get down to business, you see. But actually, he mis-
took my motives. If he had said, ''Well, here is a brassy young
American youth, a tourist who is imposing on my good nature and
time," he would have been somewhere near correct, I think. But to
ascribe my motivation to the unconscious in this case was definitely
wrong. (1964 filmed interview, somewhat reworded in Evans, 1970,
pp. 4-5)
In his autobiography, Allport summarized the Freud incident's longer-
term professional effects on him:
This experience taught me that depth psychology, for all its merits,
may plunge too deep, and that psychologists would do well to give
full recognition to manifest motives before probing the unconscious.
Although I never regarded myself as anti-Freudian, I have been
critical of psychoanalytic excesses. A later paper entitled ''The 'lhmd
in Motivational Theory" (1953) is a direct reflection of this episode
and has been reprinted, I believe, more frequently than any other
of my articles. (1967, p. 8)
ALLPORT'S THEORY OF THE CLEAN PERSONALITY
Allport's vigorous and repeated denial that he was the little boy
with the dirt phobia intrigued me as it has intrigued others (e.g., Faber,
1970; Morey, 1987). I do not consistently practice what Erikson (1969,
p. 98) calls "originology," the strategy of attempting to explain adult
psychological patterns always by reference to their earliest and most
primitive form. But it seemed to me that if Allport insisted so stren-
uously on denying his resemblance to the clean little boy on the tram-
car, we should at least find out what sort of little boy Allport himself
had been.
In looking for such information, I quickly struck pay dirt. In his
autobiography, Allport described his childhood in only one cautiously
phrased paragraph. That paragraph, however, appeared to explain a
great deal about his reaction to the Freud encounter. Allport's para-
graph is worth repeating in full:
Our home life was marked by plain Protestant piety and hard
work. My mother had been a school teacher and brought to her
sons an eager sense of philosophical questing and the importance of
searching for ultimate religious answers. Since my father [a physi-
cian] lacked adequate hospital facilities for his patients, our house-
hold for several years included both patients and nurses. Tending
office, washing bottles, and dealing with patients were important
aspects of my early training. Along with his general practice my
father engaged in many enterprises: founding a cooperative drug

company, building and renting apartments, and finally developing a
new specialty of building and supervising hospitals. I mention his
versatility simply to underscore the fact that his four sons were
trained in the practical urgencies of life as well as in a broad hu-
manitarian outlook. Dad was no believer in vacations. He followed
rather his own rule of life, which he expressed as follows: "If every
person worked as hard as he could and took only the minimum
financial return required by his family's needs, then there would be
just enough wealth to go around." Thus it was hard work tempered
by trust and affection that marked the home environment. (Allport,
1967, pp. 4-5)
Largely on the basis of that passage, I drew several connections
between Allport's early history and his later career. They key para-
graphs in my 1972 paper were these:
We may now wonder whether Freud's question was received
so traumatically ... because unconsciously he [Allport] was still
carrying within him the super-clean little boy whose schoolteacher
mother stressed to her children the importance of "plain Protestant
piety," philosophical questions, and "ultimate religious answers";
whose physician father had no time for play but transmitted to the
children his concern with hard work and tight money; and who
himself spent much of his childhood living in the abnormally clean
environment of a home-hospital, washing bottles and coping with
patients who must either be kept from infection ... or be avoided
as infectious....
Not only might such an early history have made the adult
Allport peculiarly sensitive to small boys with dirt phobias; it may
also have been translated into Allport's expressive movements, into
a ''prim" and "well starched" pattern of gesture, dress, and ·speech ...
that would have been Freud's only other clue besides the story
itself to Allport's presence. Allport later in his life proposed ... that
the "style of execution" of a person's behavior "is always guided
directly and without interference by deep and lasting personal dis-
positions" [1937, p. 466]. Freud is reported to have been particu-
larly adept at reading such nonverbal cues....
If our inferences are reasonably accurate, then the apparent
force of this brief encounter with Freud becomes more understand-
able. It was [not merely] one misguided comment by Freud ... that
set Allport upon a career of discounting unconscious forces.
Allport's reaction can instead be viewed as compatible with his de-
velopmental history. The high-minded and clean little boy had be-
come a high-minded and clean young man, unwilling to examine any
"dirty" foundations for his choice of conversational gambit, for his
44 ALAN C. ELMS
"flabbergasted" reaction to Freud's question, or for his behavior in

general. This high-minded young man eventually became an im-
portant psychological theorist who promulgated a view perhaps ap-
propriately described as [a theory of1 "The Clean Personality." He
argued against digging into the unconscious except in unusual
cases .... He rarely discussed sexual motivation except to protest
Freud's emphasis upon it . . . . He rejected psychological data on
such unsavory creatures as rats, children, and neurotics as being
largely irrelevant to an understanding of the mature personality.
His key motivational concept was functional autonomy, the propo-
sition that most adult motives become somehow totally indepen-
dent of their baser origins. He saw the ultimate criterion of the
mature personality as a unifying philosophy of life, perhaps ideally
a religious one. Allport appears to have recognized a certain degree
of continuity between his [positive] adult theory-building activities
and his parents' "plain Protestant piety"; but it may have been
precisely [the same] early training in piety, cleanliness, and order
that led him to reject the importance of unconscious and infantile
motives to the normal adult personality. (Elms, 1972, pp. 629-631)
I cautiously noted, at the end of my paper, that "Our knowledge of
Allport's childhood is so slight that we must not push our speculations
further." I also pointed out that "Allport's theoretical formulations are
clearly the immediate product of much intellectual effort of the highest
quality'' (p. 631). Nonetheless, it seemed clear to me that Allport's early
training in cleanliness and virtue had left him with an unusual sensitiv-
ity to certain aspects of personality that other people, including Freud,
tended to overlook; and with a distinct insensitivity to, or avoidance of,
certain aspects of personality and life to which Freud was particularly
sensitive.
I have subsequently obtained other information that supports this
interpretation. For instance, the Allport Papers at Harvard contain a
letter from Henry Murray to Allport's widow, describing various as-
pects of his friendship with Gordon:
This brings me once more to that old unresolved question ...
Freud's reason for asking: ''Was that little boy you?" Gordon (and
you) seemed to think that Freud had gained the (mistaken) im-
pression that Gordon was (unconsciously) seeking therapeutic relief
in visiting the Master. Is that right? But suppose that Freud made
no such mistake (having frequently been visited by similar seer-
hunters): he simply noted a certain psychological similarity between
(i) Gordon's perfect neatness and cleanliness of body and clothes as
well as his perfect gentility of speech and (ii) the little boy's obses-
sional determination to keep free from dirt (a similarity which
would account for Gordon's special attention to the boy's behavior),

and then Freud, not without a touch of malice, asked his question,
as if to say: "Were you not, at one time, as intent as that little boy
to avoid dirt (dirty words)?" (Murray, 1972)
Dan Ogilvie, at one time a student of Allport's, has recalled an
occasion in 1964 on which Allport asked him to retrieve some copies of
Allport's first book from a dusty storage space. Allport handed Ogilvie
a clean rag to wipe the dust from his hands while Allport autographed
a copy of the book for him.
In the meantime I had removed the dust, but he insisted I use the
rag anyway. I did that and seeing that the rag was in no worse
shape than when it had been handed to me, I refolded it and placed
it on the corner of his desk. With an expression of disgust, he
gingerly deposited the rag in a waste basket. Suddenly I was glued
to the floor with the realization that in a critical way the little boy
[of the Freud anecdote] ... did represent an important aspect of
Professor Allport after all. (Ogilvie, 1984, p. 13)
Another of Allport's graduate students has described Allport to me
as "extremely orderly," as a man who often referred to ''buttoning up
things," and who indeed described himself as ''buttoning up his life"
during the final months of his fatal illness (Thomas Pettigrew, personal
communication, 11 August 1977)-an image of precise control reminis-
cent of that well-starched Viennese hausfrau whom Allport was so fond
of characterizing.
In my original paper, I did not speculate on toilet-training history
or anal personality traits as sources of Allport's later responses, be-
cause I had no specific information on such matters and because I did
not wish to be so crudely reductionistic. Both Thomas Pettigrew (per-
sonal communication, 30 September 1988) and Robert Allport, Gordon's
son (personal communication, 26 October 1988), have suggested to me
that describing Allport's personality solely in terms of dirt avoidance
and orderliness would leave a distorted picture of a complex man.
Allport's heavy smoking habit, which probably contributed to his death
from lung cancer, also appears inconsistent with impressions of a com-
pulsively clean and totally self-controlled individual. Nonetheless, the
evidence for his strong cleanliness concerns and for his reluctance to
acknowledge them is substantial. Further evidence may be found in
Allport's only detailed written description of the rest of his encounter
with Freud, previously unpublished:
Mter this opening episode [ending with Freud's question] I
realized that my attitude was too much that of a "tourist" and that
46 ALAN C. ELMS
Dr. Freud had been expecting a professional consultation. I then

brought out a more personal idiosyncrasy: my dislike of cooked
raisins. I told him I thought it due to the fact that at the age of
three, a nurse had told me they were ''bugs." Freud asked, ''When
you recalled this episode, did your dislike vanish?" I said, "No." He
replied, ''Then you are not at the bottom of it." ...
I then spoke of a common sexual problem of youthful males of
my age. [Here Allport may have been referring to anxieties about
masturbation or wet dreams; "self-pollution" was a popular term
for such matters in those days.] His [Freud's] reply: "Yes, nature
tells a many to marry at 18; and society tells him to marry at 28."
Finally I asked, in general, what analyst he would recommend
in America in case I ever wished to undergo analysis. He replied
that he would recommend only one man: Brill. (This was in 1920.)
I then departed with a vivid feeling of respect and liking for
Freud, even though our short conversation had started at cross
purposes. (Allport, 1958, p. 3)
THE EVOLUTION OF A CLASSIC TEXTBOOK
From the latter account, it appears that although Allport flatly

rejected Freud's equation of him with the little boy on the tramcar, his
rejection of Freudian theory in general did not start as quickly or as
emphatically as he implied in other accounts. His continuing ambiva-
lence over the next several years was expressed in his correspondence
with a friend and former graduate school classmate, William S. Taylor.
In July 1922, two years after his Freud encounter, Allport wrote to
Taylor that he would soon be returning to Vienna and Berlin, "and hope
that we may correspond rather seriously and fully on some of the mat-
ters which puzzle us both, and which we might disentangle in writing.
I may be able to give first hand the psycho-analytic views, or I may
have an opportunity of securing Freud and Jung's answers to our ques-
tions" (Allport, 1922b). As matters turned out, Allport did not meet
Freud again and apparently had no significant contact with other Euro-
pean psychoanalysts. Nor were his questions readily disentangled, as
his further letters to Taylor indicate. Five months later, for instance, he
wrote from Berlin:
The course I have with [Max] Dessoir on "Parapsychologie and
Psychoanalyse" is an interesting exposition and destruction of the
Freudian formulae. He cannot add much, I believe, to our own crit-
icisms of Freud. Is it not true that some of the fundamental contri-
butions of Freud are as original and significant as the ideas of any
one man can ever be in science? Is it not true also that no psycho!-
ogist who has balance would hesitate to condemn the silly Freudian
superstructure? Freud in Vienna is now the old story of a famous
man in his own town. He is hooted and hooted. Everyone here
laughs at him, and he has no recognition except in a narrow circle
of orthodox admirers. Freud would be far better off should he mi-
grate to America. (Allport, 1922c)
After another three months, Allport wrote further from Hamburg: "I
have just finished reading Freud's Vorlesungen [Introductory Lectures]-
miles of them-and remain no more nor less instructed than before
taking up the book. He is totally without respect in his own land. Such
is the fate of a prophet or a would-be prophet" (Allport, 1923).
Allport did gain a great deal of new information and new perspec-
tives on psychology in Europe, and he occupied himself mainly with
issues related to these nonpsychoanalytic approaches during the next
decade. But Freud's crucial question to him remained an irritant consciously
and perhaps unconsciously, a challenge to his personal and professional iden-
tity at several levels. It would be finally addressed-though not addressed
finally-during the writing of a pioneering textbook in 1936.
The textbook was Personality: A Psychological Interpretation,
published in 1937. It was not merely one of the frrst textbooks in per-
sonality psychology; as Maddi and Costa (1972) have observed, "The
book had a profound impact on the field, virtually defining for many
years what was and was not to be considered important in the study of
personality" (p. 139). Allport's official version of how the textbook came
to be written deals with rather abstract and intellectual matters:
My ambition was to give a psychological definition of the field of
personality as I saw it. My vision, of course, was influenced by my
encounters with social ethics, Anglo-American empiricism and Ger-
man structural and personalistic theories. I wanted to fashion an
experimental science, so far as appropriate, but chiefly I wanted an
"image of man" that would allow us to test in full whatever demo-
cratic and humane potentialities that he might possess. I did not
think of man as innately "good," but I was convinced that by and
large American psychology gave man less than his due by depicting
him as a bundle of unrelated reaction tendencies. I did not write
the book for any particular audience. I wrote it simply because I
felt I had to define the new field of the psychology of personality as
I saw it. (Allport, 1967, p. 15)
However, in one of his previously unpublished autobiographical
statements, Allport offers a much more personally involving account of
the book's origins. In so doing, he returns to a familiar topic:
That book has a strange feature that I have never mentioned to
48 ALAN C. ELMS
anyone. I wrote it for no audience. Now, when you're taught to

write you're told to have an audience in mind. But, if there was
ever a book written because someone had to write it, I think that
was it. I didn't think it was a textbook. I didn't think of it as an
argumentative treatise. I thought of it as summing up my ideas
about human personality. In it, you can see the impact of my en-
counter with Freud as the origin of my idea that adult motivation
is not necessarily a channelling, or conditioning, or overlay of ca-
thected instincts or infantile motivations or fixations.
The result was the concept of functional autonomy. I had no
idea at the time that it would be picked up and made an issue of,
but it merely seemed to me obvious that motivation was often func-
tionally autonomous of its historical origins in a life. The main-
springs of life get re-wound in the course of development. This was
a direct answer to Freud, for Freud had thought that I was suffer-
ing from an infantile trauma. I wasn't. If he said I was a brassy
young American, he would have been right. But, he didn't. He
didn't perceive my contemporary motivation. His error impressed
me very much, so that I kept working on the problem of motivation
from a point of view of the developing adult. I don't deny that
there may be traces of infantilism in all of us or traces of neurosis
in all of us. I'm not denying or disparaging at all the Freudian
contribution. But, Freud's theory to my mind just is not adequate,
because some people do grow up sometimes, in some respects.
What interested me was what it meant to grow up and be adult
and normal in personality function. That is the focus of my concept
of functional autonomy and in general of all the 1937 book, as well
as much subsequent writing. (1962, p. 5; Allport's emphases)
That is a very interesting statement, for a number of reasons.
First, it reports in greater detail on something we have already heard
about elsewhere: how Freud's question had a powerful effect on All-
port's subsequent psychological thinking. Second, it emphasizes the
magnitude of that effect, by saying that a two-minute exchange with
Freud in 1920 led to the writing of a 600-page theoretical-position-
paper-cum-textbook 16 years later. This is further evidence that
Allport had been well primed to react to Freud's question, which hit
home in ways that Allport never fully acknowledged.
NOT A LITTLE BOY
Allport's series of denials, toward the end of the passage just quoted,
indicates one of the ways in which Freud's question had hit home.
" ... Freud had thought that I was suffering from an infantile trauma. I
wasn't. If he had said I was a brassy young American, he would have

been right. But, he didn't . ... I don't deny that there may be traces of
infantilism in all or us or traces of neurosis in all of us" (my italics). But
that is just what Allport had denied about himself, perhaps silently at
first but promptly and vigorously, then aloud, over and over again: I
am not that little boy with the dirt phobia. Keep that sentence in
mind-Allport's core response to Freud's interpretation of his behav-
ior-as we examine each element of it.
In my initial analysis of Allport's Freud anecdote, I stressed the
denial implicit in the final portion of the core response. "I am not some-
one with a dirt phobia," insisted young Gordon Allport, who had been
trained by his parents to be clean in mind and body, and who would
later develop a theory of the clean personality in contrast to dirty
psychoanalysis. ''Who, me, a neurotic dirt-phobe? Never!" was Allport's
apparently instant reaction to Freud's unsettling probe. When Allport
later felt it necessary to justify presenting what he called "so critical
and so brief' an account of psychoanalysis in his Personality textbook,
he offered a similar denial of its theoretical applicability to nonneurot-
ics: "Psychoanalytic concepts are drawn exclusively from neurotic and
pathological material ... and for this reason their applicability to nor-
mal personality is in many respects questionable" (1937, p. 181, fn. 27).
But there is more to Allport's reaction, as his 1962 statement on
the origins of the Personality textbook suggests. He says there, " ...
Freud's theory to my mind just is not adequate, because some people
do grow up sometimes, in some respects" (p. 5; his emphases). Now
back to our core sentence: "I am not that little boy...."
Regardless of what neurotic symptoms were or were not involved,
it was important for Gordon Allport, at age 22 in Sigmund Freud's
office, to deny being a little boy. Allport had been the baby in his
family, five years younger than the youngest of his three brothers. To a
considerable degree, he had remained in the shadow of his brother
Floyd, seven years older than he. Floyd Allport graduated from Har-
vard before Gordon thought seriously of enrolling there. Floyd was a
decorated military hero in World War I while Gordon merely served in
the Students' Army Training Corps, performing what Gordon himself
called "sophomoric" tasks. Floyd was awarded a Ph.D. degree from
Harvard on the day that Gordon got his bachelor's degree. At several
points in his short autobiography, Gordon refers either to his "general-
ized inferiority feeling'' as a young man (1967, p. 7) or to specific feel-
ings of inferiority toward Floyd (p. 12). In an undated description of his
"Personal Experience with Racial and Religious Attitudes," Gordon de-
scribed himself as having been "a youth of great inferiority feeling ....
50 ALAN C. ELMS
I know what it is to be the object of scorn, but for personal and not
[group] membership reasons. As I said, this sensitivity gave me a vari-
ety of compensatory strivings in myself, and as I overcame my handi-
caps, I also grew in sympathy with any under-dog'' (Allport, n.d.). Early
in his academic career he told one correspondent, "I have published
several articles of no great importance, and am not to be confused with
my more eminent brother, F.H. Allport, of Syracuse University" (All-
port, 1928).
Gordon Allport did manage after college graduation to go off on his
own for a year, to teach in a small English-speaking school in Constan-
tinople, with nobody around who was likely to make him feel inferior.
But at the end of that year he headed back to Harvard, ready to be-
come a graduate student in the psychology department where brother
Floyd was already on the faculty-and at this point Freud had the
audacity to tell him he was still a little boy! No wonder so much of the
1937 Personality textbook and of Gordon Allport's later writing about
personality development stress the discontinuity between child and man,
through the development of the functional autonomy of motives and in
many other ways.
ALLPORT AS UNIQUE INDIVIDUAL
Back to our core sentence again: "I am not that little boy...."
Allport treated as self-evident the fact that he was not the little boy he
had seen on the tramcar. Regardless of any possible similarities that
Freud might have hypothesized, Gordon W. Allport was one person and
the little boy was another person. Allport's first sentences in his Per-
sonality textbook are, "As a rule, science regards the individual as a
mere bothersome accident. Psychology, too, ordinarily treats him as
something to be brushed aside" (1937, p. vii). Allport was not about to
permit his own individuality to be brushed aside. Allport ends the book
with what is, for a textbook, an unusually stirring paragraph:
Thus there are many ways to study man psychologically. Yet to
study him most fully is to take him as an individual. He is more
than a bundle of habits; more than a nexus of abstract dimensions;
more too than a representative of his species. He is more than a
citizen of the state, and more than a mere incident in the gigantic
movements of mankind. He transcends them all. The individual,
striving ever for his own integrity, has existed under many forms
of social life-forms as varied as the nomadic, feudal, and capitalis-
tic. He struggles on even under oppression, always hoping and plan-
ning for a more perfect democracy where the dignity and growth of
each personality will be prized above all else. (pp. 565-566)
This stress on the individual was not altogether uncommon in psy-
chology by the 1930s. Gordon's older brother Floyd had already made it
a feature of his own influential approach to social psychology-for in-
stance, by attacking the "fallacy" of the "group mind" early in his 1924
textbook and by arguing instead that the so-called "mental structure"
of groups actually consists of "sets of ideals, thoughts, and habits re-
peated in each individual mind and existing only in those minds" (pp. 8-
9). What really distinguished Gordon Allport's position from that of his
brother, and from most other psychologists of his day, was Gordon's
emphasis on the uniqueness of the individual. Early in the final chapter
of Personality, he states:
Implicit in the modern point of view is the demand that psychology
expand its boundaries, revise its methods, and extend its concepts
to accommodate, more hospitably than in the past, the study of the
single concrete mental life.
This demand is thoroughly radical. It is directed against the
practice in general psychology of drawing the blood and peeling the
flesh from human personality, leaving only such a skeleton frame-
work of mind as is acceptable to the sparse canons and methods of
nomothetic science. By stripping the person of all his troublesome
particularities, general psychology has destroyed his essential na-
ture. The newer point of view reverses the perspective. The person
is no longer regarded as a neutral tinted background upon which
the all-important design of mind-in-general stands out. Quite the
reverse: the uniform design traced by general psychology becomes
the ground upon which the integral, three-dimensional, and unique
individual emerges as the salient feature. (p. 549)
That is a dramatic declaration of independence from the nomo-
thetic psychoanalysts and behaviorists of Allport's day. Even more, by
advocating the study not just of the individual but of the unique in-
dividual, Allport declared his uniqueness in comparison with brother
Floyd or with virtually anyone else in psychology at the time. "I am
not that little boy," he was saying; "I am nobody but me."
ALLPORT AS AN ADULT
Finally, having dealt with every other element of the core sen-
tence, "I am not that little boy with the dirt phobia," we are left with
its beginning: "/ am not." Erik Erikson, in his psychobiography of Mar-
52 ALAN C. ELMS
tin Luther, makes much of Luther's so-called "fit in the choir," when
Luther is said to have shouted out, "It isn't me!" or "I am not!" (1958,
p. 23). Declaring what you are not, according to Erikson, is often a way
to establish who you are, what your identity is (p. 36). Gordon Allport,
by his testimony, was not a neurotic, not a little boy, not a generic
personality interchangeable with other personalities; so what was he?
As he suggested in his 1962 "reminiscence" on the origins of the Per-
sonality textbook (" ... some people do grow up ... "), he was a grown-
up, a psychologically healthy adult.
And what does that imply? Well, as he said in the book, "There are
as many ways of growing up as there are individuals who grow, and in
each case the end-product is unique. But," he added, "if general criteria
are sought whereby to distinguish a fully developed personality from
one that is still unripe, there are three differentiating characteristics
that seem both universal and indispensable" (1937, p. 213).
Allport did little to demonstrate the universality or indispensabil-
ity of the three characteristics that he listed. But they are surely char-
acteristics that he felt he himself possessed in full measure, indeed
more than most people-as, on the whole, he did. He identified them as
self-extension, self-objectification, and a unifying philosophy of life
(1937, pp. 213-214). Self-extension involves having "a variety of auton-
omous interests" (p. 213), the incorporation into oneself of the many
things one loves: "Possessions, friends, one's own children, other chil-
dren, cultural interests, abstract ideas, politics, hobbies, recreation, and
most conspicuously of all, one's work" (p. 217, his emphasis). One is
reminded here of the busy Allport, who even as an undergraduate de-
voted himself enthusiastically to volunteer work with boys' clubs, social
service agencies, the Humane Society, and so forth, while joyfully dis-
covering the new intellectual world of Harvard, Cambridge, and Bos-
ton-a far cry from the Freud anecdote's narrowly focused dirt-phobic
boy and his well-starched mother.
Self-objectification includes the development of self-insight and of
that highly correlated quality, a sense of humor. The latter involves,
according to Allport, "the ability to laugh at the things one loves (in-
cluding of course oneself and all that pertains to oneself), and still to
love them" (p. 223). We might wonder about Allport's degree of self-
insight, as he resolutely insisted that that little boy had no relevance to
himself. But Allport certainly did display a sense of humor about the
matter, and he was very self-aware regarding the only level of person-
ality he thought important for normal adults-the conscious level.
Finally among Allport's criteria for a fully developed personality is
a unifying philosophy of life, one that represents to a person ''his place
in the scheme of things" (p. 214). Allport acknowledged that "there are
many ... unifying philosophies," but, he insisted, "Religion is the search
for a value underlying all things, and as such is the most comprehen-
sive of all the possible philosophies of life" (p. 226; his emphasis). Here
the very religious Gordon Allport could feel superior even to his older
brother Floyd, who had publicly given up the idea of a "transcendental,
monistic god" (Allport, 1933, p. 458). On this point Gordon could also
feel superior to Sigmund Freud. He observed that in the recently pub-
lished New Introductory Lectures, "Freud declares himself 'perfectly
certain' that this particular class of 'illusions' [i.e., religion] springs from
infantilism of the mind" (Allport, 1937, p. 227). Allport thought he knew
better; and in knowing better, he recognized himself as the true adult
among this company. While discussing his criteria for personal matu-
rity, Allport listed the major limitations of other approaches to the
psychology of personality, and here he made one simple complaint about
psychoanalysis: "Freudian psychology never regards an adult as truly
adult" (1937, p. 216; his emphasis). Like every personality theorist be-
fore and after him, Allport clearly knew of at least one adult who was
truly adult: the theorist himself.
CONCLUSION
I realize that in focusing on the personal sources of his 1937 per-

sonality textbook, I am selling Gordon Allport short. I wish he were
still around to rally the forces of idiographic psychology; we need some-
one of his vigor and eloquence. I have been greatly impressed by the
wise and pithy editorial correspondence I have seen from his many
years as editor of the Journal of Abnormal and Social Psychology.
If Allport's editorial suggestions had been headed more carefully by
Abraham Maslow, for example, the third-force humanistic psychology
movement might have proceeded on more solid ground. I have seen
much evidence of Allport's private kindness to individuals as different
from him and from each other as Jenny Masterson and Harry Murray.
I agree with Allport's statement, in the Personality text, that "In
biographies ... an inevitable exaggeration of consistency occurs. 'Ir-
relevant' activities and traits are discarded, and the act of discarding
makes for over-simplification .... The writer wishes to extract the 'es-
sence' or meaning of the life. In so doing remarkable unity emerges,
more than was ever present in the animate person" (p. 352, fn. 8).
Surely there was much more to Gordon Allport than the clean little boy
of his Freud encounter. But Allport himself kept telling that story,
54 ALAN C. ELMS
telling it more often than any other story about himself, and tying it
directly to his life's work.
A year after Allport's personality textbook was published, another
milestone in personality psychology appeared: Henry Murray's Explor-
ations in Personality (1938). Murray dedicated his book in part "To
Sigmund Freud, whose genius contributed the most fruitful working
hypotheses ... and to Carl G. Jung, whose writings were a hive of great
suggestiveness." Allport had properly dedicated the Personality text to
his mother, whose virtues and teachings did indeed leave their strong
mark upon the book. But he might well have added to his dedication
these words: "And to Sigmund Freud, whose question stung me to
make this belated (but thoroughly adult) reply."
ACKNOWLEDGMENTS: Extracts from the unpublished papers of Gordon

W. Allport are quoted by permission of Robert B. Allport, the Harvard
University Archives, the Archives of the History of American Psychol-
ogy, and the Library of Congress. The excerpt from an unpublished
letter by Henry A. Murray is quoted by permission of Nina Murray
and the Harvard University Archives. Individuals who provided infor-
mation during the preparation of this chapter include Robert B. All-
port, James W. Anderson, George Atwood, K. R. Eissler, Anthony
Greenwald, Henry A. Murray, Dan Ogilvie, Thomas Pettigrew, M.
Brewster Smith, Philip J. Stone, and Eugene Taylor. Their helpfulness
of course does not make them in any way responsible for my interpre-
tations or conclusions.
REFERENCES
Allport, F. H. (1924). Social psychology. Boston: Houghton Mifflin.

Allport, F. H. (1933). Institutional behaviar. Chapel Hill: University of North Carolina
Press.
Allport, G. W. (1922a). An experimental study of the traits of personality. Unpublished
doctoral dissertation, Harvard University, Cambridge, MA.
Allport, G. W. (1922b). Letter toW. S. Taylor, 25 July. WilliamS. Taylor Papers, Archives
of the History of American Psychology, University of Akron, Akron, OH.
Allport, G. W. (1922c). Letter to W. S. Taylor, 18 December. William S. Taylor Papers,
Archives of the History of American Psychology, University of Akron, Akron, OH.
Allport, G. W. (1923). Letter to W. S. Taylor, 21 March. William S. Taylor Papers, Ar-
chives of the History of American Psychology, University of Akron, Akron, OH.
Allport, G. W. (1928). Letter to W. E. Leonard, 5 April. Gordon W. Allport Papers,
Harvard University Archives, Cambridge, MA.
Allport, G. W. (1929). The study of personality by the intuitive method: An experiment in
teaching from The Locomotive God. Journal of Abnormal and Social Psychology,
24, 14-27.
Allport, G. W. (1937). Persanality: A psyclwlogical interpretation. New York: Holt.
Allport, G. W. (1942). The use of personal documents in psychological science. New
York: Social Science Research Council, Bulletin 49.
Allport, G. W. (1953). The trend in motivational theory. American Journal of Orthopsy-
chiatry, 25, 107-119.
Allport, G. W. (1958). G. W. Allport recalls a visit to Sigmund Freud. Unpublished manu-
script, Sigmund Freud Archives, Library of Congress, Washington, DC, October 29.
Allport, G. W. (1962). My encounters with personality theory. Unpublished manuscript,
recorded and edited by W.G.T. Douglas, Boston University School of Theology, Octo-
ber 29.
Allport, G. W. (1965). Letters from Jenny. New York: Harcourt.
Allport, G. W. (1967). Autobiography. In E. G. Boring & G. Lindzey (Eds.), A history of
psychology in autobiography (Vol. 5, pp. 1-25). Boston: Appleton.
Allport, G. W. (n.d.). Personal experience with racial and religious attitudes. Unpublished
manuscript, Gordon W. Allport Papers, Harvard University Archives, Cambridge,
MA.
Elms, A.C. (1972). Allport, Freud, and the clean little boy. Psychoanalytic Review, 59,
627-632.
Erikson, E. H. (1958). Young man Luther. New York: Norton.
Erikson, E. H. (1969). Gandhi's truth. New York: Norton.
Evans, R. I. (1970). Gordan Allport: The man and his ideas. New York: Dutton.
Faber, M. D. (1970). Allport's visit with Freud. Psychoanalytic Review, 57, 60-64.
Maddi, S. R., & Costa, P. T. (1972). Humanism in personology: Allport, Maslow, Murray.
Chicago: Aldine.
Morey, L. C. (1987). Observations on the meeting between Allport and Freud. Psychoan-
alytic Review, 74, 135-139.
Murray, H. A. (1938). Explorations in personality. New York: Oxford.
Murray, H. A. (1972). Letter to Ada L. Allport, 22 September. Gordon W. Allport Papers,
Harvard University Archives, Cambridge, MA.
Ogilvie, D. M. (1984). Personality and paradox: Gordon Allport's final contribution. Per-
sonality Forum, 2, 12-14.
CHAPTER FOUR
Allport and Murray on

Allport's Personality
A Confrontation in 1946-194 7*
M. BREWSTER SMITH
In the winter of 1946-1947, graduate students in the Department of

Social Relations at Harvard, then in its second year, heard Gordon
Allport, Henry Murray, and Kurt Lewin each talk about his distinctive
approach to personality in the presence of the others and comment
about the others' theories. New graduate students were then required
to participate in two proseminars: one on methods chaired by Jerome S.
Bruner, the other on concepts led by Talcott Parsons. Most of the ad-
vanced students turned out for the occasions (December 15, 1946, and
January 6, 1947) when these three originators of personality psychol-
ogy held forth. I was a privileged participant and took full telegraphic
notes. I remember sharing the common feeling that we were participat-
ing in a joust of Olympians. When I returned to my notes a quarter
century later, they still evoked for me the excitement of the occasion. I
select here the interchange between Allport and Murray on Allport's
theory. What follows is a paraphrase rather than a verbatim recording.
I try, however, to err on the side of fidelity rather than felicity of
expression. I let each protagonist speak in the first person.
*This chapter is adapted from Smith (1971) with permission.
M. BREWSTER SMITH • Board of Studies in Psychology, University of California, Santa

Cruz, California 95064.
New York, 1993.
57
58 M. BREWSTER SMITH
ALLPORT ON ALLPORT
I will reply here to the critics of my book, Personality: A Psycho-

logical Interpretation, and attempt to meet their criticisms insofar as
possible. There have been three main criticisms: First, with respect to
culture, the book disregards the cultural matrix of personality; second,
in regard to its emphasis on uniqueness; and third in regard to its
motivational theory.
THE NEGLECT OF CULTURE

In the new Department of Social Relations, the neglect of culture
is an unpardonable sin. But the sin was deliberate in this book, and I
would commit it again. The book's preface indicates its intended scope,
which is concerned with the structural problem, with what personality
is rather than how it got that way. In my thinking there is a sharp
distinction between personality and social psychology. One must think
of personality as a biophysical or biopsychical structure. It may be
modified by the field, but it is something. Social psychology is every-
thing else. It doesn't make sense to define personality in terms of its
social implications. This leaves it with nothing substantial.
I was reacting against Kimball Young's (1930) social psychology
and also against the anthropological relativists who said that personal-
ity is merely the subjective side of culture-roles and statuses. My
position here was supported by the German life history studies (Allport
et al., 1941), which showed that personality was not disrupted propor-
tionally to the disruption in culture. I was also reacting against Cooley
and G. H. Mead, against James's exaggerated statement about the plu-
rality of social selves, and against Lewin's emphasis on the influence of
the momentary field. Murray and Lewin have treated the environment
better than I in their concepts of press and field. But I was trying to do
something quite different. I may, of course, have exaggerated things in
the other direction.
In response to my critics, I readily admit that culture largely de-
termines the formation of the growing personality. It also determines
much or most of the variability of personality, and its effects. (The
great man theory is wrong). The social factor has everything to do with
the evaluation of personality.
UNIQUENESS
The criticism here has come from the statistically minded and from
Murray. I am unrepentent. There have been three objections. First,
ALLPORT AND MURRAY ON ALLPORT'S PERSONALITY 59
personality becomes inaccessible to science if we are required to regard

it as unique. Second, uniqueness is merely quantitative, not qualitative.
We can encompass "uniqueness" by finer measures of more variables
that jointly produce a fine enough psychography. Third, even though
everything concrete is unique, that has not prevented the development
of the nomothetic physical sciences.
The first of these issues is one of semantics. Following Windelband, I
distinguish two aspects of knowledge: the idiographic and the nomo-
thetic. Most intellectual work requires both of these aspects. In my
book, this is visible in the chapter on understanding personality,
where inference is contrasted with intuition (attending to, concen-
trating on the unique object). The two points of view also apply to
the treatment of traits as common versus unique and in the chapter
on methods. The issue concerning uniqueness is settled in terms of
predictive power, which is greater when idiographic knowledge is
employed. The evidence is not fully conclusive, but suggests that
direct acquaintance works better than general knowledge. But Sarbin
(1943) objects.
In regard to the second objection, it is simply an impracticable
procedure to approach uniqueness via a million variables, since you
would still have to reckon with all interrelationships. One might as well
stay on the molar level.
As to the third objection, I am unimpressed by reference to nomo-
thetic physics, since we are dealing not with stones but with person-
ality. Uniqueness is not the primary characteristic of stones. My
argument with Cyril Burt in the British Journal of Educational Psy-
chology summarizes this (Allport, 1946). Burt seeks his key qualities
from factor analysis. My key qualities would be personal, not universal;
not derived from a hash of all different kinds of personalities. The
results of factor analysis are limited by the tests that are put in and by
the fact that a hash of the personalities of all participants cannot repre-
sent the ego structure of any single one of them.
ISSUES OF MOTIVATIONAL THEORY
In this area I am more vulnerable. Three criticisms have been

made of my position: That it does not provide genetic continuity; that it
is superficial, lacking in depth, antipsychoanalytic; and that it fails to
supply the basis for patterning or cohesion among motives and ought,
strictly speaking, to result in a kind of entropy. All of these criticisms
have some validity, and I am willing to arbitrate.
First, in regard to geneticism. Bertocci (1940) says that I am an
emergent evolutionist, in regard to the emergence of personality and
seltbood. Along with McDougall and Murray, Bertocci would like a

steady, rechanneled source of motivation. In contrast, I feel (Allport,
1940) that we genuinely learn our motives. The genetic continuity is a
matter of overlapping shingles. In the army, illiterates were trained to
read. There were powerful motives supporting their learning to read in
the army: they had to, shame, homesickness. Once they leave the army,
these motives are gone. Will the ex-soldiers lose their skill? No. Why
do they continue to read? Is it because of new motives or of old mo-
tives to which the reading skill is attached? This makes little differ-
ence, since some transformation is necessary.
Next as to the criticism of lack of depth. Remember the times in
which my book was written-the excesses of Freudianism prior to
Horney, French, Rogers, and Alexander. I follow Adolph Meyer (Lief,
1948) in holding that the mechanisms of abnormality differ from those
of normality. In normality, balancing mechanisms predominate; this ho-
meostasis is missing in abnormality. The balancing factors include inte-
gration, benign inhibition and repression, ability to adjust to novelty,
sthenic emotions of hope and love, insight, and a philosophy of life
compatible with reality. In abnormality, we find the dissociative, the
unwholesomely repressive, the compulsions and obsessions, the morbid
fantasies. Freud was interested in the unbalancing mechanisms and
saw them everywhere. He developed a paradigm of personality out of
these. In modern psychoanalysis, a more dynamic ego has been admit-
ted. I overreacted to Freud's statement that the ego is not dynamic,
the metaphor of the riderless horse. This is very false, but it is not the
view of post-Freudians. So the ground is prepared for agreement. I
admit I underestimated the remnants of juvenility and unbalancing ele-
ments in all of us.
As to the third criticism, my treatment of functional autonomy-
mechanisms becoming motives-in Chapter 7 gives the impression of a
lot of separate motives going off in different directions without any
cohesion. This was my greatest blunder in the 1937 volume. The prob-
lem here is somewhat the same as that involved in Lewin's concept of
differentiation, which also leads to entropy. But he gets out of the diffi-
culty by his concept of central regions. I tried to handle the problem in
an eclectic way in my chapter on the unity of personality. The only new
idea in the chapter was that of congruence. I drew from Goethe the
idea of unity through striving, through one's major goals. I was afraid
of two things-German rhapsodic verbalism and the very sterile Amer-
ican self-psychology represented by Mary Calkins. Soon after my book,
Sherif and Cantril (194 7) wrote extensively on the ego as did Lewin
(1935) and Koffka (1935). I felt the need for empirical study of the ego,
and summarized the literature in my 1943 paper on ego psychology

(Allport, 1943). This made a persuasive case for the need for a concept
of the ego. The studies demonstrating the experimental effects of ego
involvement provide an operational definition of the ego, as what makes
the difference in these results. This operationally defined ego repairs
my concept of functional autonomy.
A motive may be autonomous of its origin, but never of the ego.
Mechanisms will not turn into motives unless they serve the life-style
of the person. Striving is a descriptive concept; ego structure is the
underlying one.
MURRAY ON ALLPORT
Allport is concerned with the ego system; with what is conscious,

voluntary, and rational or self-consistent and coherent; what is observ-
able in everyday life or overt and public. He is also concerned with
social issues. His concern with a few of the aspects of the ego system is
good, but not enough.
The child begins with many energies, needs, and emotions not inte-
grated into the ego system. Unlike Allport, I see personality as begin-
ning before integration is attained. In this period, there is expressive
action as differentiated from goal-directed action. This results from the
overflow of internal energies on the motor side and from sensory plea-
sures on the sensory side. This is all id activity, Freud's pleasure prin-
ciple broadened to include things that Freud was not concerned with.
The child is doing this within the sphere of the mother. He gets re-
wards of admiration, love, and compassion.
From the beginning, barriers are put up against some behaviors.
The internalized cultural regulations deriving from the environment
are among the most important constituents of personality. Time,
place, mode, and object rules run athwart the pleasure principle.
Most of the person's needs get satisfied, but within this pattern, not
spontaneously.
When a conscious goal-directed system has developed, the ego comes
in. The child has begun to think how to arrive at a certain goal. He has
to learn, has to make an effort. This is distinct from the early expres-
sive state when he is just blowing off steam. Early the child gets his
rewards from his mother without effort. Soon, he has to work to get
rewards. He has to learn good from bad. Consciousness and the ego
system emerge from barriers and frustration.
Allport presents an ego psychology. I want to include the id-the
child, the involuntary, the unconscious. All of us have a vast Sargasso

Sea of an id: temperament and mood, emotions (we cannot control these-
only the objects and images), needs, and images that have been associ-
ated with needs, that is, fantasies. Whenever we engage in undirected
thinking, there is id activity. If love were an ego process, it would be
planned and rational. But it is not. It is usually a compromise between
id and ego. The good as well as the bad emerges from the id. The id
goes along under the ego all the time. I am interested in superordinate
dynamic systems straddling the ego and id.
ALLPORT IN REJOINDER
Our respective approaches to personality can be put in capsule

thus: For Murray, personality is the life history. The most important
question that he wants to know about a person is, what are his fan-
tasies? For Lewin, the person is the life space. The most important
question for him is, what is the situation? For me, personality is the
individual's intentions. The most important question is, what are his
intentions and values?
I have always dealt with two differences with Murray: (1) His
excessive geneticism. I think that the first three years of life don't
matter very much. The child is flexible. (2) His need theory, which is
too McDougallian, too arbitrary. I don't claim to know about native
energies, but feel sure that they are recast as life goes on. A third
major difference I have with Murray concerns the ego concept.
Back in the 1880s, John Dewey used the concept of soul, not theo-
logically but to represent the functions of unity, coherence, and purpos-
iveness. When the soul went out of fashion, it took with it the cognate
concepts and problems-the self. Freudianism brought back the ego as
a related concept-an integrative organizer-but lost its purposive-
ness, seeing the id as the source of all energies.
I have distinguished eight meanings of the ego concept, of which
Murray adopts one. I embrace all eight (See Allport, 1943).
1. The ego as knower. This is rare in psychology, except for Bren-
tano and the phenomenologists. William James denied the neces-
sity of pure ego in this sense.
2. The ego as object known (e.g., studies of the location of one's
self, the concept of depersonalization).
3. The ego as primordial selfishness (for example, Sterner,
LeDantec). Metaphysical egoism has been mainly European.
4. The ego as dominator or status seeker. Ego involvement often

means that status or pride is at stake.
5. The ego as passive organizer, in Freud, or as conscious organ-
izer, in Murray.
6. The ego as fighter for ends, the center of purpose.
7. The ego as a segregated system of personality. Here I would
classify Koftka and Lewin, except insofar as they say that the
ego always tends upward, which involves senses 4 or 6.
8. The ego as subjectified cultural norms, as exemplified in Sherif
and Cantril.
When there is ego involvement, experiments show that general
traits of personality are obtained. Without ego involvement, there is
specificity.
Peter Bertocci pulls out senses 1 and 6, the knowing and wanting
ego, which is phenomenologically separate as the self. The other senses
represent central aspects of personality which the self evaluates.
The ego need not be conscious. Ego involvement is not coextensive
with consciousness or with voluntarism.
COMMENTARY
The themes that Allport and Murray developed in this confronta-

tion are ones that each of them presented more systematically else-
where. Yet I think there is value in noting what they chose to highlight
in each other's presence for an audience of graduate students. Allport
and Murray were competing for the favor of their graduate students.
Their stylistic differences, the emotional flavor of which eludes my
notes, find a record in them: Allport's magisterial marshaling of argu-
ment from on high as compared with Murray's less orderly plunge into
the midst of the teeming phenomena of personality. What my notes do
not convey, but I remember, are Murray's playfully provocative taunt-
ing of Allport with id-like phenomena, and Allport's high, pink forehead
scalp turning progressively deeper shades of red in response. Indeed,
their former student of much earlier times, Saul Rosenzweig (1970),
had it right when he matched Murray, Allport, and Boring (the histo-
rian of experimental psychology) with the id, the ego, and the superego
in Freudian psychoanalysis: the fit is uncanny.
Understandably, from a graduate student's standpoint, Murray got
the better of the match. Of the sparring partners, Murray was redolent
with charisma that Allport somehow lacked. As a graduate student just
before and just after World War II and then as a young faculty mem-
her, I was more acutely aware of Allport's limitations and less apprecia-
tive of his extraordinary strengths than I came to be in later retrospect
(Smith, 1972, 1990). The vignette stands out in my memory of an early
morning in December 1948, when on my hurried way to teach, having
just greeted my newborn first son at the Boston Lying-in Hospital, I
encountered Gordon on the steps of Emerson hall. As he took the cus-
tomary cigar from my outstretched hand, Gordon beamed his reserved
smile and said, "Very nice, Brewster. Every psychologist should have a
child and a dog." At the moment I could have throttled him! Thereaf-
ter, I understood his doctrine of the ''functional autonomy of motives"
as his ingeniously designed charter to let him leave untidy animals and
children to the behaviorists and psychoanalysts.
But in spite of the undoubted fact that Murray made him uncom-
fortable and represented a view of personality that was distinctly un-
congenial to him, Allport defended Murray with complete magnanimity
when it counted most. Only since Triplett's (1983) dissertation have we
known about the intense struggles behind closed doors concerning
Murray's tenure at Harvard, a closely fought academic battle in which
Allport was Murray's indispensable supporter.
REFERENCES
Allport, G. W. (1940). Motivation in personality: Reply to Mr. Bertocci. Psyclwlogical

Review, 1,.7, 533-554.
Allport, G. W. (1943). The ego in contemporary psychology. Psychological Review, 50,
451-478.
Allport, G. W. (1946). Geneticism vs. ego-structure in theories of personality. British
Jaurnal of Educational Psychology, 16, 57--68.
Allport, G. W., Bruner, J. S., & Jandorf, E. M. (1941). Personality under social catastro-
phe: Ninety life-histories of the Nazi revolution. Character and Personality, 10,
1-22.
Bertocci, P. A. (1940). A critique of G. W. Allport's theory of motivation. Psyclwlogical
Review, 1,.7, 501-532.
Koffka, K. (1935). Principles of Gestalt psychology. New York: Harcourt, Brace.
Lief, A. (Ed.) (1948). The commonsense psychiatry of Dr. Adolph Meyer. New York:
McGraw-Hill.
Rosenzweig, S. (1970). Boring and the Zeitgeist: Eruditione gesta beavit. Journal of
Sarbin, T. R. (1943). A contribution to the study of actuarial and individual methods of
prediction. American Jaurnal of Sociology, 1,.8, 593-602.
Sherif, M., & Cantril, H. (1947). The psyclwlogy of ego-involvements. New York: Wiley.
Smith, M. B. (1971). Allport, Murray, and Lewin on personality theory: Notes on a
confrontation. Journal of the History of the Behavioral Sciences, 7, 353--362.
Smith, M. B. (1972). Toward humanizing social psychology. In T. S. Krawiec (Ed.), The

psychologists (pp. 212-239). New York: Oxford.
Smith, M. B. (1990). Personology launched. Retrospective review of H. A. Murray, Ex-
plorations in Personality, Contemporary Psychology, 35, 537-539.
Triplett, R. G. (1983). Henry A. Murray and the Harvard Psychological Clinic, 1926-
1938: A struggle to expand the disciplinary boundaries of psychology. Unpublished
doctoral dissertation, University of New Hampshire.
Young, K. (1930). Social psychology. New York: Crofts.
PART THREE
CURRENT STATE OF
PERSONALITY PSYCHOLOGY
AND ITS TEXTS
CHAPTER FIVE
Pattern and Organization

Current Trends and Prospects for the Future*
LAWRENCE A. PERVIN
In reading over the 1937 personality texts of Allport and Stagner in

preparation for a colloquium series in honor of them, I was struck with
a passage from the 1980 Annual Review chapter by Jackson and
Paunonen: "We share with Block the view that these earlier authors
frequently expressed central issues in personality theory more co-
gently than do the writings of some contemporary authors who seem to
have rediscovered the same issues" (p. 505). Reference was being made
to Allport, Lewin, Murphy, and Murray, but Stagner could have been
included as well. t
Although clearly one reads into these writings some of what one
believes today, I am struck with the many common points of emphasis
in the writings of Allport and Stagner: traits and the underlying consis-
tency of personality, the dynamics of personality functioning, the im-
portance of affect, concern with the development of personality, and the
importance of culture and an appreciation for cultural differences.
*This chapter is adapted from an address given at the Institute of Personality and Social
Research, University of California at Berkeley, Spring 1987.
tIt was particularly interesting for me to go back and read Allport, something I have been
prone to do over the years, since he was a professor of mine in 1957. At the time I failed
to appreciate his enormous wisdom.
LAWRENCE A. PERVIN • Department of Psychology, Rutgers University, New Bruns-

wick, New Jersey 08903.
New York, 1993.
69
70 LAWRENCE A. PERVIN
I would like to focus on the shared emphasis that I found most

striking and that forms the theme of this chapter: the organization of
personality. Allport was very concerned with the individuality of per-
sonality. Sometimes he attempted to focus on the uniqueness of the
individual through an emphasis on the uniqueness of the units of per-
sonality (e.g., unique traits) while at other times he emphasized the
uniqueness of the organization of personality. Particularly in terms of
the latter, it is clear that Allport was concerned with synthesis, organi-
zation, patterning, and the "unity of personality." And, while Stagner
emphasized the importance of analysis (traits), over time he increas-
ingly became concerned with personality as a dynamic system.
So, taking the concern of Allport and Stagner with the organiza-
tion of personality as a point of departure, I would like to consider
three issues: (1) the nature of the units and structure of personality,
(2) the organization of affects, and (3) the organization of the self.
THE UNITS OF PERSONALITY

AND THEIR ORGANIZATION
Concern with the organization of personality confronts us first with

the question of the units of personality (Allport, 1958). What is it that
is organized? Here I would like to consider how both Allport and Stag-
ner struggled with the conceptual status of traits as the fundamental
organizational units of personality and then go on to consider the con-
cept of goals as a unit that they arrived at and which I believe contin-
ues to have merit today.
CONCEPTS OF TRAIT AND MOTIVE IN ALLPORT AND STAGNER

Although both Allport and Stagner are considered to be trait theo-
rists, their discussion of the trait concept is of considerable interest.
Not only did Allport not view traits as suggesting that the individual
behaved the same way in all situations, a point I and others have made
previously (Pervin, 1985; Zuroff, 1986), but he suggested that traits
expressed a motivational significance. I believe that this reflected an
ambivalence and conceptual confusion on Allport's part, one that he
never completely resolved. On one hand, he rejected narrow motiva-
tional concepts such as instinct and drive because they failed to provide
for the individuality of the organism, whereas on the other hand, he
remained concerned with the dynamics of behavior-that which acti-
vated and directed the organism. In a sense, traditional motivational
PA'ITERN AND ORGANIZATION 71
concepts were seen as too limited to be expressive of the individual,

that is, expressive of the overall organization of functioning that re-
sulted in an expressive style for the individual. At the same time, trait
concepts were insufficiently dynamic. Thus, he asked whether traits
were self-active, and after first suggesting that strictly speaking they
were not, he went on to suggest that "in another sense traits do initiate
behavior'' (Allport, 1937, p. 321). His conclusion was that although not
all motives were traits and not all traits were motives, there was some
overlap between the two: "Some traits thus seem to have motivational
(directional) significance and some mere instrumental significance"
(1937, p. 323). Whereas some traits referred to the expressive-stylistic
aspect of personality, other traits referred to the motivational compo-
nent of personality, what Allport called its "telic significance." And it
was the traits with motivational significance that played a particularly
significant role in the unity of personality.*
Allport also related the distinction between expressive traits and
motivational traits to the phenotype-genotype issue. He pointed out
that the same behavior in different people could have different roots
and that seemingly different behaviors in the same person could have
the same root. Whereas stylistic, expressive traits were more pheno-
typic, motivational traits were more genotypic (Allport, 1937, pp. 324,
358). And while personality psychologists needed to be concerned with
both, the latter were "the very springs of conduct" (p. 326). Further-
more, Allport suggested that personality psychologists needed to be
concerned with interdependent substructures, some of which would be
integrated and others of which would be in conflict. As he expressed it
in a 1939 address at Berkeley, "the organism is, after all, but a living
system of interdependent motives" (p. 108).
It is here that we may draw an interesting comparison with the
views of Stagner, who suggested that although traits were the units of
personality, they remained descriptive of behavior rather than explan-
atory of behavior. And what of explanations? These involved inner
drives or impulses: "We cannot then understand personality without
understanding motivation" (Stagner, 1937, p. 12). This view is stated
even more explicitly in the second edition: "Too many psychologists
write as if a trait were an effective cause of behavior. This is quite
incorrect and misleading.... The trait, then, should be considered de-
scriptive, but not explanatory. The causes of human behavior insofar as
we can identify them are the motives which impel us to act and the
*Buss (1989) similarly includes both temperament and motive concepts under the trait
rubric.
environment which shapes our actions" (Stagner, 1948, p. 144).* And, in

the third edition, published in 1961, Stagner went on to view personal-
ity as a goal system, a view that he related to Allport's earlier empha-
sis on goals. An interesting aspect of this apparent shift to a more
dynamic, explanatory view is that Stagner suggested that the goal con-
cept was "hardly different from our earlier usage of trait" and that "the
concept of trait should be reevaluated as having dynamic implications"
(1948, p. 358). Further, he noted that "Allport, after having earlier ac-
cepted the separation, seems to have shifted and accepted the notion of
identity of trait and motive" (p. 359). Perhaps this is why McClelland
was led to suggest that ''it appears that Allport has stretched the term
trait a little too far. He is now using it to account for inconsistencies in
behavior, new responses which seem primarily determined by the per-
son's wishes or goals rather than by his past adjustment in similar
situations" (1951, p. 214).
In sum, both Allport and Stagner were concerned with the units of
personality and with the organization of these units. Both began with
a trait concept and struggled with the relation between traits and mo-
tives, descriptions and explanations, phenotypes and genotypes. Both
came to emphasize the importance of goals, or what Allport (1955)
called propriate strivings, for the organization and unity of personality.
Both appreciated the complexity of behavior from a systems perspec-
tive in terms of the varied roots that the same behavior can have
(equifinality), the varied paths that the same motive can follow (equi-
potentiality), the multidetermination of most behavior, and the poten-
tial for conflict as well as coherence or integration in system
functioning.
THE GoAL CONCEPT
I would like next to consider goals, both as units of personality and

as a basis for understanding the dynamic, organized aspects of person-
ality functioning. In particular, I want to call attention to two aspects of
system functioning, multidetermination and conflict, since they have
been of interest to me as a clinician and researcher. That most behavior
is multidetermined or-as psychoanalysts would have it, over-
determined-would appear to be self-evident. Yet, my sense is that it is
relatively neglected in our thinking and research. My own work on the
goal concept made clear the extent to which people view their own
behavior in important situations as involving multiple goals. Rarely do
*Wiggins (1973) similarly describes traits as descriptive rather than explanatory.

we do something for one reason alone. Consider, for example, the con-
fessions of a lawyer who seeks to answer the question of why he has
fought so hard for the interests of the guilty. In this Newsweek article,
the author suggests such motives as the following: emotional identifi-
cation with the underdog, performing well during a trial, a need for
power and admiration, winning, beating witnesses, voyeurism, and ex-
hibitionism. He concludes that "in trying to understand myself and my
work, I am led ineluctably to the murky and subjective realm of what I
brought with me when I first stepped into a courtroom. It was clearly
not just the belief that every criminal defendant has a right to counsel"
(Wishman, 1981, p. 25). Or, consider the more recent article in the New
York Times that suggested that women who choose to become surro-
gate mothers have such diverse and multiple motivations as money,
feeling confirmed as a woman, doing something special, altruism, grati-
tude for the life given them after they were adopted, and guilt over a
previous abortion (Goleman, 1987, p. C1). Indeed, this article suggests
many of the principles of system functioning that are suggested in the
works of Allport and Stagner and that seem particularly key to me:
equipotentiality, or the potential for similar motives to lead to quite
different (even contrasting) behaviors; equifinality, or the potential for
the same behavior to express different motives; and multidetermina-
tion, or the principle that behavior expresses the interaction among
multiple motives.
Similarly self-evident would appear to be the importance of con-
flict in our lives, here referring to conflict internal to the system,
though clearly conflict is a major issue in close interpersonal relations
(Peterson, 1989) as well as in the international relations level of system
functioning. Lewin (1935) gave considerable attention to the issue of
conflict, as did Dollard and Miller (1950) in their extrapolations from
psychoanalytic theory to stimulus-response theory, but the concept has
ceased to play a major role in most current personality theory and
research.
In my earlier research on goals (Pervin, 1984), I examined the
implications of goal conflicts for how situations are experienced. Each
of ten subjects rated the relevance of 11 conflicts to 20 situations, the
conflicts being standard for the group but the situations being idio-
graphic for the subject. The list of conflicts had been drawn from those
suggested by a small group of subjects and included items such as
"Relax, enjoy versus Do what 'should' do." and "Assert self versus
Gain acceptance, approval." Generally, situations that the subjects as-
sociated with negative affect were also found to be associated with
high conflict.
I have just begun to pursue further this line of research, again

having subjects provide situations representative of their daily lives,
and have extended the list of goal conflicts to include some suggested
by the work of Karen Horney, as well as others suggested by the work
of Timothy Leary on interpersonal relations. Some of the issues being
investigated are: critical features of situations of goal conflict, the de-
terminants of how conflicts are handled, and sex differences in the
nature of goal conflicts and patterns of handling conflict.
While the data are still being analyzed, some of the preliminary
findings may be of interest. Of course, it is no surprise to find enor-
mous individual differences in the amount of overall conflict experi-
enced, in the conflicts reported to be most central, and in the reported
pattern of handling various conflicts. Overall, there is a tendency for
the amount of conflict reported to be associated with separate self-
report measures of anxiety, depression, and the tendency to blame one-
self for negative events, though not with a measure of defensiveness.
These results appear to fit with those reported by Emmons (1989),
suggesting a relation between life satisfaction and a "personal strivings
system" free of conflict, as well as a relation between conflicts over
emotional expression and problems of psychological and physical well-
being.
Most of the sex differences considered so far have been small, but
a few patterns appear to be emerging. First, males report a higher
overall level of conflict than do females. Second, whereas males appear
to experience more conflict over intimacy, females appear to experience
more conflict in relation to issues of expression of anger and domi-
nance-submission. Finally, whereas males report a greater tendency to
handle conflict through doing what feels best or acting impulsively,
females report a greater tendency to handle conflict through compro-
mise and avoidance.
A final aspect of the goals concept worthy of consideration is its
place in social cognition; that is, the extent to which we anticipate and
account for behavior in terms of assumptions we make about the mo-
tives or goals of others and ourselves (Pervin, 1989; Read, 1987; Read
et al., 1989). This is a point emphasized by Heider (1958) but often
neglected in the attribution literature. In a relevant study, Adrian
Fumham and I (Pervin & Fumham, 1987) had subjects rate the proba-
bility of each of 16 behaviors in 16 situations, with the 16 situations
representing all combinations of four situation categories and four
goals. The resulting ratings indicated that personal goals influenced
behavioral expectations far more than did situational types, with the
exception of behaviors associated with affect and the face (e.g., crying).
This tendency for emotions to be perceived to be highly situationally

determined is a common finding in the literature.
Furthermore, in an apparent reversal of the "fundamental attribu-
tion error," subjects expected their own behavior to vary less than the
behavior of others! In other words, they expected their own behavior
to be more stable than they expected the behavior of others, whereas
the "fundamental attribution error'' suggests that people see the be-
havior of others as stable and traitlike and "more accurately" perceive
their own behavior as relatively situation-specific. One possibility sug-
gested for this apparent discrepancy with the attribution literature is
that goals rather than traits were considered. Another possibility is
that expectations rather than postevent attributions were considered.
The latter suggests the possibility that people really expect themselves
to be far more stable than social psychologists have suggested, at least
more stable than they expect others to be. When they do find them-
selves behaving "inconsistently," they are even more prone to attribute
their own behavior in the situation to the situation than they are to
attribute the behavior of others. In other words, people may be more
trait theorists, or at least motivation theorists, for themselves as well
as others than social psychologists have recognized. It may also be of
interest to note here the findings of Asch and Zukier (1984) that sug-
gest that people perceive themselves and others as psychological
units-units that have patterning and order, units that are complex,
and units that have the potential for conflict.
Consideration of the question of social cognition may bring us back
to the question of traits versus motives as units of personality.
Fumham and I intended to conduct a study comparing trait-based ex-
pectations of behavior with goal-based expectations. We never did get
around to conducting this study, but my sense is that traits and goals
represent two different ways of organizing information about people
and social action. Traits have heuristic value in enabling us to package
together a great deal of information, to smooth out irregularities, and
form relatively global impressions of people that provide for general-
ized expectancies. Motives have heuristic value in enabling us to make
more specific predictions based on an assessment of particular aspects
of the individual's personality and situational factors. Overall, one
would expect there to be a relationship between traits and motives in
the same person, as Allport and Stagner suggested. And, as Asch and
Zukier (1984) noted, motives may help us to reconcile apparent discrep-
ancies between traits and specific behaviors. However, traits and
motives are distinct concepts, each with its own heuristic value and
limitations.
If this is the case in social cognition, then it may suggest a parallel

view for us as personality psychologists; that is, as Murray (1938) and
McClelland (1951) suggested, we treat traits and motives as distinct
concepts, with some overlapping relationships but with differing func-
tional utilities-the former (traits) involving broad consistencies in be-
havior and expressive of regularities and the latter (motives) involving
the interplay between the regular and unusual in the person, or what I
have more broadly described as the stasis and flow of behavior (Pervin,
1984). Recent research indicates that it is possible to map trait and goal
concepts onto one another, suggesting that study of relationships be-
tween the two units may yet be possible (Costa & McCrae, 1988; Read
et al., 1989).
I would like to conclude this section by emphasizing the difficulty
of devising a system for representing the organization of personality,
in particular the organization of goal systems. Ordinarily we think of
organization in terms of hierarchical structures and, where assessment
is involved, in terms of score profiles. However, it seems to me that the
organization of personality is a much more complex matter, involving a
more complex relationship among the parts than a hierarchical analysis
or profile of scores provides for. What is needed is a system of repre-
sentation that captures both the dynamics of action and the degree of
integration or conflict among the parts. Allport (1937, pp. 246, 358) was
critical of the factorial conception of personality as a system of indepen-
dent elements and, as noted, argued instead for a view of interdepen-
dent substructures. Development of a scheme for representing the
complexity of organization of personality would go a long way toward
providing for appreciating the individuality that Allport so cherished,
without resorting to the investigation of unique units for each person.
To my mind, George Kelly (1955) came closest to the systems view
being emphasized, suggesting that the person's construct system had
both structure and fluidity to it.
AFFECT
I would like now to turn briefly to the question of the organization

of personality in relation to two additional areas: affect and the self. It
is my sense that for some time the importance of affect for personality
functioning was a neglected area, with the exception of such individual
affects as anxiety and such psychologists as Tomkins (1962). In part,
this was because of the emphasis on overt behavior during the 1940s
and 1950s, followed by the emphasis on cognition in the period since
PATIERN AND ORGANIZATION 77
then. The situation has changed of late, to the extent that affect is a
major concern of "cognitive-behavior therapists." However, much re-
mains to be done in the area of the organization of affects within the
person.
My own interest in the area of affect was spurred by a number of
factors: a long collegial relationship with Silvan Tomkins, dating back to
1962; ongoing practice as a psychodynamically oriented clinician; re-
search on free-response descriptions that primarily elicited affect de-
scriptors of situations; and a study of consistency of response across
situations that suggested that similar situations might not elicit similar
behaviors if they differed in a characteristic that elicited strong affect
(Champagne & Pervin, 1987).
I will describe some of the results of my research on affect and
personality, particularly as they relate to the issue of organization and
pattern, but let me note that this proved to be the most difficult and
frustrating piece of research in which I have engaged. I believe that
there is gold there, but the mining is difficult, involving the hazards of
digging deep and following false leads.
In any case, the research basically involved subjects rating the
relevance of an extended series of affects to situations and people they
had identified as being relevant to them. In addition, in one phase of
the research, subjects rated a list of prototypic situations and proto-
typic situation features for various affects to see if subjects could agree
about the nature of the situations and defining characteristics of situa-
tions critical for each affect. This research followed that suggested by
Buss and Craik (1983) in their work on traits. Finally, subjects indi-
cated the second-most likely affect they would experience in each
prototypic situation. This procedure was used to consider further rela-
tionships among affects and individual differences in the organization of
affects in prototypic as well as idiosyncratic situations. A feature of this
research, which has tended to be typical of the research I have con-
ducted over the years, is a mix of standard and free-response data, of
nomothetics and idiographics, of an interest in general principles estab-
lished through the study of patterns in individuals.
Generally, similar affect factors were found for individuals, with
three positive affect factors (Happy, Attraction-Love, Friendship) and
four negative affect factors (Anger Out, Anger In-Blows to Self-
esteem, Distress, Envy-Jealousy) being typical. In other words, gener-
ally individuals appear to organize affects in reasonably similar ways,
at least at the highest level of organization. This was true regardless of
whether subjects were rating affects in relation to situations, people, or
other affects. At the same time, there were considerable individual
differences in the relative importance of individual affects as well as in

the relations or correlations between specific affects. In relation to the
former, two individuals might have the same mean value for an affect
across the situations, but for one subject this might indicate an affect of
considerable import while for another it might indicate an affect of
minimal import. For example, two individuals had similar overall mean
ratings for the affect jealousy. In one subject this meant that it was the
least important of all 50 affects rated while for another subject it meant
that it was of average importance.
In relation to the latter-the relations among affects-while virtu-
ally all subjects reported a high correlation between friendship and
trust, there was considerable variability in the extent to which love and
trust were seen as related to one another. Similarly, while subjects
generally agreed on the prototypicality of situations and situation char-
acteristics, there were considerable individual differences in the next
most likely affect to be experienced in a prototypic situation, with this
measure of association found to be related for the individual to that
determined by an analysis of other affect association data. In other
words, even in situations that are so powerful as to elicit virtually the
same predominant affect in most individuals, the further affective elab-
oration will vary considerably from individual to individual.
The data from this research clearly indicate that most situations
are experienced in terms of combinations of affects, including at times
positive and negative affects. I am reminded here of a patient who
recently told me of the range of emotions he experienced while caring
for his terminally ill father: love, hate, guilt, sadness, and anxiety. Fur-
ther, while one can find evidence of an overall, general organization of
affects, there exists considerable individual variation in the patterning
of relative importance of affects and relationships among affects. These
data fit with the results of research I have conducted on the relation
between affect and addiction (Pervin, 1988), where different patterns of
affect are associated with different drugs and drug versus nondrug
states, and with the research of Lewis and Michalson (1982) on the
organization of affects in children.
In concluding this section, I would like to note that with the study
of a very restricted range of affects, one may miss some affects that,
while infrequently experienced, may be of considerable import. Even
where an affect is rarely experienced by a particular subject, it may
play a key role in the person's functioning, as is so often observed by
clinicians. For example, I have recently become quite impressed with
the importance of the affect of shame in the organization of personali-
ties of some individuals, regardless of the frequency with which it is
PATTERN AND ORGANIZATION 79
experienced. Indeed, much of a person's life may be organized around

the feeling of shame, involving how the person interacts with others
and feels about the self.
In sum, the importance of the organization of affects in personality
functioning is easily lost in studies where but a few major affects are
investigated and only nomothetic data analyses are used.
THE ORGANIZATION OF THE SELF
Finally, I would like to consider the question of the organization of

the self. Although I have not conducted research in this area for some
time, I would like to consider it because it concerned both Allport and
Stagner and because it raises profound questions concerning the issue
of organization.
In writing the first edition of Current Controversies and Issues in
Personality (1978), I debated whether to include a chapter on the self.
As I noted, interest in the self had waxed and waned, and I was not
sure where it was headed. Indeed, in 1955, Allport found it necessary
to address the question "Is the concept of the self necessary," conclud-
ing, of course, as did Stagner (1948), that the concept of the self was
necessary for capturing the organization of personality. Since that time,
research on the self has grown enormously so that it now plays a signi-
ficant role in the areas of personality and social cognition. Yet, it seems
to me that little of this research concerns the question of organization
or how the person is able to organize variability in behavior over time
and across situations into a coherent representation of the self.
We find considerable evidence that people behave differently at
different times in their lives, behave differently in different situations,
and play different roles or present themselves differently in different
situations. As Schlenker (1985) notes in a recent edited book on the
self, there is both the inner self, concerned with personal experience,
and the outer self, concerned with self-presentation. There is, then, the
question of the organization of the inner self, the outer self, and the
integration of the two. Scheibe (1985), presenting a historical review in
the same book, notes the following frequently quoted passage from
James (1890, p. 294):
Properly speakillg, a man has as many social selves as there are indi-
viduals who recognize him and carry an image of him in their head.
But as the individuals who carry the images fall naturally into classes,
we may practically say that he has as many different social selves as
there are distinct groups of persons about whose opinion he cares.
However, Scheibe goes on to criticize those who quote this passage
with a bias toward division, fragmentation, and pluralism rather than

with a more complete view of the self as an inward and active agent.
Interestingly, he neglects to mention the work of Allport who, it seems
to me, did as much as anyone over the years to keep the concept of the
self in the forefront of our attention.
We are aware of the fragmented nature of the self that can often
be experienced, reaching extreme proportions in the cases of schizoid
personalities and multiple personalities. How, then, do we account for
such pathological developments as well as for the ability of most of us
to retain some degree of cohesion and integration within inner and
outer, between inner and outer, between earlier and later? How is it
that we look, feel, and seem very different as adults than as children,
and yet retain a sense of continuity? I know that there is important
evidence of such continuity (Block, 1971), but I think that even these
data leave considerable room for consideration of this question. And
how is it that we can feel and look so different in social situations and
yet here, too, retain a sense of self, as if one was looking at a holo-
graphic image from different angles?
Again we come to fundamental questions of pattern and organiza-
tion, questions of integration and conflict, questions that were of funda-
mental concern to Allport and Stagner, and that all too rarely are the
focus of concern in today's literature. Perhaps an exception to this, and
an indication of things to come, is the interest expressed by a number
of authors in the 1985 Schlenker volume in the work of Prescott Lecky
on self-consistency. Perhaps another is the recent effort by Cantor and
Kihlstrom to consider the self system as a family of selves-a fuzzy set:
The meaning of the self is given in the family of selves. The unity
of self comes from the many overlapping resemblances among the
different selves.... You are yourself because of this network of over-
lapping features which are characteristic of the family of selves.
And, you are many things in many places with different people....
The self-concept, therefore, must represent both the variety and
the unity within each person's family of selves.... As such, the cog-
nitive definition of self can be varied, encompassing more aspects of
self than would a representation of a core self, yet still serving as a
unified mental structure. (1987, p. 124)
CURRENT STATUS AND PROSPECTS FOR THE FUTURE
In a 1939 APA address given by Allport in Berkeley, he reviewed

50 years of progress in psychology. How do we stand now in the field of
personality as we review 50 years of progress since the publication of
PATI'ERN AND ORGANIZATION 81
Allport's and Stagner's texts? Alternatively, a Handbook of Personal-

ity: Theory and Research (Borgatta & Lambert, 1968) was published
over 20 years ago; how far have we come since then? Having just
finished editing a handbook of personality (Pervin, 1990), I fmd it an
opportune time to consider this question as well as the field's prospects
for the future. I believe that a review of the current literature presents
a relatively optimistic picture of the field, though one that must be
tempered by an awareness of periods of past optimism followed by
disappointment, specifically in the later 1950s and early 1960s. What
most impresses me about the field today is the increased emphasis on
the complexity of personality and increased diversity of concepts and
research methods utilized. For example, I am encouraged by the
greater attention being given to the question of motivation and rela-
tions among cognition, affect, and behavior, as well as the use of expe-
rience sampling, field study, and case study methods in addition to the
traditional use of questionnaires and experimental laboratory methods
(Craik, 1986). In addition, my sense is that the division between per-
sonality and other parts of psychology, in particular social psychology,
has become blurred. Although some see this as a threat and feel that
the turf of personality must be safeguarded against intrusion by outsid-
ers, I believe that this is a healthy development since it encourages us
to broaden our horizons and make use of conceptual and methodological
advances in other areas. I also believe that Allport and Stagner would
have been sympathetic to this view. In fact, Allport argued for a psy-
chology of personality, which focuses on the human person without de-
stroying the traditional structure of general psychology, and against a
personalist psychology, which seeks to focus on the individual and de-
molish the entire edifice of general psychology (Allport, 1937, p. 550).
As I have noted, both also emphasized the importance of culture and an
awareness of cultural differences, an emphasis that I believe is return-
ing to the field.
Historically there have been many issues that have been of contin-
uing concern to personality psychologists. I have attempted to outline
these in my 1990 Handbook chapter, which reviews the history of mod-
ern personality theory and research, but here I would like to focus on
three issues that I see as particularly problematic at this time. Not
surprisingly, we will once more see connections to the work of Allport
and Stagner. The first issue concerns the interplay between stability
and change, what I have called the stasis and flow of behavior. Histori-
cally we have become bogged down in the person-situation controversy
in a way that I think has been unproductive. The issue is not whether
persons or situations are more important, but rather how we can con-
ceptualize the dynamic interplay between the person and the environ-
ment over time and across situations; that is, the ways in which people
bring coherence to their functioning while remaining adaptive to chang-
ing external circumstances. As Mischel (1990) has noted, his influential
1968 book was an effort to call attention to exactly this question, for
the most part neglected by trait and psychoanalytic approaches, rather
than an attack on the field of personality or the existence of individual
differences.
The second issue concerns the kinds of laws and predictions that
may be possible for us as personality psychologists. Perhaps the most
succinct way to express this is to state the possibility that human be-
havior is so multiply determined and influenced by highly idiosyncratic
meaning structures that general laws and specific predictions may be
unrealizable goals. Frequently of late I have come across the sugges-
tion that ultimately psychology will be composed of three segments:
neuroscience or neurobiology, artificial intelligence or cognitive science,
and some version of folk psychology or hermeneutics. One already sees
splits along these lines in many major departments of psychology, the
first two often being seen as part of the "hard sciences" and the latter
as part of the "soft sciences." Although I do not share this view, it is
raised, in one form or another, by some very respectable thinkers who
are not necessarily hostile to "soft psychology" generally or personality
in particular. One of the most provocative thinkers along these lines is
the anthropologist Richard Shweder (1989, 1990), who, in defining the
field of cultural psychology and the semiotic person, argues against the
view of people as the same wherever you go and instead for a view of
people as intentional beings who live in a world that is domain-specific,
cross-culturally diverse, and historically variable.
Finally, there is the issue of the organization of personality. Here I
fear that little progress has been made. Whether because of the dif-
ficulty of the problem, forces in the field that operate against such
exploration, or the lack of methodological tools appropriate for such
investigation, we find little consideration of pattern and organization in
personality functioning. In this sense I am in agreement with Carlson
(1984), though I wish to distinguish this issue from consideration of the
issues of idiographic versus nomothetic and laboratory versus natural-
istic research. In his 1937 text, Allport observed that "the truth of the
matter is that the total organization of personality is still a new and
poorly formulated problem in psychology. It is a many-sided issue
whose solution yet lies in the future" (p. 365). Fifty years later, the
problem is hardly new, though the rest of what Allport had to say
PATIERN AND ORGANIZATION 83
would still apply: it is poorly fonnulated, it is many-sided, and the

solution lies in the future.
REFERENCES
Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt, Rine-
hart, & Winston.
Allport, G. W. (1939). Review of fifty years of personality. Paper delivered at the meet-
ings of the American Psychological Association, Berkeley, California.
Allport, G. W. (1955). Becoming: Basic considerations for a psychology of personality.
New Haven: Yale University Press.
Allport, G. W. (1958). What units shall we employ? In G. Lindzey (Ed.), Assessment of
human motives (pp. 239-260). New York: Holt, Rinehart and Winston.
Asch, S. E., & Zukier, H. (1984). Thinking about persons. Journal of Personality and
Social Psychology, 46, 1230-1240.
Block, J. (1971). Lives through time. Berkeley, CA: Bancroft.
Borgatta, E. F., & Lambert, W. W. (Eds.) (1968). Handbook of personality theory and
research. Chicago: Rand MeN ally.
Buss, A. H. (1989). Personality as traits. American Psychologist, 44, 1378-1388.
Buss, D. M., & Craik, K. H. (1983). The act frequency approach to personality. Psycho-
logical Review, 90, 105-126.
Cantor, N., & Kihlstrom, J. F. (1987). Personality and social intelligence. Englewood
Cliffs, NJ: Prentice-Hall.
Carlson, R. (1984). What's social about social psychology? Where's the person in person-
ality research? Journal of Personality and Social Psychology, 47, 1304-1309.
Champagne, B., & Pervin, L. A. (1987). The relation of perceived situation similarity to
perceived behavior similarity: Implications for social learning theory. European
Journal of Personality, 1, 79-91.
Costa, P. T., Jr., & McCrae, R. R. (1988). From catalog to classification: Murray's needs
and the five-factor model. Journal of Personality and Social Psychology, 55, 258-
265.
Craik, K. H. (1986). Personality research methods: An historical perspective. Journal of
Dollard, J., & Miller, N.E. (1950). Personality and psychotherapy. New York: McGraw-
Hill.
Emmons, R. A. (1989). The personal striving approach to personality. In L. A. Pervin
(Ed.), Goal concepts in personality and social psychology (pp. 87-126). Hillsdale,
NJ: Erlbaum.
Goleman, D. (1987). Motivations of surrogate mothers. New York Times, January 20, Cl.
Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.
Jackson, D. N., & Paunonen, S. V. (1980). Personality structure and assessment. Annual
Review of Psychology, 31, 503-551.
James, W. (1890). Principles of psychology. New York: Holt.
Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton.
Lazarus, R. S. (1990). Theory-based stress management. Psychological Inquiry, 1, 3-13.
Lewis, M., & Michalson, L. (1982). Children's emotions and moods: Developmental the-
ory and measurement. New York: Plenum.
McClelland, D. C. (1951). Personality. New York: Sloane.

Mischel, W. (1990). Personality dispositions revised and revisited: A view after three
decades. In L. A. Pervin (Ed.), Handbook of personality: Theory and research
(pp. 111-13.1,). New York: Guilford.
Murray, H. A. (1938). Explorations in personality. New York: Oxford University Press.
Pervin, L.A. (1978). Current controversies and issues in personality. New York: Wiley.
Pervin, L. A. (1984). The stasis and flow of behavior: Toward a theory of goals. In M. M.
Page (Ed.), Personality: Current theory and research (pp. 1-53). Lincoln: University
of Nebraska Press.
Pervin, L.A. (1985). Personality. Annual Review of Psychology, 36, 83-114.
Pervin, L. A. (1988). Affect and addiction. Addictive Behaviors, 13, 83-86.
Pervin, L. A. (Ed.) (1989). Goal concepts in personality and social psychology. Hillsdale,
NJ: Erlbaum.
Pervin, L. A. (Ed.) (1990). Handbook of personality: Theory and research. New York:
Guilford.
Pervin, L. A., & Furnham, A. (1987). Goal-based and situation-based expectations of
behavior. European Journal of Personality, 1, 37-44.
Peterson, D. R. (1989). Interpersonal goal conflict. In L. A. Pervin (Ed.), Goal concepts
in personality and social psychology (pp. 327--362). Hillsdale, NJ: Erlbaum.
Read, S. J. (1987). Constructing causal scenarios: A knowledge structure approach to
causal reasoning. Journal of Personality and Social Psychology, 52, 288--302.
Read, S. J., Jones, D. K., & Miller, L. C. (1989). Traits as goal-based categories: The
importance of goals in the coherence of dispositional categories. Paper delivered at
the meetings of the Society of Experimental Social Psychology, Santa Monica, CA.
Scheibe, K. (1985). Historical perspectives on the presented self. In B. R. Schlenker
(Ed.), The self and social life (pp. 33-58); New York: McGraw-Hill.
Schlenker, B. R. (1985). Preface. In B. R. Schlenker (Ed.), The self and social life. New
York: McGraw-Hill.
Shweder, R. A. (1989). Cultural psychology: What is it? In J. W. Stigler, R. A. Shweder,
& H. Herdt (Eds.), Cultural psychology: The Chicago symposia on culture and
human development (pp. 1-43). New York: Cambridge University Press.
Shweder, R. A. (1990). The semiotic person of cultural psychology. In L. A. Pervin (Ed.),
Handbook of personality: Theory and research (pp. 399-416). New York: Guilford.
Stagner, R. (1948). Psychology of personality (2nd ed.). New York: McGraw-Hill.
Tomkins, S. S. (1962). Affects, imagery, and consciousness. New York: Springer.
Wiggins, J. S. (1973). In defense of traits. Unpublished paper. University of British
Columbia, Vancouver, BC.
Wishman, S. (1981). Newsweek, November 9, p. 25.
Zuroff, D. C. (1986). Was Gordon Allport a trait theorist? Journal of Personality and
CHAPTER SIX
The Continuing Relevance of

Personality Theory
SALVATORE R. MADDI
In academic circles, the past 20 years have shown a decline in the

influence of personality theorizing. For the first part of that period, the
person-situation debate fulminated, with the social psychologists and
behaviorists joining with some in the personality field to argue that
situation rather than person variables account for behavior. Although
there were counterarguments raised, some psychologists came to won-
der whether personality exists at all, and even those who did not go
this far were likely to conclude that the personality area is a mess.
Finally, cooler heads prevailed, and an interactional accord was reached
in which behavior was regarded as a joint function of situational and
person variables, with the emphasis of empirical study being on the
interaction of the two.
The interactional accord did not, however, bring back into the aca-
demic fold the elaborate form of personality theorizing aiming at com-
prehensive understanding that characterized earlier times. To the
present, personality research has focused on single dependent variables
rather than broader swatches of behavior and circumscribed or middle-
level theorizing rather than more elaborate conceptualizations. In clin-
ical practice, the comprehensive personality theories still hold sway,
though that is a major reason why clinicians are regarded by academi-
SALVATORE R. MADDI • School of Social Ecology, University of California, Irvine, Cali-

fornia 92717.
Fifty Years of Perso'/W,lity Psychology, edited by Kenneth H. Craik et al. Plenum Press,
New York, 1993.
85
86 SALVATORE R. MADDI
cians as scientifically inadequate. Gone are the research efforts of an

earlier time that aimed at subjecting comprehensive personality theo-
ries to empirical evaluation. The closest we are to this earlier emphasis
now is with the act and prototype theories and research, but these
approaches cover only a small portion of the previous agenda of person-
ality study.
Along with this shift away from comprehensive personality theo-
rizing has come a decrease in influence for psychologists who consider
themselves to be in the personality area. In the 1970s, personality re-
search became harder to publish and professional organizations hith-
erto emphasizing personality became progressively more defined by
social psychology. Even academic jobs in personality became hard to
get, as this area appeared more and more irrelevant. As the 1980s
ended, there were signs that this trend might be decelerating or even
reversing, as long as the person pursuing personality study combined it
with other more central emphases, such as developmental, cognitive, or
social psychologies, and eschewed those comprehensive personality the-
ories that appear so antiquated.
At the risk of also appearing antiquated, I wish to argue for the
continuing relevance of comprehensive personality theorizing. Indeed, I
find it increasingly more important, as academic psychology ever more
clearly shows the telltale shortcomings of its 20-year disavowal of ef-
forts at comprehensive understanding of human behavior.
PSYCHOLOGY: THE FRAGMENTED DISCIPLINE
There are many signs that psychology is currently an extremely

fragmented discipline. To explicate this contention in what follows, I
will find it useful to give some examples, though my aim is not to lay
blame. Indeed, blame laying is a sign of fragmentation, and there has
been too much of this in our discipline. My stance is that the excessive
fragmentation is hurting us and that comprehensive approaches to per-
sonality constitute one kind of antidote.
THE PREVALENCE OF WARRING FACTIONS
In recent times, psychology has shown a pronounced tendency to-

ward what Gordon Allport (1955) called "simple and sovereign theoriz-
ing." By this, he meant an approach to theorizing that singled out one
(or at most a very few) variable or assumption as essential, as all that
is needed for understanding. Also part of this approach is a penchant
CONTINUING RELEVANCE OF PERSONALITY THEORY 87
for extolling the virtues of one's own variable at the expense of the
variables being extolled by others.
Allport clearly believed simple and sovereign theorizing to be too
limited and belligerent, in short, fragmenting to foster the scientific
development of a field. It is a kind of misuse of criticality that is unfor-
tunately very much alive in psychology. Indeed, most of the person-
situation debate already mentioned hinged on simple and sovereign
theorizing. Social psychologists and behaviorists (e.g., Bern, 1972;
Mischel, 1968) pounced on some evidence that personality measures
show only modest correlations with behavioral indices to argue that it
is situational pressures that are really important. Pushed on to the
defensive, some personality psychologists responded by insisting that
person variables control the lion's share of behavior (e.g., Alker, 1972;
Epstein, 1984). The ensuing fight over whether the situationists or the
personalists would triumph proceeded merrily despite the existence of
findings indicating that the interaction between situation and person vari-
ables accounted for the largest portion of behavior variance (e.g., Bowers,
1973). It took years before cooler heads could prevail and the interactional
accord was formulated (Bowers, 1973; Ekehammer, 1974). And even now,
some psychologists would prefer to keep fighting (e.g., Fiske, 1974).
Nor was this the first time the person-situation debate flared up.
Early this century, William McDougall (the Harvard instinct theorist)
and John B. Watson (the founder of behaviorism) engaged in long, heated
debate over whether instincts or situations were the real cause of be-
havior. Some feel that one or the other protagonist triumphed, but the
debate appears to have been rather inconclusive. Before too long, it
was Carl R. Rogers and B. F. Skinner who engaged in another pro-
longed debate concerning whether person-centered counseling freed per-
sons to follow their own inherent potentialities or merely shaped their
behavior in directions favored by their counselors. Despite its focus on
what the psychotherapist is really doing, this is clearly a form of the
person-situation debate. So inconclusive was the confrontation between
these two psychologists that the debate was continued by their stu-
dents. If there was any upshot of all this effort, it suggested that if the
unconditional positive regard person-centered counselors think they are
extending to their clients is really a shaping process, its end result is to
increase the likelihood of unusual or idiosyncratic responses that could
be described as freedom.
Whenever the person-situation debate bursts forth, it ends incon-
clusively. The reason for this is that it is a false debate in the first
place. It takes just a bit of thought to realize that behavior could never
be exclusively a function of either person or situation variables. Indeed,
behavior is a complex amalgam of both sources of influence and, hence,

all that is viable is an interactional accord. Early in the history of
American psychology, William James formulated this position as "the
transitoriness of instinct." By that he meant that if there are instincts,
they are only pure the first time they are expressed. After that, we are
confronted with a compound of the instinct and what has been learned
in its previous expressions, or what might these days be called an
interactional variable.
Had James been heeded, there might not have been person-
situation debates in the ensuing years. For that matter, the inconclu-
siveness of the earlier forms of the debate might have precluded our
most recent version of it. Far from giving up criticality of mind, we
might have used that capability on issues that can be resolved. That
unresolvable issues can generate so much excitement and effort sug-
gests just how deeply entrenched in our discipline is "simple and sover-
eign theorizing." We have a tendency to want to win wars even more
than to deepen understanding of behavior.
Indeed, there is a debate accelerating right now that bears resem-
blance to the person-situation debates of the past. It involves the be-
havioral geneticists and the social developmentalists. The attempt to
demonstrate that behavior is largely or totally explained by genetic
factors is gaining considerable impetus. The empirical argument hinges
on the greater similarity in behavior of genetically related children
reared apart than of genetically unrelated children reared together. In
the hands of ~orne investigators, such observations lead to the conclu-
sion that rearing is unimportant by comparison with genetics. The re-
joinder, articulated well by Lois Hoffman (personal communication) is
that behavioral similarity is not a good gauge of socialization if family
units treat different children differently, as they undoubtedly do. It is
common in families for one child to be treated as the leader and an-
other as the scapegoat, for female children to be treated differently
from male children, and for birth order to involve different parental
behaviors. Thus, it is self-serving for the behavioral geneticists to choose
behavioral similarity as a measure of socialization effects. And so the
debate goes, inconclusively, generating more competition than clarity,
as William James warned us.
Nor is the tendency toward warring factions limited to specifically
articulated debates. More general than a debate is the continuing an-
tagonism between academic psychologists and practitioners. Many aca-
demicians see themselves as the only true scientists, because they
theorize, do research, and teach. Accordingly, they consider the practi-
tioners, whether they be clinical or industrial in their efforts, as some-
where between hacks and hucksters. On their part, many practitioners

see themselves as the only really adult psychologists, because they deal
with real problems of real people. Accordingly, they consider the acade-
micians to be irrelevant, romantic, trivial, and weak. It is quite rare for
a psychologist to believe that academic and practical functions are two
sides of the same scientific coin. Each needs the other, the academic to
remain grounded in relevancy and the practitioner to resist mindless-
ness as a reaction to everyday pressures. Needless to say, there are
competent and incompetent psychologists in both academics and prac-
tice, but this proves nothing about science or relevancy.
Indeed, within the academy there exists the same war over who
and what in psychology is the more scientific. The biologically and cog-
nitively oriented psychologists generally regard themselves as scien-
tific insiders, with the others being somewhere between tender-minded
and softheaded. In contrast, the social, personality, and developmental
psychologists see themselves as truly understanding what is important in
the discipline, with the others being somewhere between sterile and silly.
Let me not belabor the point. Virtually wherever one turns in our
discipline there are waning factions, and this is a sign of our fractionation.
THE PREVALENCE OF MIDDLE-LEVEL THEORIZING
For some time now, academic psychology has been dominated by

middle-level theorizing. This approach avoids elaborate assumptive sys-
tems concerning matters far removed from immediate observation and
makes no pretense of comprehensiveness concerning human behavior
(Conant, 1947; Merton, 1957, 1968). A definite outcome of commitment
to middle-level theorizing is that many discrete hypotheses regarding
the same or similar phenomena are likely to exist concurrently in an
area. This is because each investigator may well start with just one
assumption (e.g., Durkheim's famous hypothesis that suicide rate is a
function of the normlessness of a society), applying it in one or more
concrete contexts by procedural elaborations that seem dictated merely
by methodological concerns or common sense. One positive feature of
having many discrete hypotheses is that a lively empirical competition
ensues.
But the danger inherent in middle-level theorizing might be called
a vanity of small differences or the entrenchment of several hypothe-
ses, the differences between which would appear small and unimport-
ant if only a broader purview were taken. A concerted attempt to theorize
comprehensively might collapse the hypotheses together, on the grounds
of their being special cases of each other or differing expressions of the
same underlying process. But the distrust of comprehensive theorizing

precludes such an attempt.
There are too many contemporary research areas in danger of such
a vanity of small differences to be included here. One that may suffice
as an example is intrinsic motivation. The basic phenomenon studied is
the paradoxical decrease in performance of a preferred activity that
takes place when an external reward is introduced. Many investigators
have proposed explanations of this phenomenon. Deci (1971) suggested
a cognitive evaluation hypothesis emphasizing that if a reinforcer is
perceived as controlling behavior from an external source, then intrin-
sic motivation for performing is reduced. Kruglanski (1975) proposed
endogenous versus exogenous attribution, similarly focusing on the rea-
sons for instead of the causes of behavior. Then there is the over-
justification hypothesis of Lepper and Greene (1975), stressing how
behavior originally justified by intrinsic motivation becomes over-
justified with the addition of extrinsic reinforcement. Reiss and Sushin-
sky (1975) offered a competing response hypothesis, suggesting that
subjects will be less interested in playful activities to the extent that
extrinsic rewards elicit responses that interfere with play responses.
There is even a delay of gratification hypothesis (Ross et al., 1976) that
focuses on the inhibiting effects of delay between promise and receipt
of reward.
The empirical studies done in furtherance of these various hypoth-
eses are intriguing and sophisticated. Each hypothesis has its own kind
of empirical support. But, as DeCharms and Muir (1978) conclude in
their review, the great bustle of activity in this field has led to little
consensus or accumulated sense of understanding, and the problem may
be the insufficient theoretical elaboration of the many proposed expla-
nations. The hypotheses remain rich in implicit assumptions. I suspect
that at the underlying level of implicit assumptions, there is consider-
able overlap if not identity among these hypotheses. But our commit-
ment to middle level theorizing precludes exploration of hypothesis
overlap and contributes to the fragmentation of psychology.
OVERSPECIALIZATION
So specialized have the subfields of psychology become that there
is little or no sense of common cause or even communication across
them. Odd though it may be, there is virtually nothing tying together
the subfields of cognition, personality, learning, development, social
psychology, biopsychology, industrial-organizational psychology, and
clinical psychology. There are separate journals with separate reader-
ships, simultaneous sessions in different hotels at annual conventions,
misunderstandings and antagonisms in the conduct of academic depart-
mental affairs. In some universities, biologically oriented psychologists
belong to different divisions or schools than do socially oriented and
clinically oriented psychologists. In other universities there are several
departments that include psychologists. Introductory psychology courses
and textbooks have become increasingly more fragmented and unwieldy.
Indeed, at some universities the traditional introductory psychology
course has given way to several alternative courses, each emphasizing
some but not all aspects of the field. So fragmented is our field that it
has become a Tower of Babel in which many no longer feel that there is
any underlying discipline.
Perhaps this fragmentation is one reason why psychology is not a
central feature of university organization. Although it is a popular un-
dergraduate major, psychology appears rather peripheral to academic
structure and definition. Nor on an individual basis do many psycholo-
gists tend to play central intellectual roles among their academic col-
leagues from other disciplines. Although some psychologists are now
beginning to fill administrative roles at universities, this activity does
not seem to extend to influencing their colleagues in scholarly ways.
This may well be because psychologists have become so specialized that
they lack any overall format and agenda that can be persuasive to
scholars in other areas.
Certainly, the overspecialization renders it difficult for academic
psychologists to address such major problems of a psychosocial nature
as poverty, injustice, criminality, physical and mental illness, character
development, homelessness, and the like. The research problems that
fill our journals tend to be narrower and more esoteric. And there is
the alarming tendency to retreat even further from content than that
and define our science in terms of methodological prowess. Even in
reviewing each others' manuscripts for publication, we are more likely
to critique the methods than the content and conclusions. An aspect of
this is the tendency to avoid topics for study that appear too complex
to be researched easily. It is as if a retreat to methodology is the only
thing holding our fragmented field together.
PERSONALITY THEORIZING AS ONE ANTIDOTE
One way to overcome the fragmentation in our field is for psychol-

ogists to spend time and imagination considering entire persons in their
social milieu. This will help to blend together the various bits and pieces of
knowledge currently existing in unintegrated form. Henry Murray

(1938) dubbed this holistic study of humans, personology, and personal-
ity theorizing is its conceptual aspect.
TASKS AND COMPONENTS OF PERSONALITY THEORIZING
In order to appreciate the integrating effect on psychology that

personality theorizing could have, it would be well to consider what are
the tasks in and components of personality theorizing. Elsewhere
(Maddi, 1968/1988) I have contended that personality theories need to
include core, peripheral, and developmental statements.
In the core statement, theorists make assumptions about what binds
all human beings together, what they all share. Typically, these assump-
tions refer to unlearned, inborn aspects of motivation, temperament,
and potentialities. Implicitly or explicitly, the core statement concerns
the overall purpose of human life. For some personality theories, the
core statement is heavily biological. Freud (1925), for example, as-
sumed that the mental preoccupation with sexual and aggressive in-
stincts mirror the organismic facts of metabolism. Other personality
theories are less concerned with biology, such as Allport's (1955) core
assumption that humans all try to achieve propriate (self-expression
and self-determination) functioning. Although the aim of the core state-
ment is to identify how we are all alike, some personality theories
record that, within this, inborn individual differences exist. For exam-
ple, Rogers (1959) assumes that although we are all striving to actual-
ize our inherent potentialities, we are radically differentiated by these
potentialities.
The peripheral statement in a personality theory conceptualizes
the various life-styles that characterize the thoughts, feelings, and
actions of adults (Maddi, 1968/1988). The life styles are typically com-
posed of types (e.g., the Freudian oral, anal, phallic, and genital charac-
ter types) consisting of related traits, motives, or defenses (e.g.,
dependency, dominance, gregariousness, need for achievement, reaction
formation). The types, traits, and defenses proposed in a peripheral
statement refer to motivational and expressive characteristics that are
learned, rather than inborn. The peripheral statement is the way in
which the theorist explains individual differences in behavior. Although
past critiques (e.g., Bern, 1972; Mischel, 1968) have insisted that periph-
eral theorizing does not give sufficient weight to situational determi-
nants of behavior, personality theories (e.g., Allport, 1937; McClelland,
1951; Murray, 1938) have often defined peripheral characteristics as
having joint situational and personal instigation. In any event, the
current interactional accord (Ekehammer, 1974; Endler & Magnussen,

1976) indicates the importance of including both sources of behavior in
peripheral theorizing.
In its developmental statement, a personality theory conceptual-
izes how expressions of the core we all share lead to the individual
differences in life-style constituting the periphery (Maddi, 1968/1988).
Involved here is a consideration of how learning takes place. Typically,
personality theories consider child-parent interactions to be formative
in the development of life-style. Some personality theories also consider
individual-society interactions as important as well. The latter theories
tend to assume that personality development continues throughout life,
whereas the former consider life-style to be the product of learning in
the childhood years.
A touchstone to personality theorizing is a sense that there are
behavioral phenomena that can be explained that way (Maddi, 1968/1988).
Without this, there would be little reason to theorize at all. An implica-
tion of the interest in explaining certain phenomena is the development
or adoption of a data language that will specify the units of analysis
that will frame the relevant dependent variables. It is true that the
agenda of personology is so broad-to understand all the regularities
that lend pattern to human behavior-that it has been difficult to be
very concrete about a data language. Nonetheless, some theorists, no-
tably Murray (1938), have attempted to start on this problem. Others
(e.g., McClelland, 1981) have implied a data language in their dedication
to explaining specific behaviors and their covariation.
Added to the components of a personality theory just outlined is
an evaluative concern for explanatorily differentiating ideal from com-
mon and from psychopathological behaviors that is so frequent among
theorists as to be virtually universal (Maddi, 1968/1988). The life-style
conceptualized as ideal is considered virtually invulnerable to stressors
and maximally expressive of the core (or universal human purpose). In
contrast, the other, more common life-styles are considered to permit
less full expression of core characteristics and this defines their vul-
nerability to stressors. The conceptualized breakdown of common life-
styles in the face of stressors is the typical way in which personality
theories explain psychopathological states. The evaluation of life-styles
as ideal or nonideal carries with it a related evaluation of the develop-
mental patterns that lead to them as ideal or nonideal. All of this con-
stitutes a springboard to a conceptualization of psychotherapy, as that
which can help persons overcome breakdowns by substituting for the
nonideal developmental pattern they experienced previously, a facsim-
ile of the ideal developmental pattern.
INTEGRATING POTENTIAL OF PERSONALITY THEORIZING
Even in this brief outline of the tasks in and components of per-

sonality theorizing, it is easy to discern a formidable integrating po-
tential. Properly done, personality theorizing needs input from such
areas as learning, cognition, physiological, social, measurement, devel-
opmental, and abnormal psychology. Through its integrative format and
goals, personality theorizing might also make contributions back to these
areas. This contribution to other areas might well be enhanced by the
expression of personality theorizing in relevant personality research.
With a few exceptions here and there, however, this bright hope has
not been realized, and fragmentation and divisiveness reigns in psychology.
One reason why the integrative potential of personality theorizing
has not been realized is the result of the very fragmentation and divi-
siveness itself. When other psychologists shun personality theorizing as
unscientific, personologists too often respond by withdrawal into their
own scholarly agenda, essentially giving up on their colleagues as
trivializers. In the academic world, these personologists often find com-
mon cause with more receptive scholars in other fields. The advantage
in this is some sense of intellectual community for personologists. The
disadvantage is that there is no curb on fragmentation and divisiveness
in psychology. My advice to personologists is to use the sense of worth
they get from their own work and its acceptance by scholars in other
disciplines to continue the struggle to have an integrative effect on their
fellow psychologists in other areas. After all, personologists did bother to
enter psychology rather than some other discipline. This implies or even
defines their commitment to that field, and could help them to overcome
any bitterness they might feel at being rejected as unscientific.
Another reason why personality theorizing has not reached its in-
tegrative potential in psychology is more internal to personology. Even
within personology, there is a kind of fragmentation and divisiveness
that takes place. Although each personality theory may well express
the breadth and depth that integrates information from other areas of
psychology, personologists tend to become committed to one of these
theories to the exclusion of the others. Thus, personologists are Freud-
ians, or Adlerians, or Jungians, or Rogerians, or existentialists and
have little time or inclination to encounter or be influenced by other
approaches. Indeed, this internal fragmentation is often augmented by
rivalry, as one's own approach comes to feel more and more familiar
and comfortable.
The antidote to this fragmentation within personology is compara-
tive analysis (Maddi, 1968/1988). The first step in a comparative analy-
sis of personality theorizing is to steep oneself in the various theories

that exist, trying to understand what the theorists were trying to ac-
complish in understanding behavior. There are, of course, many exist-
ing theories of personality, even if one restricts oneself to just those
known to psychologists.
The second step of comparative analysis involves classifying the
theories into the smallest number of models that appear to do justice to
the aims and approaches of the theories. Here, it is not enough to
consider just parts of the theories, as has been done in the common
classification of cognitive, trait, behavioristic, and dynamic theories. For
example, too much is left out when one classifies Allport as a trait
theorist (as he was at least as much cognitive and behavioristic) or
Freud as a dynamic theorist (as he heavily emphasized cognition and
trait considerations as well). These common classifications get imposed
on personality theorying primarily because they reflect the preoccupa-
tions of other fields of psychology. But to persist in this is to miss the
integrative potential of personality theorizing by reducing it to what
will minimally interfere with other areas of psychology. A more instruc-
tive classification of personality theories will simultaneously take into
account their core, developmental, and peripheral statements. This is
admittedly hard to do, and we will probably go through several classi-
ficatory efforts before one feels acceptable to most personologists.
My own efforts at this second step of comparative analysis have
led to three models, each with two variants. I have called them conflict
(psychosocial and intrapsychic) fulfillment (actualization and perfec-
tion), and consistency (cognitive dissonance and activation) models
(Maddi, 1968/1988). A theory fits into the conflict category if it postu-
lates at the core level two great forces, present at birth, unchangeable,
and in opposition to each other; at the peripheral level a life-style that
is considered ideal because it is a compromise between the opposing
great forces, whereas the nonideal life-styles fail to achieve a compro-
mise; and a developmental statement that defines the ideal interaction
between youngsters and parents as a razor's edge, not too permissive
and not too punishing. Such theories express the psychosocial variant if
one great force is presumed to arise in the individual while the other
arises in groups or societies and the intrapsychic variant if both great
forces are considered to be present in the individual without any neces-
sary consideration of groups or societies. The prototype of the psy-
chosocial conflict model is Freud's (1933) theory and of the intrapsychic
conflict model is Jung's (1933).
In contrast, a theory expresses the fulfillment model if it postu-
lates at the core level only one great force that, because inherent, is
unchangeable; at the peripheral level a life-style that is considered ideal in

that it expresses the great force more fully than the other, nonideal
life-styles; and a developmental statement that emphasizes as ideal in-
teractions between youngster and parents that facilitate expressions of
the great force. In this model, psychosocial conflict is recognized as a
possible occurrence. But when it occurs, society is regarded as having
failed the individual. Such conflicts are considered resolvable and indi-
cate the need for social reform. A theory expresses the actualization
variant if the great force involves ever greater expressions of pre-
sumed inherent potentialities and the perfection variant if it involves
striving to reach behavior patterns regarded as ideal though not pro-
grammed in potentialities as such. The prototype of the actualization
fulfillment model is Rogers's (1959) theory and of the perfection fulfill-
ment model is existential psychology (Maddi, 1970; May, 1958).
Finally, the third, or consistency, model involves at the core level a
negative feedback principle in which there is discomfort with and ef-
forts to decrease discrepancies between some personal norm and actual
occurrences; and at the peripheral and developmental levels the reten-
tion of behaviors that have proven successful in reducing such discrep-
ancies. In the cognitive dissonance variant, the norm is an expectation
or prediction, whereas in the activation variant the norm is a custom-
ary level of tension or arousal. Prototypic of the cognitive dissonance
variant is the theory of Kelly (1955) and of the activation variant is that
of Maddi (1968/1988).
Thus, the end result of the second step of comparative analysis is
to highlight the smallest number of categories or models that char-
acterize the existing personality theories. Then the third step is to
pinpoint the issues that arise from the disagreements between these
categories or models (Maddi, 1968/1988). These issues can be primarily
at the core, peripheral, or developmental levels. For example, a largely
core issue is whether all, some, or no behavior is defensive. A seminal
issue that arises at the developmental level is whether or not there is
radical change in behavior after the childhood years are past. Issues at
the peripheral level tend to involve differences in the content of life-
styles that are to be regarded as ideal.
The formulation of issues leads readily into the fourth step of com-
parative analysis, which is to resolve the issues. In an empirical field
like psychology, resolution will be sought through research. To be truly
relevant, the research should employ a design that permits the favor-
ing of one or the other side of the issue, depending on what results are
obtained. Such comparative analytic research is most productive when
the design to be employed has been carefully planned so that obtained
results can convincingly favor theories on one side of the issue while
damaging those on the other. Through a concerted effort at such re-
search, the issues separating models could well be resolved. Once it is
clear which model is the most promising, psychologists could either
endorse the personality theory or theories well expressing this model
or build a new theory that explicitly capitalizes on what has been learned
in the comparative analytic research.
Whether or not you find the particular models I have proposed
convincing (they are certainly not the only ones it is possible to ab-
stract from existing personality theories), I hope you will appreciate
the enormous integrative potential in the comparative analytic approach.
At every step of this approach (and especially in the posing of issues
and their resolution), individual personologists would be well advised to
consult with their colleagues who hold differing theoretical perspec-
tives. There would be, through this task-oriented communication, pres-
sure on all personologists to be clearer and more definite about their
theoretical views and the research implications in them. Engaging in
comparative analysis would go a long way toward obviating fragmenta-
tion within personology. It would also facilitate the integrative effect of
personology on other areas of psychology in that many personologists
would have to use more explicitly and comprehensively information
available from these other areas in order to be specific enough to en-
gage in the collegial aspects of comparative analysis just outlined.
THE EMPIRICAL SPIRIT
If personologists adopted comparative analysis as their method of

choice, the personality field would be defined more by the issues need-
ing resolution than by one or another of the existing theoretical ap-
proaches. For this kind of commitment one needs an empirical spirit, a
willingness to change one's beliefs if convincing research compels that.
This does not mean, however, that a personologist would have to or
should lose his or her belief in one personality theory as probably bet-
ter than the others. Loss of all belief pending the outcome of some
future comparative analysis comes perilously close to lack of commit-
ment or alienation. Personologists need to struggle simultaneously to
find or develop the personality theory that seems convincing to them
and be willing to subject that theory to the empirical rigors of compar-
ative analytic test. Without the former commitment, the hard work of
theory refinement and development will languish. Without the latter
commitment, the pressures toward fragmentation will stifle progress
toward true understanding of human behavior.
TEACHING PERSONALITY
At present, it is most common for psychology curricula to include

one advanced undergraduate course on personality and for there to be
one or more courses on aspects of this subject matter at the graduate
level. Typically, these days, the undergraduate course either empha-
sizes personality research rather than theory or considers a major the-
ory or theories as examples of what is available. The same pattern
appears at the graduate level, though one is more likely to find there
the research emphasis rather than the theoretical emphasis.
Although the research emphasis in personality teaching expresses
the scientific tradition and aspiration of psychology, it has a serious
drawback in terms of the position I have taken in this chapter. Scrutiny
of the research appearing in even the most prestigious personality jour-
nals illuminates this drawback by showing that rarely do the papers
articulate, let alone derive their concerns from, anything approaching
comprehensive theories of personality. As personality research is done
these days, it involves, at most, middle-level theorizing (Merton, 1957,
1968) in which hypotheses to be tested or questions to be answered are
derived from either common sense or other studies similarly conceived.
Because there is little comprehensive theorizing standing behind
personality research, courses emphasizing it are not particularly useful
in overcoming fragmentation. At most, the research studies cluster in
areas of human behavior, such as creativity or aggressiveness, and the
course based on them goes from one such area to another. Nowhere is
the student encouraged to put the areas together into an overall sense
of human behavior. Nowhere is the student stimulated to consider core,
peripheral, and developmental concerns in an integrated fashion. This
approach to personality teaching recapitulates the fragmentary approach
currently seen in the teaching of other areas of our field, notably intro-
ductory, social, and biological psychology. Thus, to teach personality by
restricting coverage to research studies alone is to fail to alert new
generations of budding psychologists to the value of integrative effort.
We will be passing on to them a fragmented, divisive field as if that
were some sort of ideal.
By contrast, courses covering major personality theories at least
encourage the student to consider core, peripheral, and developmental
assumptions in overall attempts to understand human behavior. But
such courses come with dangers of their own. The major pitfall of courses
taught from one theoretical perspective to the exclusion of others is
what I have elsewhere (Maddi, 1968/1988) called partisan zealotry. In
his or her zealous championing of one personality theory, the teacher
may insulate the student (especially if this is the only personality course
in the curriculum) from the possibility of picking and choosing among
various available alternatives. This pitfall is minimized if the teacher
goes out of his or her way to alert students to alternative views though
they are not emphasized in the course. It is especially useful if there
are also courses championing some of these alternative personality the-
ories in the curriculum.
That leaves the type of course that surveys personality theories
rather than restricting itself to a favored one. Done well, such a course
can certainly alert the student to the comprehensive, integrating func-
tion of personality theorizing. The major pitfall of this approach inheres
in its survey format and might be called benevolent eclecticism (Maddi,
1968/1988) or an uncritical reiteration of the same old theories in the
same old ways. Too often, students get through such courses by memo-
rizing madly immediately before the examinations and promptly forget-
ting everything as soon as the evaluation is over.
It will by now come as no surprise to you that I advocate a com-
parative analytic approach in personality teaching (Maddi, 1968/1988).
With such an approach, the student experiences the comprehensive,
integrative format of personality theorizing, learns the content of the
major existing theories, experiences an intellectual framework within
which to classify and evaluate theories, ponders some of the issues
characterizing the personality field, and sees examples of how research
can hope to resolve those issues. Should students taking such a course
actually become psychologists, they are more likely to approach their
field with integrative standards and goals, and value in themselves and
others an empirical spirit that transcends but does not replace their
personal convictions. Especially because it is hard for those psycholo-
gists who have long conducted their careers in more fragmentary and
divisive fashion to change that, it is important to educate the succeed-
ing generations of psychologists differently. And comparative analysis
is a useful teaching tool in this effort.
A FINAL WORD
Although I started with an analysis of personality theorizing, I

have ended with an encouragement to teach the existing personality
theories. To some, these theories, though perhaps having a kind of
grandeur, are a bit old-fashioned and outmoded. Indeed, Jerome Singer
even characterized textbooks surveying these theories as a kind of so-
journ through the graveyard. To be sure, his critique was largely about
the uncritical nature of such surveys. But he was also objecting in part
to emphasizing these old theories.
To my mind there is no reason to either cherish or reject the com-
prehensive personality theories merely because they exist or may have
been formulated some time ago. If they are the subject matter of a
comparative analysis, we will see soon enough where they fit into mod-
els for understanding human behavior and how their model fares in
research-based issue resolution. Then there will be a reason to retain,
change, or reject existing personality theories that is based in intellec-
tual concerns rather than fashion. After all, if the concrete personality
theories that exist today had not been formulated already, there is
every reason to believe that personologists would invent them soon.
There is a small number of models for understanding human behavior
and a finite number of examples of these models that human imagina-
tion can construct.
REFERENCES
Alker, H. A. (1972). Is personality situationally specific or intrapsychically consistent?
Jaurnal of Personality, 40, 4-16.
Allport, G. W. (1955). Becoming: Basic considerations for a psychology of personality.
New Haven: Yale University Press.
Bern, D. J. (1972). Constructing cross-situational consistencies in behavior: Some thoughts
on Alker's critique of Mischel. Journal of Personality, 40, 17-26.
Bowers, K. S. (19730. Situationism in psychology: An analysis and a critique. Psycholog-
ical Review, 80, 307-336.
Conant, J. B. (1947). On understanding science. New Haven: Yale University Press.
Deci, E. L. (1971). The effects of externally mediated rewards on intrinsic motivation.
Jaurnal of Personality and Social Psychology, 18, 105-115.
DeCharms, R. C., & Muir, M. S. (1978). Motivation: Social approaches. In M. R. Rosen-
zweig & W. Porter (Eds.), Annual review of psychology (pp. 000--000). Palo Alto,
CA: Annual Reviews.
Ekehammer, B. (1974). lnteractionism in personality from a historical perspective. Psy-
chological Bulletin, 81, 1026-1048.
Endler, N. S., & Magnussen, D. (1976). Toward an interactional psychology of personality.
Psychological Bulletin, 83, 956-974.
Epstein, S. (1984). The stability of behavior across time and situations. In R. A. Zucker,
J. Aronoff, & R. I. Rabin (Eds.), Personality and the prediction of behavior. New
York: Academic Press.
Fiske, D. W. (1974). The limits for the conventional science of personality. Jaurnal of
Freud, S. (1925). Instincts and their vicissitudes. In S. Freud, Collected papers (Vol. 4).
London: Institute for Psychoanalysis and Hogarth Press.
Freud, S. (1933). New introductory lectures in psychoanalysis (W.J.H. Sprott, trans.).
New York: Norton.
Jung, C. G. (1933). Modern man in search of a soul. New York: Harcourt, Brace &
World.
Kelly, G. A. (1955). The psychology of personal constructs (Vol. I). New York: Norton.
Kruglanski, A. W. (1975). The endogenous-exogenous partition in attribution theory.
Psychological Review, 82, 387-406.
Lepper, M. R., & Greene, D. (1975). Turning play into work: Effects of adult surveillance
and extrinsic rewards on children's intrinsic motivation. Journal of Personality and
Maddi, S. R. (1970). The search for meaning. In M. Page (Ed.), Nebraska symposium on
motivation (pp. 000-000). Lincoln: University of Nebraska Press.
Maddi, S. R. (1988). Personality theories: A comparative analysis (5th ed.). Homewood,
IL: Dorsey. (Original publication 1968).
May, R. (1958). Contributions of existential psychotherapy. In R. May, E. Angel, & H. F.
Ellenberger (Eds.), Existence: A new dimension in psychiatry and psychology
(pp. 000-000). New York: Basic Books.
McClelland, D. C. (1951). Personality. New York: Dryden.
McClelland, D. C. (1981). Is personality consistent? In A. I. Rabin, J. Aronoff, A. M.
Barclay, & R. A. Zucker (Eds.), Further explorations in personality (pp. 87-113).
New York: Wiley.
Merton, R.K. (1957). The role-set: Problems in sociological theory. The British Journal of
Sociology, 8, 106-120.
Merton, R. K. (1968). Social theory and social structure. New York: Free Press.
Mischel, W. (1968). Personality and assessment. New York: Wiley.
Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality.
Psychological Review, 80, 252-283.
Murray, H. A. (1938). Explorations in personality: A clinical and experimental study of
fifty men of college age. New York: Oxford University Press.
Reiss, S., & Sushinsky, L. W. (1975). Overjustification, competing responses, and the
acquisition of intrinsic interest. Journal of Personality and Social Psychology, 31,
116-125.
Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal relationships, as
developed in the client-centered framework. In S. Koch (Ed.), Psychology: A study
of a science (Vol. 3, pp. 000-000). New York: McGraw-Hill.
Ross, M., Karniol, R., & Rothstein, M. (1976). Reward contingency and intrinsic motiva-
tion in children: A test of the delay of gratification hypothesis. Journal of Personal-
ity and Social Psychology, 33, 442-447.
CHAPTER SEVEN
It's Time to Put Theories of

Personality in Their Place, or,
Allport and Stagner Got It Right,
Why Can't We?
GERALD A. MENDELSOHN
Among the various subfields of psychology, Personality is just about

unique in its continued dedication to the teaching of courses and thus to
the writing of textbooks devoted to "theory." Where else, save perhaps
in the closely allied field of Abnormal Psychology, can one find com-
parable courses and texts? Surely not in Cognitive or Biological Psy-
chology, for example, and only with difficulty in Social Psychology. The
contrast between the fates of Hilgard's (1948) classic Theories of
Learning and Hall and Lindzey's (1957) comparably classic Theories of
Personality is telling in this regard. The former, whose last edition was
written a decade ago, can scarcely be considered representative of the
current generation of textbooks in learning while the latter, in its most
recent incarnation (Hall et al., 1985) (and with two additional authors),
remains a leading textbook in the field. In what follows, I will be less
concerned with the question of why this state of affairs exists than
with the question of whether it is a desirable state of affairs. It should
come as no surprise after so tendentious an opening that I intend to
GERALD A. MENDELSOHN • Institute of Personality and Social Research, University of

California at Berkeley, Berkeley, California 94720.
New York, 1993.
103
104 GERALD A. MENDELSOHN
argue that the continued emphasis on theories of personality is not

desirable, that it is both anachronistic and a misleading representation
of what, in fact, personality psychologists think about and do.
THE DISTINCTION BETWEEN THEORIZING ABOUT

PERSONALITY AND THEORIES OF PERSONALITY
Before I develop the argument whose conclusion I have just stated,

two clarifications are in order. First, a distinction between "Theories of
Personality" and theorizing about personality needs to be made. By the
latter I mean the kind of conceptual analysis that both precedes and
follows empirical work. Such theorizing is essential and, if one accepts
the proposition that there are no facts apart from theories, inevitable.
We have, in fact, been passing through a particularly vigorous period of
such theoretical activity, most obviously (but not only) in the debate
about the meaning and status of "traits" and situations. In the course of
that debate, any number of basic assumptions about personality, includ-
ing the legitimacy of the concept of personality itself, came under in-
tense scrutiny. If, as I believe, the debate is now largely resolved, that
is because of the understandings that it engendered or revived; under-
standings about, for example, the meaning of traits and "consistency,"
the compositing of observations and the interaction between persons
and situations. Thanks to this kind of conceptualizing, personality is
now a healthier field than it has been for years. And when we add in
such current developments as the revived interest in long-term motiva-
tional sequences and in the self, public and private, the elaboration of
the concepts of social intelligence and ego development, the increase in
taxonomic activity, the study of the dynamics of stress and coping, and
the attempt to comprehend the significance of evolutionary biology for
human behavior, it becomes apparent that personality theory is by no
means languishing in neglect.
In contrast, I mean the term "Theories of Personality" to apply to
those ambitious, broadly encompassing and universalistic statements
that claim to identify the basic elements and processes underlying in-
dividual psychological activity. These theories are generally identified
with a particular person (e.g., Freud, Kelly, Rogers) and by a label
written in capital letters (e.g., Analytic Psychology, Personal Construct
Theory, Social Learning Theory). More often than not, they originate
in clinical observation or armchair reflection and are closely tied to
psychotherapeutic considerations. To be sure, there have been many
efforts to "validate" several of these theories, or rather, particular ele-
PUTTING PERSONALITY THEORIES IN THEIR PLACE 105
ments of these theories, by empirical procedures, but it is essentially to

the speculations, insights, and persuasiveness of the individual theo-
rists that they owe their existence. Theories of Personality are, in sum,
at an altogether higher level of abstraction and at a far greater remove
from empirical research than what I previously described as "theoriz-
ing about personality."
Judging from the 15 textbooks (all with publication dates of 1984
or later) that I have before me as I write, there is good agreement
about the theorists to whom students should be introduced. The list of
the top 20 runs as follows: Adler, Allport, Bandura, Cattell, Dollard
(and Miller), Erikson, Eysenck, Freud, Fromm, Horney, Jung, Kelly,
Maslow, Mischel, Murray, Rogers, Rotter, Skinner, and Sullivan. Though
they form the core group, not all of them appear in every text; Maddi
(1989), in his distinguished comparative analysis of personality theories,
for example, excludes Sullivan and Horney "because their thinking no
longer seems to have great direct impact on the personality field" (p. 14).
A number of theorists appear with less frequency than the top 20 but
often enough to warrant mention: Guilford, Hartmann, Kohut, Lewin,
May, McClelland, Rank, Sheldon, and White. Finally, cameo appear-
ances are made by Angyal, Berne, Binswanger, Boss, Ellis, A. Freud,
Goldstein, Maddi, and Perls. A few other names might justifiably be
added to the list, but this is, I believe, a reasonably inclusive roster of
the writers who define the canon of Theories of Personality as it can be
inferred from contemporary textbooks.*
THE GENRES OF TEXTBOOKS IN PERSONALITY
The second clarification has to do with the content of textbooks in

Personality. I cannot claim to have done a systematic survey of pub-
lisher's lists, but it seems clear that the prototypic textbook is organ-
ized to a considerable degree around the presentation of Theories of
Personality. There are, of course, exceptions; in fact three genres of
textbooks can be distinguished. First, there are those that are explic-
itly designed to provide coverage of the positions of the theorists deemed
most significant by the author(s). Hall and Lindzey is the locus classi-
cus of this genre-after the obligatory introductory material, a dozen
or so chapters devoted to a theorist or to a group of related theorists,
*There are some mysteries about who is included and who is not in textbooks of person-
ality. For example, in what meaningful sense can Skinner be considered a personality
theorist and why has Gardner Murphy so completely disappeared from view?
followed by some comparative or summarizing comments. Other for-

mats can be found in which theorists who presumably share a general
orientation are grouped into families that may be variously labeled
strategies, approaches, perspectives, and so forth. Indeed, this seems
to be the predominant trend recently. The cast of characters is the
same, however, and the essential substance of the volume remains the
exposition of their views. I say the "essential substance" because it is
usual to include as well material on some or all of the following: rele-
vant assessment techniques and therapeutic approaches, examples of
pertinent research, critical evaluations, and intertheorist comparisons.
The selection of this additional material is, of course, a function of the
particular theorists chosen for inclusion in the text; it does not result
from an effort to provide representative or systematic coverage of the
issues of concern to contemporary personality psychologists.
The second genre of textbook is organized around topics and issues
in personality rather than Theories of Personality. Stagner's Psychol-
ogy of Personality (1974) provides a good example of the genre. In it
one finds chapters on methods, development, units of analysis (e.g.,
traits, styles), dynamics, and determinants (e.g., biological and socioeco-
nomic factors). Many of the familiar theorists make an appearance, but
briefly and only inasmuch as they have relevance to the topical organi-
zation of the book. Examples of this genre vary in the balance they
strike among conceptual analysis, methodological concerns, and presen-
tations of specific areas of empirical research, but they all seek to pro-
vide an introduction to and overview of the subject matter of the field
of personality. It may perhaps be unnecessary to point out that the
"subject matter of the field of personality'' is by no means coextensive
with the subject matter addressed in the work of the canonical Theo-
rists of Personality.
Finally, there is a third genre of textbooks that is a hybrid of the
two just discussed. Textbooks of this kind are organized into two major
sections, the first devoted to the expositions of the standard Theories
of Personality and the second to reviews of specific areas of empirical
investigation. The areas chosen for inclusion naturally vary from text
to text; some typical choices are aggression, sex roles and gender iden-
tity, stress and coping, self-concepts and so on, though in some cases
the topics are broader (e.g., personality development). Inclusiveness is
the hybrid's advantage-there is something for everybody-but the in-
clusiveness inevitably leads to less depth and detailing than one finds
in the best examples of the other two genres. A more serious limitation
of the hybrids is that the two major sections are rarely integrated in
more than a superficial way. That, too, is inevitable given the casual
relationship that exists in actuality between Theories of Personality

and research in personality. I shall elaborate on this point shortly.
It is perhaps ironic that even though the first two significant text-
books in personality, the textbooks that inspired this volume, were of
the second genre, the example of the founders has been followed fit-
fully at best. A number of empirically oriented texts have appeared
over the years, it is true, but they have not often run to multiple
editions and exemplars are not common today. Rather, the diet offered
to undergraduates has been and continues to be based on Theories of
Personality, however varied the preparation and the side dishes.
What conclusions can our students, and indirectly our colleagues,
be expected to draw from this way of presenting the field of personal-
ity? First, that the work of personality psychologists is inspired by and
organized around the content of the Theories of Personality; second,
that these theories have adequate validity and utility; and finally, that
they represent the current state of thinking in the field. If this were
not so, why would we in our primary vehicle of self-presentation, the
textbook, give pride of place to this content as opposed to some other?
But, in fact, none of these conclusions is correct, for Theories of Per-
sonality are at the periphery, not at the core, of the contemporary
field of personality as it is practiced by professionals and, I believe, for
good reason.
THE CRITIQUE OF THEORIES OF PERSONALITY
Now that I have made the necessary clarifications and staked out
a position, it is time to defend that position with more than sniping
comments. My brief will be in three parts. I will argue first that Theo-
ries of Personality are not really theories in any acceptable scientific
sense; second, that they are anachronistic; and finally, that they have
little to do with the research actually done by personality psychologists.
QUESTIONABLE SCIENTIFIC STANDING OF THE THEORIES
Although it would be most difficult to obtain agreement on pre-

cisely what qualities a scientific theory should have, the following statement
taken from Hall and Lindzey (1985) seems appropriate for present pur-
poses and reasonably consensual: ''We have defined personality theory
as a set of assumptions about human behavior, together with the empir-
ical definitions required, first, to move from the abstract statement of
the theory to the empirical observations and, second, to test and/or
support the theory'' (p. 17). Do the Theories of Personality that appear
in textbooks conform to this definition? It would be hard to argue that
they do. They are rarely systematic in their statement of basic assump-
tions or concepts and the relationships among them. Key concepts are
ill-defined and central issues of methodology and measurement are not
addressed. Alternative hypotheses are numerous but unexamined, ad
hoc and post hoc interpretations abound, and many, perhaps most, of
the propositions of these theories are neither confirmable nor discon-
firmable, that is, they are not capable of generating precise predictions.
Furthermore, the empirical and observational sources of these theories
are questionable on several grounds. Generally, they have been based
on small, special, and historically limited samples. Introspection and
armchair reflection have more often been their sources than data of a
less subjective character. When some sort of evidence is adduced, it is
frequently contaminated by the means of its collection (e.g., clinical
interactions) and it is frequently impossible to distinguish between the
data and the interpretation of the data. This makes replication and
intersubjective agreement well-nigh impossible. Thus, the systematic
collection and analysis of evidence has had little to do with the origin,
modification, acceptance, or fall from grace of the Theories of Person-
ality. Finally, I have never understood why Personality should place
such heavy emphasis on theories originating in the study of atypical
groups. The study of pathology frequently provides illuminating infor-
mation, but it seems to me more sensible to regard pathology as a
deviation from general patterns than to make inferences about general
patterns on the basis of the exceptional. As Maddi (1989) points out, "A
theory of psychotherapy is one thing; a theory of personality is an-
other'' (p. 617).
The set of objections I have just rattled off obviously apply to
some Theories of Personality more than to others, but none is immune
to criticism on several of these grounds. Does that mean that I regard
them as worthless? Not at all, for they often raise issues of substantial
psychological importance and sometimes provide insights about them.
The classical Theorists of Personality, however, can scarcely be consid-
ered the only thinkers to have struggled with issues of psychological
importance or to have provided insights about them. So, too, did the
Greek dramatists and philosophers, the medieval scholastics, the Prot-
estant reformers, the Elizabethan playwrights, the Encyclopedists, and
the Russian novelists, to note only a few groups of Western writers.
The question then becomes what, if anything, have the Theorists of
Personality identified as issues or as conceptions of those issues or as
approaches to the study of those issues that distinguishes them from
the many, many others who have observed and written about human
behavior? What is special about their work? Why in our teaching of
personality should we pay particular attention to them as sources of
ideas in preference to artists, philosophers, and theologians?
The answers to those questions ought to have something to do
with psychology as a scientific endeavor. If the Theorists of Personality
have particular significance for psychologists who seek to be empirical
scientists, it ought to be the case that they present their ideas in a
form that facilitates a scientific understanding of personality. But that
is precisely what, I have argued, they fail to do. I am not alone in that
judgment; the authors of textbooks by and large raise the same criti-
cisms as I have in their evaluations of the theories they present. Hall
and Lindzey (1985) for example, explicitly pose the question of whether
the personality theories they examine in their book conform to their
definition of a theory. Their answer: "Hardly" (p. 17).
What we have, then, are theories in name only. Levy (1970), after
coming to the same basic conclusion, suggests the term "conceptions"
as more appropriate. Perhaps the term "viewpoints" would do equally
well; but whatever term is chosen, the comment by historian Peter
Brown (1967) about religious systems seems no less applicable to Theo-
ries of Personality: "The quality of a religious system depends less on
the specific doctrine, than on the choice of problems that it regards as
important, the areas of human experience to which it directs attention"
(p. 393). That is a function of genuine importance-there is no reason to
doubt that the field would be the poorer had the canonical theorists
(among others) not transmitted their understandings and speculations
to us. There remains, however, a fundamental difference between "con-
cepts" of personality and "theories" of personality; to ignore that differ-
ence is not a service to our students or our field.
QUESTIONABLE CONTEMPORARY RELEVANCE OF THE THEORIES
The second part of my brief concerns the contemporary relevance

of the theories of personality. It is a striking fact that of the 38 theo-
rists cited earlier as forming the canon, 16 were born in the nineteenth
century and another 19 were born between 1900 and the end of World
War I. That leaves only three, Bandura (1925), Mischel (1930), and
Maddi (1933) whose careers began after World War II. It is also strik-
ing that the theorists who populate the textbooks of 1990 are largely
the same as those who populated the textbooks of 1960. Is it surprising
in these circumstances that many of our colleagues regard the field of
personality as stagnant?
Of course, it does not necessarily follow that because the Theories

of Personality are getting on in years, they should be put out to pas-
ture. My father used to say, "Because it's new doesn't mean it's good."
To that observation a corollary can be added, "Because it's old doesn't
mean it's bad." One could argue, that is, that the Theories of Personal-
ity have proven so durable because they are good theories. Given their
widely acknowledged deficiencies, however, that seems an unpromising
argument. And there are other grounds as well for doubting that the
merit these theories may once have possessed has remained undimin-
ished over the years. The time is long past when one could consider
theories, scientific or otherwise, as being isolated from the intellectual,
social, and cultural milieu of the time of their development. They are
embedded in history. This is particularly true of theories such as those
in the social sciences that are essentially speculative constructions in-
nocent of systematic empirical grounding. Thus, the bulk of the Theories
of Personality reflect the concerns of bygone times. A brief anecdote
may illustrate the point. During a discussion with a graduate student in
music, I made the obvious point that in his choice and handling of
dramatic themes, Wagner was Jungian. "No," the student corrected me,
"Jung was Wagnerian." He was, of course, correct; for budding Euro-
pean intellectuals of Jung's generation, Wagner was a force whose influ-
ence was felt well beyond the sphere of music (Barzun, 1938, p. 115).
Consider, too, Freud's now-rejected theorizing about the psychology of
women or how the category of sexual perversion has bit by bit been
emptied of content in the post-1960s era.
Now I realize that intellectual issues that have their origins in an
earlier period may still have considerable contemporary relevance; we
have not ceased to be concerned about unconscious processes, the self,
identity, and the like. What has changed, however, is the cultural, intel-
lectual, and scientific context in which we conceptualize and investigate
those issues. Moreover, new issues and paradigms, as well as new data
and methods that could not have been taken into account or even im-
agined by the canonical theorists have come to prominence since their
work was done. One can conceive a psychoanalysis shaped by a com-
puter model rather than a hydraulic model, but the fact is that the
death of Freud and the birth of computers were roughly contemporane-
ous. History, and with it science, marches on; it is the fate of theories
to be discarded, modified, and replaced. There is no good reason to
believe that theories of personality have qualities that make them any
more enduring than theories in other fields.
Lest I be misunderstood, I should make it clear that I am not
counseling ignorance of the past. Thinking about personality did not
begin in 1980 or 1960 or at the turn of the century; to understand

where we are now, it is essential to understand where we came from.
Thus, an awareness of Theories of Personality should be a part of the
training of personality psychologists at all levels. Indeed, in my own
teaching of personality I devote considerable time to psychoanalytic
thinking not only because the shape of the field has been so influenced
by Freud's work but also because he was so astute in identifying the
problems that any comprehensive approach to personality must address.
But if it is an error to be ignorant of the past, it is no less an error to
present the past as if it were the present. To do so is to encourage the
identification of the field of personality with the spirit and spirits of
yesteryear and to ignore or crowd out the vigorous theorizing and re-
search about personality that, as I noted earlier, is very much a part of
the contemporary scene. In sum, the more a textbook focuses on Theo-
ries of Personality, the more it conveys a distorted and deadening view
of the field.
QUESTIONABLE HEURISTIC VALUE OF THE THEORIES
Despite all that I have said so far, there is ano~her criterion for
evaluating theories that must be considered before concluding that the
canonical Theories of Personality are of historical interest only. This
criterion can be summed up as follows: "The bottom line in the evalua-
tion of personality theories is their heuristic value. However vague and
poorly developed a theory, if it can be shown to generate significant
research, the chances are that it has something important to say about
human behavior" (Hall & et al., 1985, pp. 19-20). While the second sen-
tence is perhaps arguable, the first embodies a view that appears in
one form or another in most textbooks. The implication of this view
seems clear enough: Theories of personality should play a major role in
the ongoing research activities of personality psychologists. Whether or
not they do is an empirical question, one that can be investigated by
examining representative samples of recent articles in the major jour-
nals of the field.
To be reasonably up to date and efficient in this research, I have
selected for study the most recent complete volumes of two major jour-
nals in the field, the Journal of Personality (Volume 57, 1989) and the
Personality Processes and Individual Differences section of the Jour-
nal of Personality and Social Psychology (also Volume 57, July-
December 1989). The two combined produced a sample of 81 articles in
which there is a total of 3799 references. First, I will report the good
news for partisans of Theories of Personality. Seventy percent of the
articles include at least one reference to one or more of the canonical

theorists; moreover, 26 of the 38 and 19 of the "top 20" theorists are
cited at least once. But as it says in a good book, "Data giveth and data
taketh away." A closer look at the evidence, it turns out, provides a less
encouraging picture. Of the 3799 references, only 208 are to a publica-
tion authored or coauthored by one of the 38 canonical theorists. Five
theorists, Allport, Bandura, Cattell, Eysenck, and Mischel account for
60% of all those references; that is, the remaining 33 theorists com-
bined are cited a mere 84 times (2.2% of the references) and only 11 of
them appear more than twice. It is important to add that the great
majority of the 208 references are to theorists still living in 1989-1%
of the references are to those no longer with us. It should be noted, too,
that cognitive social learning and psychometric-structuralist perspec-
tives are the only two that are represented in these journals with any
frequency at all. The similar findings of Mendelsohn in 1983 and Levy
in 1970 indicate that the neglect of Theories of Personality by research-
ers is not of recent origin.
If we move beyond counting to qualitative analysis, it becomes
apparent that very few of these references are more than phrases or
one-liners, a sort of courteous gesture to our forebears. Here are some
typical examples: "A third view is that establishing a sex-typed identity
is a major adaptive milestone of childhood and adolescence (Erikson, 1950;
Kohlberg, 1966; Mussen, 1969)"; "This tradition [psycho biography] may
be said to have been launched by Freud's (1910/1964) classic treatment
of Leonardo da Vinci (Elms, 1988), even though its major proponent
within personality psychology may have been Gordon Allport (1961),
the champion of the idiographic perspective (see also Erikson, 1959,
1969)"; and "Many longitudinal studies have been case studies of adults
offering detailed descriptions of individuals but little evidence of gener-
ality of results beyond the individual studied (e.g., Allport, 1965; White,
1966)." (These quotations are from Alpert-Gillis & Connell, 1989, p. 98;
Simonton, 1989, p. 696; and Moskowitz & Schwartzman, 1989, p. 724,
respectively; it should be noted in passing that they include a bit more
than 3% of the 208 references to the Theorists of Personality.) Clearly,
it is correct and appropriate to make such citations, but they are
scarcely an integral or even necessary part of the text; that is, nothing
essential, conceptual, or empirical would have been lost had the great
majority of the 208 citations been omitted. To put it differently, in very
few of the studies examined was a hypothesis derived from or inspired
by a Theory of Personality tested, and in no case was the validation or
invalidation of one of the Theories or even a part of one of the Theories
at issue. That they have a generalized distal influence on research is
true to varying degrees, but it is obvious that the proximal influences

on research are theoretical positions of more recent vintage and the
contemporary empirical literature. At this point I can do no better than
to quote Levy (1970): " ... one interpretation, which seems inescapable,
is that these theories of personality are not performing the integrative
and heuristic function we expect of a theory" (pp. 84-85).
WHAT IS TO BE DONE?
Now I do not wish to argue that what is, is necessarily the way
things ought to be. One cannot dismiss out of hand the possibility that
the manifest neglect of Theories of Personality by contemporary re-
searchers has more to do with their ignorance, indifference, laziness,
lack of historical sense, and low horizons than with the inadequacies of
the classical theories. But that seems to me incorrect on several
grounds. First, it is implausible that researchers would not be drawn to
theories that are useful, ones that could serve integrative and heuristic
functions. Why would they willfully ignore so valuable a resource if it
existed? Second, as I argued earlier, the speculative, imprecise, and
anachronistic character of the canonical theories do, in fact, render
them inadequate for contemporary scientific purposes. It is important
to note in this respect that the theorists whose positions are most
nearly free of those limitations, for example, Allport, Bandura, Cattell,
Eysenck, and Mischel, have a far from trivial influence on current re-
search. Finally, I believe that it is a misreading of the contemporary
literature to regard it as devoid of theoretical inspiration or signifi-
cance. What is true is that, with the exceptions just noted, the theories
that figure in the literature are other than those that appear in text-
books. It would be difficult, however, for many, I fear most, of our
students to appreciate the fact or even to understand that personality
is an evolving field, not one mired in the contemplation of the great
figures of its past.
And yet a problem remains. If it is true, as I would have appeared
to argue, that the field is in a healthy and productive state, why is
there a widespread feeling (one I share) that much of the empirical
literature is empty and trivial. Might it not be that by turning away
from the theorizing of the canonical theorists or by concentrating on
middle-level theories (cf. Maddi, 1984), we have condemned ourselves
to the collection of "itty-bitty'' facts, to use Allport's (1961) telling phrase?
I do not think so. Recent theoretical work based on, say, evolutionary
biology or conceptions of ego development may be lacking in the liter-
ary allure of psychoanalysis, but it is scarcely lacking in breadth or

potential integrative power. And there are any number of other recent
theoretical positions of which the same can be said. In sum, the fitful
intellectual sustenance to be gained from reading the journals in per-
sonality cannot reasonably be attributed to any impoverishment of the-
oretical possibilities resulting from the neglect of the classical tradition.
Having argued that the tediousness of much current research is
not due to the neglect of the classical theorists, I should, at least for the
sake of completeness, offer some alternative hypotheses. To be brief, my
candidates are the following: (1) the rigidity and rituals of what the
field regards as appropriate methodology; (2) the pressures to produce
articles at a high and steady rate, a factor that encourages the study of
immediate outcomes with procedures that yield a maximum of data
with a minimum of effort; (3) laziness; (4) the uneven distribution of
research talent; (5) the tendency of many investigators to think like
psychologists instead of like functioning social beings when they en-
gage in research; (6) the oversimplification of complex processes and
phenomena in order to make them amenable to investigation by the
methods that are available and acceptable to the profession; and (7) the
sheer difficulty of doing good research on human beings and here I
refer to both intellectual and ethical problems. Some might suggest an
eighth factor: the failure of researchers to place their empirical work
within a firm theo:retical context; but I have never been able to detect
a correlation between how significant or interesting a study is and the
strength of its tie to some explicit theory.
By now the point should be clear: It's time to kick the habit, to
break our addiction as teachers to theories of personality. The phrase
"as teachers" is, of course, crucial, for what we do as researchers and
as theorizers has but a passing resemblance to what we habitually teach,
at least to undergraduates. But if we abandon the course in Theories of
Personality,* what should we be doing as teachers? The answer to that
question is easy-we should be telling students about what we actually
do and how and why we do it. To quote myself, "I am convinced that as
teachers and writers of textbooks, we can best convey what makes the
study of personality consequential and exciting by introducing students
to the scientific activity that engages us as professionals" (1983, p. 437).
Ironically, that is what Allport and Stagner, in their very different
*The typical contemporary textbook in personality could be reserved for its proper setting
in courses explicitly devoted to the history of ideas in personality. As it stands now, such
courses would be indistinguishable from much of what is presently being taught in the
field; but if we were to break the addiction, they could serve an important function.
ways, did some 50 years ago. We could do no better than to take their
approach as a model now. In the end, all that I am advocating comes
down to this: "Preach what you practice."
REFERENCES
Winston.
Alpert-Gillis, L. J., & Connell, J. P. (1989). Gender and sex-role influence on children's
self-esteem. Journal of Personality, 57, 97-114.
Barzun, J. (1958). Darwin, Marx, Wagner: Critique of a heritage. New York: Doubleday.
Brown, P. (1967). Augustine of Hippo. Berkeley: University of California Press.
Hall, C. S., & Lindzey, G. (1957). Theories of personality. New York: Wiley.
Hall, C. S., Lindzey, G., Loehlin, J. C., & Manosevitz, M. (1985). Introduction to theories
of personality. New York: Wiley.
Hilgard, J. (1948). Theories of learning. New York: Prentice-Hall.
Levy, L. H. (1970). Conceptions of personality. New York: Random House.
Maddi, S. R. (1984). Personology for the 1980s. In J. Aronoff, R.A. Zucker, & A. I. Rabin
(Eds.), Personality and the prediction of behaviar (pp. 7-41). Orlando, FL: Academic
Press.
Maddi, S. R. (1989). Personality theories: A comparative analysis (5th ed.). Chicago:
Dorsey Press.
Mendelsohn , G. A. (1983). What should we tell students about theories of personality?
Contemporary Psychology, 28, 435-437.
Moskowitz, D. S., & Schwartzman, A. E. (1989). Painting group portraits: Studying
life outcomes for aggressive and withdrawn children. Journal of Personality, 57,
723-746.
Simonton, D. K. (1989). Shakespeare's sonnets: A case of and for single-case histori om-
etry. Journal of Personality, 57, 695-721.
Stagner, R. (1974). Psychology of personality (4th ed.). New York: McGraw-Hill.
PART FOUR
PRESENT-DAY
PERSPECTIVES ON BASIC
ISSUES
CHAPTER EIGHT
Science and the Single Case

IRVING E. ALEXANDER
I must express my gratitude for being asked to participate in this

anniversary celebration. The task assigned, a review of "science and
the single case," or the continuing saga of the study of the individual,
allowed me to reexperience the pleasure I derived in reading Allport's
volume (1937) as a graduate student more than 40 years ago. While my
present concern is centered in the first and last chapters, I was drawn
to rereading the others as well. My impression was that despite the
passage of the years, it could still be read with profit by graduate
students today. With regard to the scientific study of the individual, the
arguments remain timely and persuasive but certainly need retelling in
light of our history over the past half-century.
Allport immediately sets the tone of his ideological position by
referring to individuality as the supreme characteristic of human na-
ture. He then contrasts this obvious bit of knowledge possessed by the
ordinary person to the attitude exemplified in the sciences devoted to
the study of life processes that he characterizes as one where "the very
existence of the individual [is] somewhat of an embarrassment and [they]
are disturbed by his intrusion into their domains" (1937, p. 3).
At the beginning of his last chapter, entitled "The Person in Psy-
Presented in the APA Symposium, "Fifty Years of Personality Psychology,'' August 28,
1987, New York, New York.
IRVING E. ALEXANDER • Department of Psychology, Duke University, Durham, North

Carolina 27706.
New York, 1993.
119
120 IRVING E. ALEXANDER
chology," he attributes to the modern point of view (undoubtedly his

own) the demand that ''psychology expand its boundaries, revise its
methods, and extend its concepts to accommodate, more hospitably than
in the past, the study of the single concrete mental life" (1937, p. 549).
In support of this position, he refers, in a footnote, to the definitions of
psychology by Wundt, James, and Titchener, all stressing individuality,
although he points out the discrepancy between their aims and their
ultimate accounts of mental life. Ironically, this turned out to be largely
true of Allport's work as well.
Allport's focus on the individual grew out of an antielementaristic
stance. He did not want to study part processes but rather how things
common to human kind were ultimately put together in any individual.
In this sense, he was influenced by various theoretical positions arising
in late nineteenth century and early twentieth century Germany, es-
pecially the ''Verstehende Psychologie" of Dilthey and later Spranger
(emphasizing value or meaning) and also the "personalism" of Wilhelm
Stern. The dynamic holism of Gestalt psychology also captured his in-
terest although its relationship to the individual was not clearly drawn
until the work of Lewin. Finally, we cannot leave the issue of influence
without mentioning psychoanalysis, which had its origin in neighboring
Austria. Allport was intimately aware of the tenets of psychoanalytic
theory and recognized its potentiality for studying the dynamic pat-
terns of individual psychological existence. However, in this fJ.rst chap-
ter, he enumerates his antipathy toward the search in psychoanalysis
for universal causes, its slavish acceptance of Freud's doctrine, its iso-
lation from general psychology, its "one-sided" interest in the problems
of psychopathology, and its overemphasis on the sexual motive and on
the power of the unconscious. These unacceptable features kept Allport
in an ambivalent posture toward Freudian theory. He certainly could
not embrace the psychoanalytic position, especially as it related to the
direct impact of early experience on later personality development, nor
could he deny the importance of information resulting from the study of
"unconscious" mechanisms. His solution, in part, was the doctrine of
functional autonomy of motives, which reduced the all-embracing power of
the original conditions surrounding drive acquisition. Without question,
Allport imparted a clear and persuasive message that stressed the im-
portance of the study of the individual to the growth of the understand-
ing of human nature.
The growing strength of a personalistic position in American psy-
chology by the end of the 1930s was obvious. In addition to Allport's
clarion voice, a strong statement for the lawfulness of the individual
had been made earlier in a brilliant essay by Kurt Lewin in 1930 enti-
SCIENCE AND THE SINGLE CASE 121
tied, ''The Conflict between Aristotelian and Galilean Modes of Thought

in Contemporary Psychology." In the very same period, the pioneering
work of the group led by Henry A. Murray at the Harvard Psycholog-
ical Clinic was in progress and culminated in their classic report, Ex-
plorations in Personality, published in 1938, the year after the
publication of Allport's book. One powerful chapter in Murray's book
contains the extensive descriptions of the results of investigation by
multiple techniques, of a single human being, "Case History: Case of
Earnst."
When one adds to these milestones the strong interest generated
by the appearance of new techniques, the Rorschach (Rorschach, 1921)
and the Thematic Apperception Test (Morgan & Murray, 1935), which
had the promise of delineating dynamic individual profiles utilizing im-
portant personality dimensions, the day of the study of the individual
appeared close at hand. Yet, here we are a half-century after Allport's
treatise on personality appeared, not all that much farther along the
path. Although there have been occasional voices continuing to remind
us of this lamentable fact, the general climate in psychology as a dis-
cipline has been noticeably unresponsive over the years to the Allport-
ian message until rather recently. I should like to reflect on why this
might have been the case.
STUDY OF THE INDIVIDUAL: INHIBITING FORCES
The messages of Allport, Murray, and Lewin regarding the scien-

tific legitimacy, utility, and interest in studying the individual were
even more unusual since they were embedded in the social climate of
the 1930s, a period whose disastrous economic impact placed particular
emphasis on the common fate. Few escaped the ravages of the depres-
sion. Certainly at that time the plight of the individual so clearly delin-
eated, nurtured, and protected in our society by the founding fathers
was masked by the plight of the many. This particular attitude ex-
tended into the following decade fueled by the cooperative, group de-
mands necessary to sustaining a critical, national war effort.
The immediate ramifications of World War II on the study of per-
sonality were powerful. Two major needs were quickly specified, both
contained under the general umbrella of selection. One was concerned
with the identification of the particularly fit (the selection of officers,
pilots, special technicians, intelligence agents, etc.), which had formerly
been the province of industrial psychologists-the matching of the
worker to the job. The other focused on the identification of the partie-
ularly unfit, those unsuited psychologically to withstand the demands of

a life in the military, a major concern of psychopathologists. The efforts
that followed, extensions of the practices that already existed in psy-
chometrics, personality assessment, and psychopathology, were directed
away from any emphasis on the understanding of any individual qua
individual and turned toward the specific question of whether any indi-
vidual was a member or potential member of any particular designated
class. Does he possess the critical attributes necessary to succeed as an
airplane pilot? Can he stand the rigors of submarine duty? The ultimate
outcome of such study, especially under the press of time, is likely to be
the search for clues or signs (usually atomistic and unconnected, some-
times not easily explainable on logical grounds) that will improve a
selection ratio. False positives are easily identifiable, false negatives
especially in a buyer's market are easily forgotten. In any case, the
study of the individual as a dynamic, functioning whole, is lost for pre-
sumably the good, utilitarian reasons of efficiency and cost. This was, in
my estimation, one of the major impacts of those war years on the
study of personality.
In the ensuing years, the study of personality was engulfed and
largely dwarfed by the dramatic rise of the field of clinical psychology.
One would have thought that its rapid growth as a special discipline
might have held out the possibility of a rebirth of interest in the study
of the individual. For a variety of reasons, however, this was not to be
the case. Part of the difficulty can be traced to conflicting ideologies in
the development of a discipline. Those conflicts were multiple. They
existed in university departments, between clinical psychologists and
their colleagues in more traditional areas of study. In other settings,
they were interdisciplinary; power struggles occurred between clinical
psychologists and the psychiatrists who administered the hospitals,
clinics, and treatment programs in which practical skills were learned.
Furthermore, there was strife among clinical psychologists themselves,
who were variously split on the importance of training for research or
practice in this newly emerging, burgeoning specialty.
Clinical psychology as a unique field was fostered by an anticipated
national need: a returning population of young veterans who had been
exposed for long periods of time to the stresses of combat and of mili-
tary life. The need to train a whole new profession to deal with these
acute and anticipated long-range problems became a governmental con-
cern. Funds were appropriated and dispensed through the National
Institute of Mental Health and the Veterans Administration to create
or expand professions dealing with the problems of mental health and
illness. One consequence of this effort was that personality as a field of
study increased in prominence. Yet the barriers to individual study

remained formidable. The training issues from a philosophical and prac-
tical standpoint were resolved early. Without professional schools de-
signed to train clinical psychologists, as there were in medicine and
social work, a training base had to be selected. Since those who had
assumed leadership in this newly emerging field were trained in grad-
uate departments of psychology, the locus of the training was assured.
New graduate programs, handsomely supported by government subsi-
dies, were introduced into somewhat reluctant but appropriately entre-
preneurial traditional departments of psychology where research was
conceived as an entirely nomothetic enterprise, largely experimentally
based. The training model adopted for clinical psychology was that of
the scientist-practitioner with nomothetic research as the accepted stand-
ard for clinical students and faculty alike. The critical messages ema-
nating from the two major landlords with which clinical psychology had
come to live-clinical psychiatry and academic psychology-were pow-
erful enough to discourage any serious attempt to study the individual.
Clinical psychiatry emphasized the assessment task in the service of
medical-type diagnosis that led to nomothetic distinctions and fostered
both therapeutic interventions and practical research based on this model.
Academic psychology continued to regard the individual in a manner
that Allport detested: as a source of "error variance," a constant distur-
bance in the search for general laws. Career incentives for idiographic
research were minimal. For young faculty, idiographic interest seemed
a likely road to academic oblivion since research funding for such stud-
ies was meager and publication outlets were limited; for graduate stu-
dents it was an unlikely path to acceptable dissertation projects. These
constraints were firmly established in the decade or two following World
War II and have tended to restrict the growth of personological inves-
tigations ever since.
STUDY OF THE INDIVIDUAL: SUSTAINING FORCES
Given such constraints, one would have expected the ultimate dis-
appearance of the study of the individual except when that could be
accomplished by the method of nomothetic comparison. Although this
pattern of investigation became dominant, fortunately it did not turn
out to be the exclusive model and I should like to briefly trace what
seemed to be some of the countervailing forces.
The largest of these remains the unabated curiosity that humans
have about other people and about themselves. This interest is reflected
both nomothetically and idiographically. Certainly we wish to know

about people in general, about women and men, Russians and Chinese,
the young and the old, the troubled and the untroubled. But we also
want to know what single others who stir strong affects in us are like
and how we can come to know with any reasonable degree of certainty
how to detect what we wish to know about them. Probability state-
ments about personality characteristics derived from comparative
group data are not likely to satisfy that curiosity, since the primary or
raw data of experience with individual people are too compelling and
complex to be explained in that way.
On another level, though, despite the powerful impact of the spirit
of the times in the psychology of the 1950s and 1960s, there remained a
few voices who either emphasized the importance of the study of the
individual or presented theoretical or methodological schemas that sug-
gested frameworks for such study. Prominent exemplars were Robert
White (Lives in Progress, 1952/1966/1975), Harry Murray ("American
Icarus," 1955), Harold McCurdy (The Personal World, 1961), George
Kelly (The Psychology of Personal Constructs, 1955), Allport (Letters
from Jenny, 1965), Charles Dailey (The Assessment of Lives, 1971),
M. B. Shapiro ("The Single Case in Clinical Psychological Research"
1966). Methodological paths were opened by Kelly's Rep Test (1955),
Jack Block's development of Stephenson's Q technique (1961), and Julian
Rotter's development of incomplete sentences methodology (Rotter &
Rafferty, 1950). Many of these writers would probably not have de-
scribed themselves as thoroughgoing personologists; nevertheless, their
contributions helped to keep that tradition alive during a very difficult
period.
The scene began to change in the 1970s heralded in part by Carlson's
(1971) trenchant analysis of the role of the person in personality re-
search. Within the period of a few years, there appeared a succession of
papers introducing new analytical techniques or new applications of
analytical techniques for the study of the individual. Rosenberg and
Jones (1972) used factor analysis to interpret Theordore Drieser's work.
Simonton (1975) introduced time-series analysis. In the ten years from
1967 to 1977, a number of volumes appeared devoted to design and
analysis of studies employing the single case (Chassan, 1967/1979; Da-
vidson & Costello, 1969; Hersen & Barlow, 1976; Kratochwil, 1978; Neu-
feld, 1977). The climate in psychology was becoming more receptive
toward idiographic analysis. By 1979, Tomkins had already introduced
some of the basic ideas of "script theory," a stance ideally suited for
examining the psychological life of an individual, which Tomkins (1987,
1991, 1992) has continued to develop.
At roughly the same time, two psychological volumes appeared-

one in psychology (Stolorow & Atwood, 1979) and one in philosophy
(Scharfstein, 1980)-that examined the life work relationship of signifi-
cant contributors to those disciplines. These works were reflective of
the fact that psychobiographical analysis, long a scholarly outlet for
those in the psychoanalytic tradition, was enjoying a more general re-
birth largely stimulated by Erikson's work on Martin Luther (1958) and
on Mahatma Gandhi (1969).
The past decade produced a continuation of this vein of interest in
the study of the individual. Lamiell's (1981) cogent methodological crit-
icisms of a group comparative approach to knowledge about any partic-
ular member of that group served a function similar to that served by
Carlson's (1971) paper. He again reminded us that our interest in infor-
mation about particular people could not be satisfied by the methods of
data analyses we were employing. At this point one might wonder why
the message about individuality and its importance has to be reintro-
duced periodically in the history of personality study. Perhaps the prob-
lem is similar to all others in which the cogency of the issue is clearly
identified but no simple remedy is apparent. Under such conditions we
tend to put the problem on the back burner, returning to study that
which is customary, feasible, and economical, despite our implicit knowl-
edge that the problems concerning the study of the individual remain
largely unsolved.
The study of narratives, or life histories, certainly received posi-
tive impetus from the appearance of Runyan's (1982) important volume
Life Histories and Psychobiography. In it he performed a great service
by reviewing both the distant and more recent efforts in these areas.
The book has rapidly become the standard introduction to the field. Its
contents include not only discussions and critiques of cogent method-
ological issues, but also clear examples of sound use of data in psycho-
biographical analysis.
Within a few short years, the Journal of Personality (March 1988)
devoted a special issue to studies employing life narrative material as
basic data for general personality questions or for specific psychobio-
graphical inquiry. The papers sufficiently varied in subject matter illus-
trated clearly the continuing and expanding interest in such research
among personality psychologists. That psychobiography and individual
life narrative study were not simply a scattered enterprise was further
attested to by the formation in the mid-1980s of the Society for Per-
sonology, whose members meet annually to discuss research develop-
ments in this field.
THE CURRENT SITUATION
The year 1990 produced two books that further illustrate the in-
creased receptivity for this kind of work in the study of personality.
One, Dan McAdams's The Person, is cast in a textbook format but
constitutes a rather wide departure from traditional personality texts.
Although it reviews both theory and research in what has become the
province of standard tomes in this area, its focus and major emphasis is
on the understanding of the person and not simply of people per se. In
this regard, he has devoted a sizable portion of the book, without ne-
glecting traditional coverage, to both theory and research in the per-
sonological mode. The book is dedicated to both Allport and Murray. It
is superbly written, filled with state-of-the-art literature, and should
have a most positive influence on young people beginning the study of
personality. The second, Personology: Method and Content in Person-
ality Assessment and Psychobiography, by Alexander (1990), will by
its title, if nothing else, legitimize the Allport and Murray heritage. It
offers methodological guides to the analysis of personal data, reviews
what may be critical in preparing young scholars for personological
inquiry, and seeks further insights into the interplay between the life
as lived and the work accomplished in the histories of Freud, Jung, and
Sullivan.
While I have reported briefly on what I see as a revival of serious
scholarly work in the personological tradition, I have focused almost
exclusively on work emanating from the general area of personality.
One cannot overlook the fact that the study of individual lives and their
products exists in a variety of fields, both in psychology and in neigh-
boring disciplines. The literature of child development and aging both
reflect recent interest in linguistic analysis of individual subjects. Psy-
chohistorical inquiry appears to be flourishing as evidenced by new
journals devoted to this kind of inquiry. Certainly political analysts
both in the popular press and biographical essays appear to be paying
increasing attention to the effect of personality factors in the perfor-
mance of political leaders. While the increase in scholarly activity is a
positive sign, it also serves to point up the multiplicity of problems
both theoretical and methodological that must be solved before such
study will stand on firm, consensual grounds. Many of these issues are
treated at length by Runyan (1982) and McAdams (1990). What role
personality psychologists and personologists particularly will play in
the development of this field of study will be critical to its essential
success. Historians, political scientists, other social scientists, and even
literary biographers are not likely to generate either the conceptual or
methodological advances necessary to establish this line of inquiry on

solid footing. They are more likely to borrow from psychology that
which seems available and usable. The extreme emphasis on Freudian or
Ericksonian interpretive schemas in psychobiographical work would
lend testimony to this conjecture.
These remarks about the role of psychology in the future of per-
sonological study leads me to think about more structural and systemic
matters. How does one obtain relevant training to do personological
work? The answer turns out to be neither clear nor simple. There are
no graduate or even postgraduate programs with designations that would
lead one unequivocally to competency in the pursuit of personological
goals. Interestingly enough, this has always been true in psychology
even as far back as the golden days of the Harvard Psychological Clinic
in which many of the leading personologists-White, Tomkins, Sanford,
Rosenzweig, and MacKinnon-were trained. The usual training path
for people with personological interests has been either through clinical
psychology or psychoanalytic training programs. The closest institu-
tional setting that could foster such a goal in the recent past was at the
University of California, Berkeley, which through its Institute for Per-
sonality Assessment Research maintained a strong graduate program
in personality.
The developments of recent years would seem to point toward the
need for cross-disciplinary cooperation in the preparation of person-
ological specialists. Whether there will be such interdisciplinary pro-
grams in the future is a question whose answer does not invite a great
deal of optimism on my part. Programs in American education in order
to be launched at the present time require rapid, face value, bottom-
line payoffs, or else must be entirely self-supporting. The study of sin-
gle human beings is complex, time consuming, and not always certain
to include obvious, immediate benefits. If my analysis is correct, it is
very likely that as in the past personologists will develop slowly and in
their own unique ways.
REFERENCES
Alexander, I. E. (1990). Personology: Method and content in personality assessment and

psychobiology. Durham, NC: Duke University Press.
York: Social Science Research Council.
Block, J. (1961). The Q-sort method in personality assessment and psychiatric research.
Springfield, IL: Thomas.
Carlson, R. (1971). Where is the person in personality research? Psychological Bulletin,
75, 203-219.
Chassan, J. B. (1979). Research design in clinical psychology and psychiatry (2nd ed.).
New York: Irvington. (Original work published 1967)
Dailey, C. A. (1971). The assessment of lives. San Francisco: Jossey-Bass.
Davidson, P. 0., & Costello, C. G. (1969). N=l: Experimental studies of single cases. New
York: Van Nostrand-Reinhold.
Erickson, E. H. (1958). Young man Luther. New York: Norton.
Erickson, E. H. (1969). Gandhi's truth. New York: Norton.
Hersen, M., & Barlow, D. H. (1976). Single case experimental designs. New York: Per-
gamon Press.
Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton.
Kratochwil, T. R. (Ed.) (1978). Single subject research. New York: Academic Press.
Lamiell, J. T. (1981). Toward an idiothetic psychology of personality. American Psychol-
ogist, 36, 276-289.
Lewin, K. (1930). The conflict between Aristotelian and Galileian modes of thought in
contemporary psychology. Journal of General Psychology, 5, 141-177.
McAdams, D. P. (1990). The person: An introduction to personality psychology. San
Diego: Harcourt Brace Jovanovich.
McAdams, D. P., & Ochberg, R. L. (Eds.). (1988). Psychobiography and life narratives.
Special issue of the Journal of Personality, 56, 1-326.
McCurdy, H. C. (1961). The personal world. New York: Harcourt, Brace and World.
Morgan, C. D., & Murray, H. A. (1935). A method for investigating fantasies: The The-
matic Apperception Test. Archives of Neurology and Psychiatry, 34, 209-306.
Murray, H. A. (1955). American Icarus. In A. Burton & R. E. Harris (Eds.), Clinical
studies in personality (Chapter XVIII). New York: Harper & Brothers.
Neufeld, R. W. (1977). Clinical quantitative methods. New York: Grune and Stratton.
Rorschach, H. (1921). Psychodiagnostik. Bern: Bircher.
Rosenberg, S., & Jones, R. (1972). A method for investigating and representing a per-
son's implicit theory of personality: Theodore Dreiser's view of people. Journal of
Personality and Social Psychology, 22, 372-386.
Rotter, J. B., & Rafferty, J. E. (1950). Manual: The Rotter incomplete sentences blank.
New York: Psychological Corporation.
Runyan, W. McK. (1982). Life histories and psychobiography. New York: Oxford Univer-
sity Press.
Scharfstein, B. (1980). The philosophers: Their lives and the nature of their thought. New
York: Oxford University Press.
Shapiro, M. B. (1966). The single case in clinical psychological research. Journal of Gen-
eral Psychology, 74, 3-23.
Simonton, D. K. (1975). Sociocultural context of individual creativity: A trans-historical
time-series analysis. Journal of Personality and Social Psychology, 32, 1119-1133.
Stolorow, R. D., & Atwood, G. E. (1979). Faces in a cloud: Subjectivity in personality
theory. New York: Aronson.
Tomkins, S. S. (1979). Script theory: Differential magnification of affects. In H. E. Howe
& R. A. Dienstbier (Eds.), Nebraska Symposium on Motivation (Vol. 26, pp. ~).
Lincoln: University of Nebraska Press.
Tomkins, S. S. (1987). Script theory. In J. Aronoff, A. I. Rabin, & R. A. Zucker (Eds.),
The emergence of personality (pp. 147-216). New York: Springer.
Tomkins, S. S. (1991). Affect, imagery, consciousness: The negative affects-anger and
fear (Vol. 3). New York: Springer.
Tomkins, S. S. (1992). Affect, imagery, consciousness: Cognition-duplication and trans-
formation of information (Vol. 4). New York: Springer.
White, R. W. (1975). Lives in progress. New York: Holt, Rinehart and Winston. (Original
work published 1952; subsequent publication 1966)
CHAPTER NINE
Describing Lives
Gordon Allport and the "Science" of Personality
BERTRAM J. COHLER
No one has approached the task of describing lives with greater humil-
ity than Gordon Allport, whose contributions have been so important in
shaping contemporary psychological inquiry. At the time of his death,
in 1967, he was a preeminent figure in the human sciences; his work
encompassed social psychology and social ethics as well as personality
inquiry. He largely fashioned the concept of social attitudes, and his
conjectures about the ways these determine our actions have led to
numerous studies of attitude-behavior relations. His integrative work,
The Nature of Prejudice (1954b), illustrated how multiple perspectives
could be brought usefully to bear on a social problem. His historical
survey [in the Handbook of Social Psychology (1954a) and reprinted in
the two editions that followed] has edified numerous psychologists and
sociologists and promises to retain its status as a basic source well into
the twenty-first century.
Allport identified the central issues regarding both methods of study
and substantive areas of inquiry in social psychology. However, he is
best known for his contributions to the study of the person. Reviewing
the history of personality psychology, Craik (1986) suggests that All-
port's 1937 text, along with Stagner's (which appeared in the same
year), had much to do with advancing the study of personality from a
BERTRAM J. COHLER • Committee on Human Development, University of Chicago,

Chicago, Illinois 60637.
Fifty Years of Persorw,lity Psychology, edited by Kenneth H. Craik et al. Plenum Press,
New York, 1993.
131
132 BERTRAM J. COHLER
"preidentity" phase to the status of a recognized subdiscipline within

psychology. Allport understood the conceptual challenges confronting
students of personality; and he skillfully formulated many of the prob-
lems that personality research has addressed in the past half-century.
Owing partly to Allport's clear and provocative presentation of the cen-
tral issues, we are now beginning to make a bit of headway on some of
them, as documented by several accounts in this volume. To appreciate
Allport's vision as a personality theorist, one can compare his early
construal of an issue (cross-situational consistency, for example) with
today's state-of-the-art achievements on that issue (see Chapter 17, this
volume, for a discussion of current theory and research on consistency).
On most issues, Allport poses the problem in a way that invites re-
search; and on many issues, there has been research progress.
Allport was an action psychologist. Concerned that psychology should
make enduring contributions toward understanding social life and re-
lieving social conflict, he sought to understand the means by which
conflict could be reduced within the community and nation and in reso-
lution of international tensions. He was also concerned with the inte-
gration of American and European psychology, forging new initiatives
for study among psychologists of many nations. His continuing interest
in Continental philosophy and psychology led to dialogues with clini-
cians; he fostered an alliance between existential phenomenology and
psychology by taking seriously the works of Binswanger, Victor Frankl,
and others. This "third force" [as Allport, following Maslow (1954),
described existential psychology] was seen as an alternative to behav-
iorism and reductionistic psychoanalysis, the two systems of psychol-
ogy that dominated Western psychology in the decade following World
War II. Allport possessed a good knowledge of philosophy. His use of
phenomenology may have initially limited his impact in the area of
personality. From the vantage point of 1990, though, it is obvious that
the phenomenological perspective has exerted significant long-term in-
fluences on personality psychology with regard to method as well as
content.
Notable features of Allport's career included a continuing commit-
ment to the empirical study of personality processes and a readiness to
explore various philosophical approaches. His writings rank among the
most comprehensive accounts of the complex relationship between per-
son and social context, including the extent to which understanding of
both self and others might be influenced by situational variables. In
view of Holt's (1962/1978) critique to the effect that Allport was not a
systems theorist, it is interesting to note that much of Allport's discus-
sion of the interplay between person and social context anticipates a
DESCRIBING LIVES 133
systems view. Indeed, in the 1961 revision of Allport's textbook, his

main criticism of Stagner's approach is that it places too much empha-
sis on homeostasis. Allport maintained that Stagner, like many other
psychologists at that time, was overly concerned with motivation as a
tension-reduction phenomenon and too unwilling to acknowledge the
teleological aspects of personality:
Most current theories of personality take full account of two of the
requirements of an open system. They allow interchange of matter
and energy, and they recognize the tendency of organisms to main-
tain an orderly arrangement of elements in a steady state. Thus
they emphasize stability rather than growth, permanence rather
than change, "uncertainty reduction" (information theory) and "cod-
ing" (cognitive theory) rather than creativity. In short, they empha-
size being rather than becoming. Hence, most personality theorists
are biologistic in the sense that they ascribe to personality only the
two features of an open system that are clearly present in all living
organisms. (Allport, 1960, p. 44)
Allport also suggested that Stagner's concept of trait was not opera-
tionally defensible, being based on too few indicators (Allport, 1961,
p. 337). For this reason, Stagner's position (as presented in his 1951
Psychological Review article on homeostasis, for example) seemed likely
to overestimate the amount of cross-situational consistency to be ex-
pected in a person's actions.
ALLPORT'S APPROACH TO THE STUDY OF THE PERSON
As one of Allport's students and graduate assistants, I had a good

opportunity to observe Allport as a teacher and a scholar; I was par-
ticularly struck with the dignity and respect he accorded to others. His
quotation from Spinoza, leading off Pattern and Growth in Personality
(1961), is noteworthy: "I have made a ceaseless effort not to ridicule,
not to bewail, nor to scorn human actions, but to understand them"
(p. vii).
Allport and I did not always agree on the scope and methods of
personality inquiry. Allport could not understand my fascination with
psychoanalytic approaches to the study of lives; he viewed psychoanaly-
sis as an unnecessarily reductive and restrictive approach. In retro-
spect, I can appreciate Allport's discomfort with psychoanalytic accounts
of personality structure and development.
Allport recognized that portrayal of particular lives and studies of
aggregates provided complementary approaches. However, he appeared
to mistrust the very person-centered approach that he championed [a

paradox noted also by Holt (1962/1978)]. During the course of my doc-
toral studies, a group of us became interested in the detailed tracking
of particular lives over time and were spending much time examining
interview and semistructured test data. When we met with Allport to
discuss the issues involved in integrating interview and test results, he
cautioned us that "idiography was no country for young men." I re-
garded this as more than a statement of political realities; it conveyed
some of Allport's ambivalence regarding the approach that he had so
long championed. On the other hand, discussing my fledgling efforts to
create what ultimately became the first automated Minnesota Multi-
phasic Personality Inventory (MMPI) procedure, Allport cautioned me
that we know far more about tests than we do about persons.
I should also add that there was some tension between Allport and
the personnel of Harvard's Psychological Clinic, including Henry Mur-
ray and Robert W. White. A certain amount of rivalry or conflict was to
be expected, with so many major figures in contemporary psychology
working as colleagues on the same campus. But both Murray and White
approached the study of lives from the clinical tradition, applying in-
sights obtained from records of disordered lives in the construction of a
theory of "normal" personality development. Allport was concerned di-
rectly and mainly with the "normal" personality; his approach relied on
philosophical concepts (drawn largely from existentialism) and relevant
findings from social and developmental psychology. Even so, Allport
was distressed when informed that the Journal of Abnormal and So-
cial Psychology, for which he had served as editor, was to become two
separate journals; thus he may have believed that both "normal" and
"abnormal" psychology could contribute importantly to the understand-
ing of personality.
Allport offered some pointed criticisms of the personality tradition
that Murray and his associates had pioneered. In the first place, All-
port disagreed with Murray's distinction between traits and disposi-
tions able to evoke action (so-called motives). While some dispositions
may appear more central or pervasive in a person's life (so-called cardi-
nal dispositions), Allport maintained that, although dispositions may
vary in intensity, all dispositions were able to evoke action. In the
second place, Allport disagreed with Murray's psychodynamic model of
personality development. Allport particularly mistrusted the epigenetic
approach to understanding lives, in which explanations of adult out-
comes were inevitably and all too readily linked to presumed childhood
origins based on a naive concept of critical period. To some extent,
Allport's notion of functional autonomy can be regarded as a protest

against and a kind of replacement for epigenetic explanation.
From his very earliest writing, Allport was concerned with the
coherence and integrity of self. He also recognized the constraints im-
posed on personality study by the fact that one person's attribution of
meaning to a particular action need not correspond to another's, and he
saw that multiple perspectives would be needed for exploring people's
constructions of the world.
With the person-rather than behavior-as the focus of investiga-
tion, Allport addressed basic questions of method: What techniques are
most appropriate for studying people's lives? Can we obtain a record of
a subject's actions, life events, and experiences that simultaneously does
justice to the coherence and integrity of the selr? What sources of
information are most useful for individuals? Which sources are best for
aggregates?
The particular "documents of life" a psychologist opts to use will
necessarily color his conclusions; and regardless of the documents one
selects, problems will be posed by the relationship between subject and
observer and the ever-present necessity of interpretation (Crapanzano,
1980; Freeman, 1985; Ricoeur, 1971a, 1971b, 1980, 1981). In these re-
spects, the study of personality differs fundamentally from the natural
sciences:
... the crisis of social science concerns the nature of investigation
itself. The conception of the human sciences as somehow necessar-
ily destined to follow the path of the modern investigation of nature
is at the root of this crisis. (Rabinow & Sullivan, 1979, p. 4)
THE INTERPRETIVE PERSPECTIVE

Charles Taylor (1985) traces the history of the modern interpretive
perspective from the nineteenth-century German philosopher Dilthey
through Windelband and Spranger, noting that its present resurgence
is due largely to the work of Ricoeur (1971a, 1971b) and Habermas
(1971). A major intellectual impetus for this tradition has been the
work of the same European scholars who formulated the concept of
human sciences to refer to those social sciences that focus on issues of
meaning and intent and whom Allport studied early in his career. Upon
receiving his doctoral degree in 1922, Allport was awarded a Sheldon
traveling fellowship and spent the next two years in Germany and En-
gland, where he worked with Stern and studied the writings of Dilthey,
Windelband, and Spranger. One product of this effort was the Allport-
Vernon (1931) A Study of Values.
In "The Study of Personality by the Intuitive Method" (Allport,

1929), a little known but important paper, Allport discussed the contri-
butions of German psychologists pioneering the Geisteswissenschaften
with their emphasis on verstehen methods. He advocated an interpre-
tive methodology:
The rigid methods employed in the current psychological analysis
of personality are prescribed in deliberate imitation of the natural
sciences. Whatever the gains may have been, this subservience has
resulted in the conviction on the part of many that such methods
are insufficient, and that there is a need, especially in dealing with
personality, of what Jaspers has called a "psychological psychol-
ogy." (p. 17)
Reviewing the contributions of Croce, Bergson, Spranger, and others,
Allport called for renewed study of personality as both an art and a
science. I interpret Allport's use of the word "art" to refer to empathic
(verstehen) understanding, and the word "science" to refer to disci-
plined scholarly inquiry focusing on meaning [in which "the image of
the man is never lost; it is merely brought into focus" (P. 21)]. Thus
Allport anticipated the "interpretive turn" of social science in the sec-
ond half of the twentieth century (Rabinow & Sullivan, 1979), present-
ing a persuasive argument against the possibility of a value-free, fully
objective psychology of personality.
THE INTERPRETIVE TURN AND THE

NOMOTHETIC-IDIOGRAPHIC DEBATE
The Continental phenomenologists and psychologists whose writ-

ings seem to have influenced Allport's views on the necessity of in-
terpretation in the social sciences probably provided some of the
inspiration for his focus on the study of the person as well. The
nomothesis-idiography distinction so important in Allport's work had
been elucidated in Windelband's 1904 monograph on history and natural
science. History was an interpretive discipline based on more or less
singular events, whereas natural science could be nomothetic and deal
with aggregated cases.
There has been a continuing controversy in psychology about the
relative merits of nomothetic versus idiographic approaches. Holt (1978),
for example, suggested that there is a lawfulness to people's lives; in
broad outline, the course of each person's life is guided by this aggre-
gate lawfulness. Therefore, the psychologist should in principle be able
correctly to infer from the individual to the aggregate, and vice versa.
In spite of its logical defensibility, however, Holt's treatment of the
problem held little interest for Allport: It did not fully address the rich
distinction laid out by Windelband.
Allport (1962) advocated a "morphogenic" perspective, which pre-
sumed (1) a unique organization or structure of personality within each
person, and (2) no uniqueness of elements (i.e., traits) of which the
structure was composed. He was more concerned with advancing the
morphogenic point of view than with predicting particular actions, dem-
onstrating behavioral regularity, or highlighting the idiography-
nomothesis polarity as initially portrayed by Windelband.
PERSONAL DOCUMENTS
There is now much debate regarding use of personal documents in

the study of persons. The issues in dispute include the representative-
ness of particular life history documents for a particular life, how best
to define units in analyzing documents, and the extent to which one
may generalize from particular lives to larger groups (Rosenwald, 1988;
Runyan, 1982, 1988; Stewart et al., 1988; Weintraub, 1975). Allport was
aware of these issues but not centrally interested in them. Rather, he
sought to adduce support for his morphogenic view: He hoped to dem-
onstrate that the structure of each person's life, though describable in
terms of the same elements assessed via the same methods as anyone
else's, was in some important respect unique.
Many of those who have recently discussed the idiography-
nomothesis problem in psychology arrive at a position similar to All-
port's. Bern (1983), for example, says
I believe a successful interactional theory ... is not likely to be id-
iographic, but it will be morphogenic or person centered. That is, it
will not invoke a unique set of variables for characterizing each
person ... but it will be concerned with the unique salience and
configuration of a common set of variables within the person rather
than with the relative standing of persons across those variables.
(p. 573)
New methodologies developed for the study of individual lives are
also compatible with the morphogenic perspective. Hogan's (1978, 1980,
1981) treatment of career sequences, Chiriboga's (1978) life chart, and
Runyan's (1982) portrayal of stage-state analysis are particularly good
examples. Although persons share conceptual timetables of optimal de-
velopment and evaluate their own progress in terms of these (Roth,
1963), the manner in which they negotiate their individual careers shows
marked variation as a function of cohort, social status, ethnicity, and
other social factors. Little's (1983) concept of personal projects and his
method of quantifying one's progress toward completion of such pro-

jects are also consistent with the morphogenic approach and can be
extended into a life course perspective.
The array of "idiographic" methods that have been explored over
the years is reviewed by Runyan (1983). Although they all have po-
tential utility as morphogenic measures, Allport was most comfortable
with content analysis of written documents; he liked the fact that these
data provided contexts and configurations in which individual traits
could be discerned. Although such traits do not permit comparisons
across persons ("common" traits fulfill this role), idiographic study is
valuable because it enables us to contemplate the structure and coher-
ence of a particular personality (i.e., to define someone's "personal dis-
position"). If the available documents span an appreciable length of
time, this method may also reveal at least some threads of the fabric of
continuity that Allport expected to find in a person's life.
In a well-known content analysis, 172 letters written by "Jenny,"
the mother of Allport's college roommate, were read by clinical judges,
who ascribed traits to her on the basis of common sense. "Many of the
selected trait names were obviously synonymous, and nearly all fell
readily into eight clusters" (Allport, 1966, p. 7). Years later, the same
letters were content coded by other judges; factor analysis of their
contents yielded seven factors, six of which corresponded to one of the
clusters obtained from the original group of clinicians. Allport concludes
that although " ... there is no possibility in this case of obtaining exter-
nal validation for the diagnosis reached ... We can say that the direct
commonsense perception of Jenny's nature is validated by quantifica-
tion, coding, and factoring'' (1966, pp. 8, 9). In other words, although it
is clear what these letters signify about Jenny's nature, they constitute
only one source of data. Allport tacitly acknowledges that some signifi-
cant convergence among different kinds of evidence would provide a
more persuasive account of her personal dispositions.
Similarity between the conclusions arising from these two ''read-
ings" of Jenny's letters is presented as a tour de force of idiographic
analysis. Its results could be useful for nomothetic investigations, too,
in that the traits dependably ascribed to Jenny might turn out to be
applicable to the description of others as well. Whereas idiographic
analysis permits identification of individual traits (which may or may
not be useful in describing someone else), nomothetic studies demand
that common traits (those which apply to just about everyone) be as-
sessed. The objective of nomothetic work---eomparisons among subjects-
is meaningful only for common traits.
ALLPORT'S CONSTRUAL OF PERSONALITY STRUCTURE
Allport opposed the mechanistic "drive" theories of his time and those
developmental theories that postulated strong links between early child-
hood experiences and one's adult personality. His psychology is a reac-
tion against behavioristic interpretations (e.g., Miller & Dollard, 1941),
which regarded personality as sets of habits or behavior tendencies
specific to the particular situations in which they were learned, as well
as against those psychoanalytic theories of development that presume
powerful and long-lasting motivational consequences of the child's mas-
tering of developmental tasks.
The formulation devised by Allport designates traits as the units
in terms of which personality is to be described and the proprium to
account for the individual's sense of selfhood.
TRAITS
Allport based his concept of trait on work by the German type

theorist William Stern, whose 1921 textbook on differential psychology
Allport had read during his postdoctoral studies in Europe (Evans,
1970) (it is also noteworthy that Allport's books contain numerous ref-
erences to Stern). Allport's definition changed only slightly over the
years and always carried an appeal to physiology: A trait is "a neural
disposition of complex order, [which] may be expected to show motiva-
tional, inhibitory, and selective effects on specific courses of conduct"
(1937, p. 319) and "a neuropsychic structure having the capacity to render
many stimuli functionally equivalent, and to initiate and guide equiv-
alent (meaningfully consistent) forms of adaptive and expressive be-
havior" (1961, p. 347).
In the 1961 edition of his textbook, Allport attempted to specify
wherein traits differed from other forms of response readiness or de-
termining tendencies of personality, such as habits (which are less gen-
eral than traits) and attitudes (which have an object of reference and a
value component, pro or con, whereas traits lack both). There are com-
mon traits, distributed in an approximately normal manner within a
culture, and personal (or individual) traits, which are unique in their
organization and interrelationships within the person. Personal traits
vary in content from one individual to another and vary in salience
within the individual; there are cardinal traits (roughly equivalent to
Murray's unity thema), central traits, and secondary traits. Allport is
less concerned with issues of intention or motivation than with detailed
description; what is important about personal dispositions is their
organization within the individual rather than whether they are instru-
mental or motivational. Indeed, the definition of personal disposition
(Allport, 1961, p. 373) is virtually identical with the definition of trait
(p. 347), except for a parenthetical addition of the phrase "peculiar to
the individual." Morphogenic study, the mapping of personal disposi-
tions in terms of their salience and interconnectedness, is a means of
achieving comparatively full knowledge of another person.
Interconnectedness is clearly of prime importance. If the theory
cannot specify how the elements go together, it ends by portraying the
person as little more than a bundle of traits, some highly salient, some
less salient. A theory of personality ought to account for one's feeling of
identity or at least attempt to explain how it is that most people achieve
and maintain a sense of coherence or integrity of experience.
REALIZATION OF PERSONAL INTEGRITY: THE PROPRIUM
The past two decades have seen an appreciable shift toward sub-
jectivism in personality psychology. This trend has led to a resurgence
of interest in the self (including self-esteem) and in phenomenological
approaches to personality. The focus is now on subjective organization
of experience; there has been a dramatic turning away from the be-
havioral tradition. Concern with overt actions and their consequences
has been replaced, to a large extent, by concern with the ways persons
construct meaning in their lives. A partial list of illustrative topics
includes the origins and functions of plans and intentions (Ainlay, 1986);
autobiographical memory (Rubin, 1986) and its incessant revision (Green-
wald, 1980); the social construction of narratives (Elder et al., 1984);
wishes, goals, strivings, and projects (Emmons, 1989; Little, 1983); main-
tenance of self-congruity in the face of inconsistent or dissonant infor-
mation (Sirgy, 1986); and the ways in which we formulate the meanings
of our actions (Wegner & Vallacher, 1987).
Within psychoanalysis, concern has shifted from study of experi-
ence-distant concepts derived by analogy from experimental psychol-
ogy (such as ego functions or processes ) to the experience-near realm,
focusing on determinants of sense of coherence and capacity for solace.
Theorists as diverse as W. D. Winnicott (1965), George Klein (1976),
and Heinz Kohut (1971, 1977) have sought to describe dynamics of the
process by which one maintains a sense of personal integrity. Within
child psychology, works by Kagan (1980, 1981) and Kegan (1982) pre-
sent constructiv~evelopmental perspectives highlighting the emer-
gence of selthood.
Lasch (1979) views this emerging focus on subjective experience as
a sign of the increased narcissism of our time (reflected in the past

decade by proliferation of popular magazines such as Self, as well as in
other media). However, it is more likely that popular culture, influ-
enced perhaps by psychological research and/or consumer research, has
come to appreciate the importance of understanding the means by which
the individual realizes personal integration and maintains positive mo-
rale. Indeed, the study of the origins and vicissitudes of self-regard
(topics considered under the heading of ''proprium" in Allport's work) is
among the central concerns of the contemporary psychology of person-
ality. Much of this current interest in the self was anticipated by Al-
lport in the two editions of his text.
Allport was much influenced by William James. In the 1937 edition
of his personality text, Allport used the concepts and terminology (e.g.,
"consciousness of self') of James's classic The Principles of Psychology
(1890/1918, Vol. 1., p. 336) and viewed the self-as James did-as a
unity of consciousness representing personal identity, continuous though
continually changing. In distinguishing between bodily and social self
and in dealing with self-preservation and sense of integrity, Allport also
echoes James's earlier formulation of the same concepts.
The 1961 edition of Allport's text presents a different account of
the self, one that elaborates James's definition of the self as a "fighter
for ends." Here, Allport introduces a new concept, the proprium, which
includes the self as both knower and object of knowledge. Propriate
striving [which subsumes "directedness" and "intentionality," but is at
the same time a dynamic unifying phenomenon, "the cement holding a
life together," (p. 126)] dictates our patterns of self-involvement as op-
posed to mere task involvement. Our sense of self arises from the pro-
prium; propriate strivings are important determinants of "attention,
judgment, memory, motivation, aspiration level, productivity, and the
operation of personality traits" (p. 128), as well as of our reactions to
personally relevant consequences of our acts.
Other propriate states, such as self-esteem and self-image, influ-
ence the expression of strivings and the ways we respond to events
that befall us. Certain patterns of propriate states, says Allport, are
recognizable as pangs of conscience, feelings of inferiority, or a sense of
obligation (1961, pp. 134-138).
During the past three decades, efforts by psychologists of various
persuasions have brought personal integrity (and closely related topics,
such as self-esteem) to a prominent position in the study of personality.
The psychoanalysts Kohut (1977, 1984) and Klein (1976), the survey
researcher Morris Rosenberg (1979), the cognitive-experiential theo-
rist Epstein (1980, 1981), the social-learning theorist Bandura (1982),
and the social psychologists Gergen and Gergen (1983) and Antonovsky
(1979, 1987) are just a few of the many writers whose descriptions of
the self are highly compatible with Allport's 1961 statement. The writ-
ings of Klein and Kohut are perhaps the clearest examples. Klein pos-
tulates the existence of an active self that functions as an integrator or
synthesizer of one's experiences; when it is operating effectively, the
person has a sense of continuity, coherence, and integrity-experience
is rendered "self-syntonic." Perceived coherence is by no means a lux-
ury, though. Kohut has documented the impact on lives of the failure to
maintain it: People who lack a coherent narrative of the course of their
lives report feelings of depletion and low morale. They sometimes re-
sort to disavowal or disowning of their actions in order to maintain a
consistent self-narrative, but this is a psychologically risky expedient
because it often leads to a sense of fragmentation.
I do not wish to imply that all the self theorists took their cue from
Allport's concept of the proprium. They may well have come by their
ideas independently or gotten them from others (in 1951, for example,
both Carl Rogers and Jean Piaget published books specifying the im-
portance of congruity between percepts and self-structures or sche-
mata). Though Allport was a penetrating thinker and writer, like other
scholars he was also both product and agent of the Zeitgeist. His works
need to be understood in historical perspective as well as on their own
merits. I want to suggest that Allport was one of the creators of the
Zeitgeist of academic psychology in the second half of the twentieth
century; as such, it is likely that he influenced some of the self theorists
who were his contemporaries and some of those who followed.
CONCLUSIONS
Allport's primary contribution to the study of the person may be

less a matter of theoretical notions, methodological prescriptions, or
empirical work than of his uncanny ability to comprehend the major
issues in the field. As early as his first papers, Allport was aware of the
fundamental problems confronting those who wished to study persons,
such as the problem of distinguishing between text and interpretation,
the advantages and drawbacks of individual difference formulations in
the study of personality structure, the fact that traits as well as en-
vironments were ever-changing, and the challenges of accounting for
continuity and change in lives over time. In his 1937 textbook, Allport
was already protesting the view that childhood experiences inevitably
shaped one's destiny; he maintained instead that one's own defmition of
the meaning of self was a powerful factor in determining long-term

outcomes.
Concepts such as functional autonomy and propriate strivings were
created to account for phenomena of growth and development, change
and stability over time and the creative capacities that sometimes en-
abled individuals to overcome adversity. Allport was able to foresee
shifts within both behaviorism and psychoanalysis away from emphasis
on the person's reactive tendencies and toward emphasis on the pro-
cesses by which morale is maintained and the sense of personal congru-
ence is enhanced. And he was able to make some telling contributions
of his own, which made it more certain that these shifts in the dominant
paradigms of the mid-twentieth century would actually come about.
REFERENCES
Ainlay, S.C. (1986). The encounter with phenomenology. In J.D. Hunter & S.C. Ainlay
(Eds.), Making sense of modern times: Peter L. Berger and the vision of interpretive
sociology (pp. 31-56). New York: Routledge & Kegan Paul.
Allport, G. W. (1929). The study of personality by the intuitive method: An experiment in
teaching from The Locomotive God. Journal of Abnormal and Social Psychology,
24, 14-27.
hart and Winston.
Allport, G. W. (1954a). The hist~rical background of modern social psychology. In G.
Lindzey (Ed.), Handbook of social psychology (1st ed., Vol. 1, p. 3-56). Reading, MA:
Addison-Wesley.
Allport, G. W. (1954b). The nature of prejudice. Cambridge, MA: Addison-Wesley.
Allport, G. W. (1960). Personality and the social encounter. Boston: Beacon Press.
Winston.
Allport, G. W. (1962). The general and unique in psychological science. Journal of Per-
sonality, 30, 405-422.]
Allport, G. W. (1966). Traits revisited. American Psychologist, 21,1-10.
Allport, G. W. (1968). The person in psychology: Selected essays. Boston: Beacon Press.
Allport, G. W., & Vernon, P. E. (1931). A study of values. Boston: Houghton Mifflin.
Antonovsky, A. (1979). Health, stress, and coping. San Francisco: Jossey-Bass.
Antonovsky, A. (1987). Unraveling the mystery of health: How people manage stress and
stay well. San Francisco: Jossey-Bass.
Bandura, A. (1982). The psychology of chance encounter and life paths. American Psy-
chologist, 37, 747-755.
Bern, D. (1983). Constructing a theory of the triple typology: Some (second) thoughts on
nomothetic and idiographic approaches to personality. Journal of Personality, 51,
566-577.
Chiriboga, D. (1978). Evaluated time: A life-course perspective. Journal of Gerontology,
33, 388-393.
Craik, K. (1986). Personality research methods: An historical perspective. Journal of

Crapanzano, V. (1980). Tuhami: Portrait of a Moroccan. Chicago: University of Chicago
Press.
Elder, Jr., G., Liker, J., & Cross, C. (1984). Parent-child behavior in the great depression:
Life-course and intergenerational influences. In P. Baltes & 0. G. Brim, Jr. (Eds.),
Life-span development and behavior (Vol. 6, pp. 111-158). New York: Academic
Press.]
Emmons, R. A. (1989). The personal striving approach to personality. In L. A. Pervin
NJ: Erlbaum.
Epstein, S. (1980). The stability of behavior. II. Implications for psychological research.
American Psychologist, 35, 790--806.
Epstein, S. (1981). The unity principle versus the reality and pleasure principles, or the
tale of the scorpion and the frog. In M. Lynch, A. Norem-Hebeisen, & K. J. Gergen
(Eds.), Self-concept: Advances in theory and research (pp. 27-37). Cambridge, MA:
Ballinger/Harper & Row.
Evans, R. (1970). Gordon Allport: The man and his ideas. New York: Norton.
Freeman, M. (1985). Paul Ricoeur on interpretation. Human Development, 28, 295-312.
Gergen, K. 91982). Toward Transformation of Social Knowledge. New York: Academic
Press.
Gergen, K., & Gergen, M. (1983). Narratives of the self. InT. Sarbin & K. Scheibe (Eds.),
Studies in Social Identity (pp. 245-273). New York: Praeger.
Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of personal his-
tory. American Psychologist, 35, 603-Q18.
Habermas, J. (1971). Knowledge and human interests. Boston: Beacon Press. (Original
work published 1968)
Hogan, D. (1978). The variable order of events in the life course. American Sociological
Review, 43, 573-586.
Hogan, D. (1980). The transition to adulthood as a career contingency. American Socio-
Hogan, D. (1981). Transitions and social change: The early lives of American men. New
Holt, R. R. (1978). Individuality and generalization in the psychology of personality: A
theoretical rationale for personality assessment and research. In R. R. Holt (Ed.),
Methods in clinical psychology: Projective assessment (Vol. 1, pp. 5-29). New York:
Plenum Press. (Original work published 1962)
James, W. (1918). The principles of psychology. New York: Holt. ('1\vo volumes; original
Kagan, J. (1980). Perspective on continuity. In 0. G. Brim, Jr. & J. Kagan (Eds.), Con-
stancy and change in human development. Cambridge, MA: Harvard University
Press.
Kagan, J. (1981). The second year: The emergence of self-awareness. Cambridge, MA:
Harvard University Press.
Kegan, R. (1982). The evolving self: Problem and process in human development. Cam-
bridge, MA: Harvard University Press.
Klein, G. (1976). Psychoanalytic theory: An exploration of essentials. New York: Inter-
national Universities Press.
Kohut, H. (1971). The analysis of the self" A systematic approach to the psychoanalytic
treatment of narcissistic personality disorders (Monograph 1 of the Psychoanalytic
Study of the Child Series). New York: International Universities Press.
Kohut, H. (1973). The psychoanalyst in the community of scholars. The Annual for Psy-
choanalysis, 3, 341-370.
Kohut, H. (1977). The restoration of the self. New York: International Universities Press.
Kohut, H. (1984). How Does Psychoanalysis Cure? (Eds. A. Goldberg & P. Stepansky)
Chicago: University of Chicago Press.
Lasch, C. (1979). The culture of narcissism: American life in an age of diminishing
expectations. New York: Norton.
Little, B. (19&'3). Personal projects: A rationale and method for investigation. Environ-
ment and Behavior, 15, 273-309.
Maslow, A. (1954). Motivation and personality. New York: Harper & Row.
Miller, N., & Dollard, J. (1941). Social learning and imitation. New Haven: Yale Univer-
sity Press.
Piaget, J. (1951). The psychology of intelligence. London: Routledge & Kegan Paul.
Rabinow, P., & Sullivan, W. (1979). The interpretive turn: Emergence of an approach. In
P. Rabinow & W. Sullivan (Eds.), Interpretive social science: A reader (pp. 1-24).
Berkeley: University of California Press.
Ricoeur, P. (1971a). The model of the text: Meaningful action considered as text. Social
Research, 38, 529-562.
Ricoeur, P. (1971b/1991). What is a text? Explanation and understanding. In K. Blarney &
J. B. Thompson (Eds. and Trans.), From text to action: Essays in hermeneutics (Vol.
II, pp. 105--124). Evanston, IL: Northwestern University Press.
Rogers, C. R. (1951). Client-centered therapy; its current practice, implications, and
theory. Boston: Houghton Mifflin.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rosenwald, G. (1988). A theory of multiple-case research. In D. McAdams & R. Ochberg
(Eds.), Psychobiography and life narratives (pp. 239-264). Durham, NC: Duke Uni-
versity Press.
Roth, J. (1963). Timetables. Indianapolis: Hobbs-Merrill.
Rubin, D. (Ed.) (1986). Autobiographical memory. New York: Cambridge University
Press.
Runyan, W. M. (1982). Life histories and psychobiography. New York: Oxford University
Press.
Runyan, W. M. (19&'3). Idiographic goals and methods in the study of lives. Journal of
Runyan, W. M. (1988). A historical and conceptual background to psychohistory. In W. M.
Runyan (Ed.), Psychology and historical interpretation (pp. ~0). New York: Ox-
ford University Press.
Sirgy, M. J. (1986). Self-congruity: 1bward a theory of personality and cybernetics. New
York: Praeger.
Stagner, R. (1937). Psychology of personology. New York: McGraw-Hill.
Stagner, R. (1951). Homeostasis as a unifying concept in personality theory. Psychologi-
cal Review, 58, 5--17.
Stern, W. (1921). Die differentielle psychologie. Leipzig: Barth.
Stewart, A., Franz, C., & Layton, L. (1988). The changing self: Using personal docu-
ments to study lives. In D. McAdams & R. Ochberg (Eds.), Psychobiography and
life narratives (pp. 41-74). Durham, NC: Duke University Press.
Taylor, C. (1985). Interpretation and the sciences of man. In C. Taylor (Ed.), Philosophy
and the human sciences: Philosophical papers (Vol. 2, pp. 15-57). Cambridge: Cam-
bridge University Press.
Wegner, D., & Vallacher, R. (1987). The trouble with action. Social Cognition, 5, 179--190.
Weintraub, K. J. (1975). Autobiography and historical circumstances. Critical Inquiry, 1,
821--848.
Windelband, W. (1904). Geschicte und Naturwissenschaft. Strassburg: Heitz.
Winnicott, D. W. (1965). Ego distortion in terms of the ''true" and "false" self. In D. W.
Winnicott (Ed.), The maturational processes and the facilitating environment
(pp. 140-152). New York: International Universities Press.
CHAPTER TEN
Gordon Allport and

"Letters from Jenny"
DAVID G. WINTER
As editor of the Journal of Abnormal and Social Psychology in 1946,

Gordon W. Allport published "Letters from Jenny" (Anonymous, 1946).
These letters were in fact selections and abridgements from a series of
301 letters written during the years 1926--1937 from "Jenny'' (at the
beginning of the correspondence, a 58-year-old widow) to "Glenn" (the
college roommate of her son "Ross") and his wife "Isabel." Eighteen
years later, Allport republished the letters as a book, adding a retro-
spective discussion by Isabel and interpretations of Jenny according to
existential, depth psychological, and structural-dynamic theory (Allport,
1965). The published letters are about one third of the total correspondence.
In Allport's words, "these letters tell the tragic story of a mother-
son relationship, and trace the course of a life beset by outward frus-
tration and defeat" (Anonymous, 1946, p. 318). During the early years
of the correspondence, Jenny's extreme possessiveness of her son erupted
in periodic suspicions, recriminations, and hatred of any women in his
life. Ross's unexpected death in 1929 (from complications of a mastoid
infection and operation) was a turning point. Thereafter, according to
Isabel's later account, "Jenny's hostility and suspicion turn into full-
fledged paranoia, [while] the more benign qualities seem to diminish"
(Allport, 1965, p. 156).
DAVID G. WINTER • Department of Psychology, University of Michigan, Ann Arbor,

Michigan 48109.
New York, 1993.
147
148 DAVID G. WINTER
To Allport, these letters were both personally fascinating and pro-

fessionally useful as a teaching device:
Speaking for myself I may say that I have found the Letters the
most effective case material I have ever encountered for provoking
fruitful class discussions of theories of personality.... I know of no
other case material so rich and exciting and challenging for those
who like to explore the mysteries of human nature, whether they
be students of psychology or devotees of literature or simply ob-
servers of life. (1965, pp. vi, x)
Two of Allport's students carried out content analysis studies of the
letters (Baldwin, 1941, 1942; Paige, 1964, 1966). In many personality
textbooks (e.g., Hall & Lindzey, 1970, pp. 288-290; Peterson, 1988, p. 292;
Phares, 1988, pp. 50, 256--257, 278; Ryckman, 1989, pp. 226--230, Scroggs,
1985, pp. 147-153), the letters are cited as the major (if not the only)
example of Allport's concern for studying individual persons through
the use of personal documents (Allport, 1942) and "idiographic" (later
"morphogenic") methods (Allport, 1962).
GORDON ALLPORT AND "GLENN"
In publishing the letters, Allport identified Glenn and Isabel as "a

married couple living and teaching in an eastern college town" (Editor's
Introduction to Anonymous, 1946, p. 318). The obvious similarity to All-
port's own life (academic positions at Dartmouth and then Harvard), as
well as the phonetic similarity of "Glenn" and "Isabel" to "Gordon" and
"Ada" (the first name of Allport's wife), naturally led many readers to
wonder whether the letters were actually written to the Allports, but
he repeatedly denied this. As an editor, he was vague about how he
acquired the letters, noting only that Glenn and Isabel "made them
available for editing, for analysis, and for publication" (1965, p. vi).
In fact, Gordon Allport and his wife Ada were "Glenn" and "Isa-
bel," and "Jenny'' actually was the mother of Allport's college roommate.
This can be demonstrated in several ways. First, there is photographic
evidence. According to the Freshman Red Book of the Harvard class of
1919 (Harvard Class of 1919, 1915), Gordon Allport lived in Gore Hall
D-41 (now part of Winthrop House) in 1915-1916. The photograph of
"Ross" published as the frontispiece in Allport (1965) is the same as the
photograph of the only other occupant of that room. Accoding to Har-
vard University records, "Ross" and Gordon Allport also roomed to-
gether in Perkins Hall 30 during 1916-1917, after which Ross joined
the United States Ambulance Service in World War I.
GORDON ALLPORT AND "JENNY" 149
Second, the dates of moves of Glenn and Isabel and the birth of
their first child match the actual details of the Allports' life. Moreover,
all of the facts about "Ross" mentioned in the letters are consistent
with the biographical details of Allport's roommate, as printed in the
class of 1919 reports published by the Harvard Alumni Office: private
preparatory school in Chicago, military service during 1917-1919, and
various merchandising positions in New York up to his death.
In publishing the letters, Allport wrote that he disguised the names
of all places (except New York and Chicago) and persons when he
published the letters (1965, p. vi). For example, Ross is said to have
attended Princeton, and Jenny's family of origin was said to be located
in Montreal. In fact, however, Allport neglected to alter a few minor
details, and these "slips" first suggested to me the real identities of the
correspondents. (1) On August 27, 1926, Jenny wrote ironically that
"This has been a wonderful week-Pres[iden]t Eliot gone-Valentino
gone .... " Charles W. Eliot, president of Harvard University from 1869
to 1909, died on August 22 (Rudolf Valentino, the movie star, died the
next day), but surely this event would be of greater interest to the
mother of a Harvard graduate than to the mother of a Princeton grad-
uate. (2) On a Sunday in November 1929, Jenny wrote:
Last night when coming home ... my eye fell on the evening paper
"Princeton 10-Yale 6." ... How lovely for our beloved College to
win again. Only 1 year ago yesterday Ross drove me to New Haven
to the game. (Allport, 1965, p. 72)
In fact, 10--6 was the score of the Harvard-Yale game in 1929, and it
was the Harvard-Yale game (not the Princeton-Yale game) that had
been played in New Haven the previous year.
Ross's obituary in the Harvard Class of 1919 fourth report (issued
in 1936) was signed by "G. W. A." (obviously Gordon Allport). It in-
cludes the following remembrance, which is both consistent with and
yet quite different from the Ross of the letters:
His classmates will remember [Ross] for his distinguished appear-
ance and bearing, for his Irish love of argument, and for his persua-
sive speech. He took unfailing delight in the gentle pleasures of
social and intellectual comradeship. Always prepared to challenge
uncritical opinions held merely on the strength of tradition, he ex-
erted a maturing influence on his associates such as might be ex-
pected from an instructor but scarcely from a fellow student.
Although vigorous in argument, he never harbored grudges nor
placed undue emphasis upon mere differences in opinion. He had
many friends. (Harvard class of 1919, 1936, p. 189)
150 DAVID G. WINTER
What a delicate task writing this obituary must have been for Gordon
Allport! The occasion called for praising Ross; yet Allport was also
regularly corresponding with Jenny, who kept pouring out her own
ambivalent feelings and bitterness about Ross. In fact, Jenny read and
edited what he wrote. (A draft of this obituary, with annotations in
Jenny's handwriting, is preserved in the Allport papers in the Harvard
University Archives.) The obituary concludes with a simple sentence
that gives scarcely a hint of the Jenny and Ross saga of the Letters:
"He is survived by his mother who resides in New York."
UNDERSTANDING THE CASE OF JENNY
While these details may make a diverting detective exercise, is it

important to know that the Glenn and Isabel of Letters from Jenny
were really Gordon and Ada Allport? Is it ethical to disclose these
facts? Since Jenny, Ross, and all their relatives, as well as Gordon and
Ada Allport, are long since dead, I do not believe that publishing this
information violates any rights of privacy or other ethical standards.*
In fact, I believe that establishing the identity of Glenn adds to our
understanding of Jenny. Knowing that the letters were actually written
to Gordon Allport enhances the value of his editorial comments on the
case. Thus when he writes that "one might label [Jenny] as hysterical,
overprotective, aggressive, asocial, extrapunitive, an isolate, paranoid,
having a character disorder [and] extraordinarily expressive" (1965, p. viii),
or when he suggests that Jenny might have had an "anal character''
(p. 181), a ''personally confused sex-identity'' (p. 183), or ''repressed guilt:'
(p. 184), we may be sure that his comments are based on his personal
acquaintance of 22 years (1915-1937) as well as on his editorial study of
the letters. Allport's quotations from various student analyses of Jenny
(e.g., Allport, 1965, pp. 173, 174, 212, 220) can now be read as having
the imprimatur of his own personal experience.
Yet as published, Letters from Jenny gives only a one-sided view
of Jenny, Ross, and their relationships to Glenn and Isabel. Allport's
editorial role necessarily gave him control of everything we know about
Jenny. For example, we know nothing whatever of Glenn and Isabel's
replies to her letters and how these replies might have affected what
she wrote. In his introduction, Allport claimed that ''what they wrote
* I have retained the pseudonyms "Jenny" and "Ross" because I do not see any reason to
use their real names (although they are given in available records). I have used "Glenn"
and ''Gordon Allport" interchangeably.
to Jenny does not alter the relationship, nor does it affect the flow of
the narrative" (1965, p. viii); but without examining both sides of the
correspondence, we cannot evaluate such a claim.* Even Jenny's side is
incomplete. For "purposes of publication," we are told, "the Letters
have been abridged to approximately one-third their original length."
Baldwin (1941, p. 19) states that originally there were 301 letters, of
which 100 were used in his analysis; but the published book contains at
least portions of over 150 letters. We know little about the selection
process beyond Allport's vague statement that he "made a special ef-
fort to preserve the original proportion of subject matter'' (1965, p. vi). t
Since the unpublished parts of the correspondence are not in Allport's
papers in the Harvard University Archives, we must assume that they
were destroyed. Perhaps scattered references to "Jenny'' will be found
among Allport's other correspondence, teaching notes, or other personal
papers that would shed more light on his feelings about the relationship.
Whatever Allport's reasons for editing, commenting on, and pub-
lishing the letters, we must realize that he was not a detached editorial
observer, but rather a participant in a demanding and difficult relation-
ship. Most readers, for example, find Jenny a difficult and double-binding
correspondent, who demanded time and effort while playing on emo-
tions and guilt. Christmas 1926 was an example. On December 15, 1926,
she wrote, "Ask dear Isabel not to send any Christmas gift this year,
* Interestingly
enough, Allport discussed these issues in his classic work on the use of
personal documents in psychology:
The use of letters in research ... is complicated by the necessity of considering
the personality of the recipient as well as that of the sender, the relationship
existing between the two, and the topics of thought that comprise the exchange
of letters. (1942, p. 108)
t Baldwin gives the most detailed description of how this selection was made:
The material eliminated was the personal information which would identify the
characters, and most of the incidental chatter common in letters. This included
thanks for gifts, answers to questions asked by the correspondents, and similar
material. The remaining material in the letters included narratives of past and
present experiences, comments on Jenny's attitude toward these experiences,
and remarks about the people she met. A fair sample of this material was
included in the selection. The selected set of letters was designed to give the
same impression of Jenny as the original. (1942, pp. 165-166)
There is some confusion about the exact materials used by Baldwin (1941, 1942). Accord-
ing to Allport (1942, p. 46), Baldwin used the complete text of Jenny's letters, from the
beginning through November 2, 1927, a period represented by only 35 whole or partial
letters 1n Letters from Jenny. As noted above, Baldwin stated that he used 100 out of 301
letters.
152 DAVID G. WINTER
but to send a note so I may receive it about Christmas time, in New

York, and so not feel so desperately alone" (Allport, 1965, p. 24). Glenn
and Isabel wisely ignored this request, and so on January 5 Jenny
wrote, "Your Christmas letter and Christmas box were wonderful. One
could never be quite forsaken or alone when one receives such things
on Christmas Eve" (p. 26). Seven years later, she thanked Glenn and
Isabel for a "splendid" St. Patrick's Day box, but then launched into
two paragraphs of criticism: she didn't relish drinking ''lukewarm thick
tomato juice," the sardines should have been in individual tins, and the
cheese was too soft (pp. 116-117).
Through the correspondence, she compelled attention to (if not ac-
ceptance of) her repeated vituperations against Ross, with whom Glenn
and Isabel also corresponded (to judge from the two letters from Ross
excerpted in Allport, 1965, pp. 64, 66). She begged their attention to
her financial affairs and dispensed unsolicited and contradictory advice
on how to live their own lives and how to raise their child. For Allport,
then, publishing the letters and using them as teaching material may
have satisfied motives arising out of this complex relationship.
"JENNY" AND ALLPORT'S THEORETICAL IDEAS
All theories are influenced by the theorist's personality and life

experience. What part did Allport's experiences with Jenny and Ross
play in his own theories? Without a complete study of Allport's life and
times, it is impossible to evaluate how much Jenny and Ross were
actual influences on his thinking versus how much they were merely
effective illustrations of themes and ideas drawn from Allport's own
beliefs, temperament, and external influences. Still, a close reading of
the letters in conjunction with Allport's theoretical writings suggests
some important connections or resonances-if not causal connections-
between his experiences with Ross and Jenny on the one hand and
some of his major contributions to personality theory on the other.
CONFLICT AND MEDIATION
For example, Allport has long been recognized as a major theorist

of prejudice and conflict (see Allport, 1954). From Isabel's recollections,
though, we get a vivid description of his practical efforts to mediate an
awkward and potentially explosive situation:
Glenn attended Ross's funeral and cremation, and spent an uncom-
fortable afternoon trying to keep peace between Jenny and the
fiancee .... On the way to the crematory Glenn sat between Jenny
and the "chip" [Jenny's word for her son's wife] in the limousine.
While the latter was decently silent Jenny kept sending barbed
verbal darts across Glenn to her. (1965, p. 153)
Allport's discretion and sense of tact, familiar to those who (like
the author) had the privilege of studying or working with him, come
through in Isabel's recollection of the time Jenny inquired about mov-
ing to their town:
Our friendship would [have been] strained by an endless series of
misunderstood actions, flare-ups of temper; and she would have no
other target for her hostility than Glenn and myself. We knew that
Jenny needed us, but with such a paranoid personality we knew
also that our only hope (and hers) was to maintain a civilized but
sympathetic distance .... So Glenn wrote a kind but frank reply,
saying that our plans were uncertain, that my health was not too
robust, and in general implying a firm No. We were far from cer-
tain how this ''rejection" would be received. To our surprise and
relief she accepted it without rancor, in fact even with sympathetic
understanding (1965, pp. 153-154).
THE IDIOGRAPHIC APPROACH
One of Allport's major contributions to personality theory was the

idiographic-nomothetic distinction-studying the unique structure of indi-
viduals versus studying general dimensions common to many people-
and his related interest in developing idiographic (or morphogenic) methods
to capture the uniqueness of the individual personality (Allport, 1962). No
doubt the challenge of understanding Jenny-an extraordinary, intracta-
ble, and complex person-resonated with this theoretical emphasis on the
individual case. Further, perhaps, the difficulty of understanding Jenny
probably reinforced his doubts about the possibilities of reducing person-
ality to a short list of universal dimensions (as he perceived Henry Mur-
ray, his contemporary at Harvard, to be attempting; see Murray, 1938).
Allport's own words illustrate this connection:
Psychologists are on safe ground so long as they talk in abstrac-
tions about personality-in-general. Their real test comes when they
attempt to explain (or guide or therapeutically treat) a single con-
crete life. In reflecting on the case of Jenny I find myself wishing
that I could take refuge in vague generalizations, but invariably
she pins me down with the unspoken challenge, "And what do you
make of me?" And so, aware as I am of my audacity, I make bold to
present this edition of the Letters. (1965, p. x)
154 DAVID G. WINTER
Even at the end of his life, Allport wrote about Jenny in his own auto-
biography: "Here surely is a unique life, calling for psychological analy-
sis and interpretation" (Allport, 1968a, p. 403).
Regardless of how much Jenny influenced Allport's idiographic con-
cerns, his description and analyses of Jenny (1965, Chapters 5-9)-
seven or eight major themes, each expressed in adjectives or phrases
chosen both to capture the unique individual characteristics of the per-
son and also to reveal whatever latent unities exist beneath superfi-
cially divergent behaviors-stands as a model of the Allportian method
of personality assessment. His descriptive methods might be compared
to art history (tracing the stylistic development of the individual) rather
than chemistry (analyzing complex substance into a few basic elements).
Allport actually developed such a comparison in a 1938 lecture that was
published in a collection of papers on personality (Allport, 1960) while
Murray explicitly used the chemistry analogy to describe his own ap-
proach (Murray, 1938, pp. 142-143).
THERAPEUTIC SKEPTICISM
Knowing his true role in the Jenny case, we can see in Allport's
editorial discussion of the role of Glenn and Isabel a sense of personal
powerlessness about the possibility of intervention:
What Glenn and Isabel wrote to Jenny did not alter the relation-
ship appreciably. Jenny wanted a pair of sympathetic listeners. Hav-
ing found them she proceeded to pour out the story of her hopes,
jealousies, striving and defeat. For eleven years Glenn and Isabel
listened willingly but made no effective attempt to alter the life-
drama being enacted before them. Indeed, they were powerless to
do so. (Anonymous, 1946, p. 316)
While Allport may have felt powerless to alter Jenny's life, he did write
letters recommending that she be admitted to a church-supported home,
some 18 months after Ross's death.*
Allport's feelings of powerlessness to intervene in Jenny's life stand
in sharp contrast to his committed activism in many other situations.
Yet this powerlessness does echo his professional doubts about the pos-
sibility of deep-seated transformations of character through therapy.
(Allport's own theory of personality, for example, has few explicit ther-
apeutic implications.) In the introduction to the published letters, All-
port asked of Jenny, "Could proper guidance or therapy at an appropriate
*Letters from Allport to D. B. Aldrich, April 26, 1931, Allport papers in the Harvard
University Archives.
time have helped alter the rigid course of her conduct" (1965, p. viii)?
In the end, he was pessimistic: "Glenn and Isabel made a good 'third
ear' as does any helpful psychotherapist. But Jenny was incapable of
advancing in insight because of the tenacity of her temperament and
the set of her world-view'' (1965, p. 218).
NORMAL AND ABNORMAL
Judging by his introduction to the book edition of the letters, All-

port saw Jenny and Ross locked in the classic drama of Jocasta and
Oedipus. Having asked ''why [the letters] should be so stimulating and
pedagogically effective," Allport then supplied his own answer:
Every male reader is himself a son.... Therefore the bitter dilemma
of Ross and his mother often seems to echo the reader's own per-
sonal (but usually milder) problem. Like a Greek tragedy the Let-
ters have a universal appeal. (1965, p. vi)
Yet if Allport believed that Ross and Jenny enacted the universal
Greek tragedy of Oedipus, in his writings on personality he usually
took pains to delimit the universality of the complex to which Freud
gave that name. In discussing interpretative autobiographical essays
written by students in his courses, for example, he argued that Freud-
ian theory had only limited usefulness (limited, as the following quota-
tion suggests, to circumstances that were almost an exact replica of
Jenny and Ross):
Their Freudian interpretations seemed to fit well if and when the
family situation in early life was disturbed. When the father was
absent or ineffectual [Ross's father died before he was born], when
the mother was notably aggressive, when there was deliberate sex
stimulation within the family-in such cases, it seems that the Oedipal
formula provides a good fit, together with all its theoretical accou-
trements of identification, superego conflict, defense mechanisms,
castration threats, and all the rest.
When, on the other hand, the family life is reasonably normal
and secure, a Freudian conceptualization seems forced and artifi-
cial. If we say, by way of rough estimate, that 60 per cent of the
students try a Freudian conceptualization of their own cases, about
10 per cent turn out to be wholly convincing and theoretically ap-
propriate. (1968b, p. 181)
This sharp discontinuity between normal and abnormal is another
familiar feature of Allport's personality theory. His views on the role of
unconscious defense mechanisms, the importance of early experience,
156 DAVID G. WINTER
and the usefulness of projective techniques all follow from this funda-
mental distinction. '1\vo examples illustrate the point:
I am fully aware of my heterodoxy in suggesting that there is, in a
restricted sense, a discontinuity between normal and abnormal mo-
tivation, and that we need a theory that will recognize this fact....
There is still a world of difference, if not between normal and ab-
normal people, then between the healthy and unhealthy mecha-
nisms involved in the development of motivation. (1953, p. 105)
The issue, as we have said, is of the highest importance for person-
ality theory. If we regard all the acquisitions of an adult (his altru-
ism, his ideals, his mature tastes, and his "ought" conscience) as
"secondary," or, as Freud has said, as "transparent sublimations" of
id processes, we have an animalistic view of the nature of normal
adult personality.... The point at issue, we repeat, is the relative
importance of unconscious and conscious functions in forming and
maintaining personality. (1961, pp. 148-149)
Elms (1972) described Allport's view of human nature as "the clean
personality ... reject[ing] psychological data on such unsavory creatures
as rats, children, and neurotics as being largely irrelevant to the under-
standing of the mature personality'' (pp. 630-631). Actually, Allport did
not deny the existence of unconscious wishes that distort reality and
defeat rational striving. Rather, he was concerned to circumscribe their
role, to limit their application, to "segregate" the normal and the abnor-
mal. This is made clear in his final summing up of Jenny:
We shall have to admit that in her personality neurotic processes
took the upper hand. Narrowness, rigidity, inappropriateness marked
her behavior. Compulsively she expressed her anger, having little
tolerance for frustration. Almost always she dwelt on the past, rig-
idly and regressively.... If character neurosis is ''inflexible self-
centeredness" ... then Jenny stands diagnosed .... Toward the end
of her life ... from the neurosis an actual psychosis seems to be
developing. (1965, p. 221)
FUNCTIONAL AUTONOMY
At the theoretical level, Allport tried to reinforce the distinction

between infantile-sexual-pathological and adult-ego-nonnal with the
concept of "functional autonomy" of motives-that the motives that
actually influence our everyday behavior are not (or are not any longer)
derived from original "primitive" or "primary" drives. Functional au-
tonomy is thus a mechanism by which adult personality processes can
be segregated from their infantile roots. More formally, Allport de-
scribed functional autonomy as an "acquired system of motivation in

which the tensions involved are not of the same kind as the antecedent
tensions from which the acquired system developed" (1961, p. 229). For
Allport, functional autonomy was "an essential element to a sound the-
ory of personality" (Evans, 1971, p. 29); yet for all his discussion over
the years, it remains one of his most elusive and controversial concepts
(see the discussion in Evans, 1971, pp. 29-39).
By Allport's own analysis quoted above, Jenny was not "function-
ally autonomous" in the usual sense of the term, for she did not show
the mature, sound, "normal" behavior ordinarily connoted by that con-
cept. Nevertheless, in his original paper on functional autonomy (pub-
lished in the year that Jenny died), Allport illustrated the concept with
a hypothetical example of a mother and child that bears a striking
resemblance to Jenny and Ross:
Many young mothers bear their children unwillingly, dismayed at
the thought of the drudgery of the future. At first they may be
indifferent to, or even hate, their offspring.... The only motives
that hold such a mother to child-tending may be a fear of what her
critical neighbors will say ... a habit of doing any job well, or per-
haps a dim hope that the child will provide security for her in her
old age. ... In later years not one of these original motives may
operate. The child may become incompetent, criminal, a disgrace
to her, and far from serving as a staff for her declining years, he
may continue to drain her resources and vitality. ... She certainly
feels no pride in such a child; yet she sticks to him. (1937, p. 81,
emphases added)
Allport used this example, so vividly reminiscent of Jenny, to illustrate
functional autonomy, in that the mother's persistence is explained not
by an "original" motive of maternality but rather by repeated child-
tending behavior that in later years acts as an acquired motive. Yet
while the behavior has become functionally autonomous, it is not neces-
sarily adaptive, healthy, mature, or wise.
JENNY AND ROSS IN GORDON ALLPORT'S LIFE
What role did the relationships with Jenny and Ross play in All-
port's own experience? Here again, without a complete biography it
is difficult to distinguish influence from illustration, or causation from
resonance. Still, a few themes can be identified as guides to further
exploration.
158 DAVID G. WINTER
JENNY AS MOTHER
As quoted above, Allport's introduction to the book edition of the

letters suggests that he saw Jenny and Ross in universal terms as
mother and son. In some respects, she was also a symbolic "mother''
to Allport. For example, a St. Patrick's Day card sent by Jenny in
his first year at college (preserved in a scrapbook of his college years)
has the following inscription, along with a drawing decoration of four
shamrocks*:
St. Patrick's Day
in the Morning, 1916
To My Dear Mr. Allport,
With warmest St. Patrick's Day greeting from the very Irish
Mother of his hyphenated room-mate.
[J. G. Masterson]
Jenny clearly assumed a maternal relation to Glenn and Isabel. She
usually addressed letters to "My dearest Boy" or "My dearest Boy and
Girl." The published letters begin with Jenny entrusting Glenn with the
final duties of a son (arranging her funeral and winding up her affairs),
having transferred these duties from Ross because she could no longer
"ever again believe one word that left his lips-to ever trust him, or
rely on him, to have the smallest faith in him" (Allport, 1965, p. 14).
Over the years, she freely offered "maternal" advice to Glenn and Isa-
bel on topics such as hair styles, raising children, and traveling.
Did Jenny resemble Allport's own mother? His autobiography con-
tains only one sentence about her: "My mother had been a school teacher
and brought to her sons an eager sense of philosophical questing and
the importance of searching for ultimate religious answers" (Allport,
1968a, p. 379). Does this sound like Jenny? Yes and no; they both seem
to have had a philosophical bent and a "tenacious temperament," but
Jenny's worldview was self-defeating (see Allport's final analysis of Jenny
quoted above).
In an analysis of Allport's famous story of his 1920 encounter with
Freud, Elms (1972) characterized Allport's father as concerned with
"hard work and tight money," two themes with which Allport himself
characterized Jenny's letters (1965, pp. 193-194, 201-202). If hard work
and money concerns also applied to his own mother, then there are
some further points of similarity between the two "mothers."
Allport's privately printed and circulated memoir of his own mother,
entitled The Quest of Nellie Wise Allport (Allport, 1944), suggests that
* Reprinted by permission of the Harvard University Archives.

Nellie (his mother) and Jenny were both similar and different. Like
Jenny, Nellie had a strong philosophical interest (but one that kept
evolving rather than getting mired in the same self-defeating grooves),
a strong sense of autonomy, and great sensitivity to beauty. Early reli-
gious training fostered in both women a sense of "unworthiness." In
later years, however, Nellie's religious views, broadened with the study
of anthroposophy, evolved into a sense of gratitude and serenity, whereas
Jenny remained an embittered skeptic, tormented by self-hatred. Per-
haps as a result, Nellie was generous and therefore generative, never
putting "obstacles in the paths of others" (Allport, 1944, p. 27), while
Jenny's Letters are the story of obstacles between mother and son.
ROSS AS DOUBLE
If Jenny was a "mother" to them both, then Ross and Gordon

Allport were brothers-even closer, perhaps "doubles" (Rank, 1922/1971).
Both were born in the Middle West less than a month apart (Ross on
October 16, 1897, Allport on November 11). Both went east to Harvard.
Both were admitted to college late [as Allport put it, "squeezing through
the entrance tests given in Cambridge in early September;" (1968a,
p. 380)]. They were roommates for two years. In an early letter, Jenny
fondly recalled their first encounter in words that emphasize the
brother-double theme:
Such a lot of things have happened since you, Glenn dear, and Ross
stood in the college office waiting to write on your exam. Tall, thin,
pale boys, the world and life all before you-anxious, tense-a long
time ago. If anyone had said then the day would come when you,
Glenn dear, the pale, slim boy, would be the only protection of the
other boy's mother, you would have been considerably surprised.
(Allport, 1965, p. 20)
World War I
At the end of their sophomore year in the spring of 1917, however,
their paths diverged. The United States entered World War I. Ross
enlisted in the ambulance corps and went off to France. Returning in
1919, he finished college but was never able to settle down and estab-
lish himself. From 1920 until his death in 1929, Ross held and lost jobs
with at least four different companies and even served in the Marines
for a short time. As he wrote to Glenn shortly before his death, "I am
worried about my job which seems shaky, and my life which seems
futile" (Allport, 1965, p. 66). As a result of the war, he had become, as
160 DAVID G. WINTER
Allport put it, a member of the ''lost generation ... distorted, badly
adjusted to old scenes and old ambitions" (Allport, 1965, p. 5).
For Gordon Allport, in contrast, "World War I dislocated my pro-
gram only slightly" (1968a, p. 381). He stayed in college and graduated
with his class in 1919. A year teaching English in Istanbul gave "free-
dom and novelty and [a] sense of achievement" (p. 383). There followed
a fellowship for graduate study at Harvard, a traveling fellowship for
two years in Germany (where he studied with Wertheimer, Kohler, Spran-
ger, Stern, and Werner) and England. With the offer of an instructor-
ship at Harvard and an invitation to develop a course in the new field
of the "psychology of personality," he was clearly launched on a suc-
cessful and distinguished career.
"Bad Son" and "Good Son"

The same divergence was reflected in their changing relationship
to Jenny: Ross became the ''bad" son, while Gordon Allport took his
place as the "good" son. 1\vo quotations from Jenny's letters make the
point:
What you say is sure to be the wise, and best, thing. You are such
a good friend to me that I trust you implicitly-your good judg-
ment, and your kindness. (May 3, 1927; Allport, 1965, p. 39)
Ah! Glenn, my dear, Ross is not a good son, nor is he a decent
fellow. (May 31, 1929; Allport, 1965, p. 65)
In his mother's view, Ross was "sex mad," consorting with a "chip ...
of the flapper type"* (May 31, 1929; Allport, 1965, p. 65); but Gor-
don Allport was faithfully married to "my dearest girl"-"Of course
I know you must be a nice girl, or Glenn would not have chosen
you" (April 27, 1926; Allport, 1965, p. 16). To Jenny, Ross was a
"contemptible cur" (May 31, 1929; Allport, 1965, p. 65), but Glenn and
Isabel were "the decentest persons I know" (March 17, 1926; Allport,
1965, p. 16).
* Baldwin, writing with at least the implicit imprimatur of Gordon Allport, gives the most
vivid description of Ross (and Jenny's view of Ross) during this time:
Ross and his [first] wife were divorced, but Jenny bitterly resented Ross's
interest in other women. She found numerous defects in the character of each
likely candidate for Ross's affection. A number of girls entered the scene, each to
be succeeded by the next in line. Vivien, the last and most prominent, was not
disliked at first but her interest in Ross made her thoroughly hated. (1942,
p. 165; see also 1941, pp. 22-25)
Return of the Double

In Rank's classic analysis, the double is the "detached personifica-
tion of instincts and desires which were once felt to be unacceptable,
but which can be satisfied without responsibility in this indirect way"
(1922/1971, p. 76). Can we find any traces of this "double" relationship
with Ross in Allport's writings? For all of their similarities of background
and early college experience, Gordon Allport and Ross diverged sharply in
temperament, life outcome, and of course relationship to Jenny. Draw-
ing on Rank's analysis, we might speculate that Gordon Allport, raised
on "hard work and tight money," would have envied Ross's easygoing,
fun-loving manner (B. J. Cohler, personal communication, December 23,
1987). Allport's sharp distinction between normal and neurotic function-
ing, often laced (as we have seen) with ''hypothetical" examples and
images suggestive of Jenny and Ross, would then seem to reinforce the
difference between his own life and that of Ross. Of course, his theoret-
ical distaste for the infantile, the sexual, and the neurotic was certainly
consistent with many other features of his background and tempera-
ment, as Elms (1972) suggests (see also Chapter 3, this volume).
Was Ross also feared as well as envied, and the double relationship
denied as well as affirmed? Rank suggested that "in the same phenom-
ena of defense the threat also recurs .... [the double] reappears in su-
perstition as the messenger of death" (1922/1971, p. 86). Freud maintained
that in our unconscious we are all perverse, infantile, and potentially
neurotic. If that is true, then how different would Glenn really be from
"sex-mad" Ross, bound to his mother in a helpless and hopeless strug-
gle? What would be the protection from Ross's fate? On what grounds
would survival be deserved? For Allport, perhaps, a theory of person-
ality that emphasized the discontinuity between normal and abnormal
helped to answer these questions.*
SUMMARY
This chapter has established Gordon Allport's true role in the Let-
ters from Jenny. I have gone beyond this established fact to speculate
*These speculations suggest that Allport's view of Freud would have become more nega-
tive after the beginning of the Jenny correspondence (1926) and especially after the death
of Ross (1929). Such a prediction could be explored through a close reading of Allport's
psychological writings and papers, notes, and professional correspondence between 1921
(when he actually met Freud) and 1937, when he published Persorwlity: A Psyclwlogical
Interpretation.
162 DAVID G. WINTER
about how Allport's role affects our interpretation of the Jenny case
and to suggest ways in which aspects of the case may relate to All-
port's life and personality theory. Of course, these speculations need to
be checked against the facts that will emerge from further biographical
research on Gordon Allport. In a time when personality psychology has
begun to focus on the personalities of its founders, this chapter is a
small contribution to that task.
ACKNOWLEDGMENTS: I am grateful to Abigail Stewart, Alan Elms,

Bertram Cohler, Jeffrey Paige, Robert Hogan, and several anonymous
reviewers for reading the manuscript for this chapter and giving
suggestions. Naturally, they bear no responsibility for any of my
conclusions.
REFERENCES
Allport, G. W. (1937). The functional autonomy of motives. American Journal of Psychol-

ogy, 50, 141-156.
York: Social Science Research Council, Bulletin 49.
Allport, G. W. (1944). The quest of Nellie Wise Allport (revised). Privately printed.
(Available in the Allport papers, Harvard University Archives.)
chiatry, 25, 107-119.
Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley.
Allport, G. W. (1960). Personality: A problem for science or for art? In G. W. Allport,
Personality and social encounter (pp. 3-15). Boston: Beacon Press.
Allport, G. W. (1961). Pattern and growth in personality. New York: Holt, Rinehart &
Winston.
Allport, G. W. (1962). The general and the unique in psychological science. Journal of
Allport, G. W. (1965). Letters from Jenny. New York: Harcourt, Brace.
Allport, G. W. (1968a). An autobiography. In G. W. Allport, The person in psychology
(pp. 376-409). Boston: Beacon Press.
Allport, G. W. (1968b). Crises in normal personality development. In G. W. Allport, The
person in psychology (pp. 171-183). Boston: Beacon Press.
Anonymous. (1946). Letters from Jenny. Journal of Abnormal and Social Psychology,
41, 315-350, 449-480.
Baldwin, A. L. (1941). A statistical method for investigating the structure of personality.
Unpublished doctoral dissertation, Harvard University.
Baldwin, A. L. (1942). Personal structure analysis: A statistical method for investigating
the single personality. Journal of Abnormal and Social Psychology, 37, 163-183.
Elms, A. L. (1972). Allport, Freud, and the clean little boy. Psychoanalytic Review, 59,
627-632.
Evans, R. (1971). Gordon Allport: The rrutn and his ideas. New York: Dutton.
Hall, C. S., & Lindzey, G. (1970). Theories of personality (2nd ed.). New York: Wiley.
Harvard Class of 1919. (1915). Freshman red book. Cambridge, MA: Harvard University.
Harvard Class of 1919. (1936). Harvard Class of 1919 faurth report. Cambridge, MA:
Harvard University Alumni Office.
Murray, H. A. (1938). Explorations in personality: New York: Oxford University Press.
Paige, J. M. (1964). Automated content analysis of "Letters from Jenny." Unpublished
honors thesis, Harvard University.
Paige, J. M. (1966). Letters from Jenny: An approach to the clinical analysis of personal-
ity structure by computer. In P. J. Stone (Ed.), The General Inquirer: A computer
approach to content analysis (pp. 431-451). Cambridge, MA: MIT Press.
Peterson, C. (1988). Personality. San Diego: Harcourt, Brace.
Phares, E. J. (1988). Introduction to personality (2nd ed.). Glenview, IL: Scott, Foresman.
Rank, 0. (1971). The double. Durham: University of North Carolina Press. (Originally
published 1922)
Ryckman, R. M. (1989). Theories of personality (4th ed.). Monterey, CA: Brooks/Cole.
Scroggs, J. R. (1985). Key ideas in personality theory. St. Paul, MN: West.
CHAPTER ELEVEN
Allport's Personal Documents

Then and Now
LAWRENCE S. WRIGHTSMAN
It was my pleasure to spend a rather extensive amount of time with

Gordon Allport during the spring semester of 1966, when he was a
visiting fellow at the East-West Center in Honolulu and I was a faculty
member at the University of Hawaii. He was a gentle and gracious
man who was willing to share his past experiences and to facilitate the
development of younger colleagues. I remember fondly that he volun-
teered to :till out an attitude scale I had recently constructed (the Philo-
sophies of Human Nature Scale), and he even scored his own scale responses.
Twenty-five years later, I still deeply value my association with him.
The very first words in the Preface to Allport's Personality were:
"As a rule, science regards the individual as a mere bothersome acci-
dent" (1937, p. vii). Allport's science, of course, was just the opposite. It
placed the individual at the very focus of investigation.
OVERVIEW
My goal in this chapter is to compare the status of personal docu-

ments 50 years ago and now. Allport's (1942) monograph, The Use of
Personal Documents in Psychological Science, published five years after
his Personality text, served as a beacon for those interested in the use
LAWRENCE S. WRIGHTSMAN • Department of Psychology, University of Kansas, Law-

rence, Kansas 66045.
New York, 1993.
165
166 LAWRENCE S. WRIGHTSMAN
of autobiographies, memoirs, diaries, collections of letters, and similar

materials as raw material for the analysis and interpretation of person-
ality and behavior.
Within a decade of Allport's monograph, there had been a fertile
development of the personal documents approach. Charlotte Buehler
(1935), Else Frenkel-Brunswik (Frenkel, 1936; Frenkel-Brunswik, 1939),
John Dollard (1935), Henry A. Murray (1938), Jerome Bruner (Allport
et al., 1941), Alfred Baldwin (1940, 1942), and other prominent psychol-
ogists used diaries, life histories, and other autobiographical material to
understand responses to catastrophe, mechanisms of self-deception, and
changes over the life cycle, among other matters. After World War II,
interest dwindled. There are exceptions, of course; Robert White's (1975)
continuing interest in the growth of personality; the intensive case analy-
ses of a small number of adult males by M. Brewster Smith, Bruner,
and White (1956) done in an attempt to understand how opinions are
developed and maintained; and the analysis of Theodore Drieser's writ-
ings by Seymour Rosenberg and Russell Jones (1972) in order to un-
derstand Dreiser's implicit personality theory.
Several years ago (Wrightsman, 1981), I argued that the "state of
the art" regarding the use of personal documents in psychology had
not-until shortly before the 1980s-advanced beyond that summarized
in Allport's 1942 monograph. Happily, I can report my impression that
the "rebirth" in the use of these materials that was beginning to emerge
in the late 1970s has continued and increased.
In looking retrospectively at 50 years of the use of personal docu-
ments, I will attempt to compare the purposes, the theoretical perspec-
tives, and the methods then and now. As space permits, I will provide
some recent and current examples; I regret that I will not have the
space to mention a greater number of the excellent contemporary examples.
In reviewing Allport's perspective, I have primarily relied on three
of his writings: his Personality text, the previously mentioned SSRC
monograph, and his book, Letters from Jenny (1965). (In actuality, he
had little to say about such materials in the book that has generated
this volume; "personal documents" is not in the index, and there is only
a brief section on "personal records.")
THE PURPOSES OF PERSONAL DOCUMENTS IN

PSYCHOLOGICAL RESEARCH
Allport saw personal documents as materials that can tell us "what

goes on in people's minds" (1942, p. vii). He defined the personal docu-
ALLPORT'S PERSONAL DOCUMENTS 167
ment as " ... any self-revealing record that intentionally or unintention-

ally yields information regarding the structure, dynamics, and function-
ing of the author's mental life" (1942, p. xii). He limited the term to
first-person documents; this restriction has continued. Allport was well
aware of the problems in the representativeness of the single case and
the problems in the validity of subjective records (1944, p. xi). But, he
wrote, " ... the psychologist ... must ask even of the deceptive and trivial
documents why they were written and, further, why they are dull or
deceptive" (1942, p. xiii).
Allport observed that the ways of using personal documents varied
considerably. Chapter 3 of his SSRC monograph lists and annotates 21
different purposes for use of personal documents. Table 1 reprints these,
using Allport's labels and providing his definitions and some examples
of his and some of mine. Allport would acknowledge that some of these
purposes overlap with each other.
Looking at these purposes through the filter of the 50 intervening
years, we can note several salient characteristics of this categorization:
1. It is a surprising hodgepodge of purposes, in the sense that
some of these reflect anticipated benefits to the author of the docu-
ment, while most deal with the value of personal documents for the
social scientist or the psychotherapist.
2. Some of the uses are now routine ones, such as analyzing collec-
tions of letters in order to understand the personality of the letter
writer or asking a psychiatric patient to write a brief autobiography. In
contrast, other purposes identified by Allport have seemingly been ig-
nored; for example, Number 19, as a source for generating question-
naire items, has not often been utilized (Richard Christie's Machiavellianism
scale being an exception).
3. Some of these purposes lead to creative ideas, based on unusual
types of personal documents. Number 18 suggests that exchanges of
letters could be used to study the nature of friendship or marriage. All-
port, in 1942, saw this resource as almost overlooked by social psychologists.
Extending this idea, a specific type of personal document-an ad-
vertisement in a lonely hearts column of a newspaper-has recently
been used by social psychologists to study self-presentation and social
needs. In one study, Harrison and Saeed (1977) content analyzed more
than 800 such advertisements, focusing on differences in the ads placed
by women and men. Women were more likely to describe themselves in
terms of physical attractiveness and men were more likely to specify
this as a want. Men more often mentioned financial security as an attri-
bute of theirs; women were more likely to seek it. Deaux and Hanna
TABLE 1. Allport's List of Purposes for

Psychologists' Use of Personal Documentsa
1. Phenome1Wlogical investigations
DEFINITION: ''interest in complex phenomenal states" (p. 37). [Allport's famous
quotation (p. 37): "If we want to lmow how people feel: what they experience
and what they remember, what their emotions and motives are like, and the
reasons for acting as they do-why not ask them?"]
EXAMPLE: Galton's accounts of the imagery of his correspondents.
2. The study of religious experience
EXAMPLE: Several investigators (Hoffding, Clark, etc.) have analyzed diaries and
autobiographies from the point of view of religious experience.
3. Study of the psychological effects of unemployment
EXAMPLE: Zawadski and Lazarsfeld (1935) analyzed autobiographies of unemployed
writers in Poland, leading to a grouping into four types.
4. Mental life of adolescents
RATIONALE: G. Stanley Hall and others have argued that personal documents are
the best means of studying adolescence because the "experiences peculiar to
adolescence are inaccessible to adults whose later encounters with love and life
have the effects of recasting to tally the nascent and turbulent groping of
adolescents in their struggle to come to terms with physical reality and social
responsibility'' (Allport, 1942, p. 39).
EXAMPLE: Norman Kiell, in 1964, collected a set of autobiographical materials
prepared by adolescents in an effort to demonstrate that the internal and
external agitations of the adolescent are present in every part of the world and
hence only partly determined by culture.
5. Didactic uses
RATIONALE: It has been claimed that practice in writing self-reports increases
insight, powers of observation, and self-control in adolescents.
6. Practical use of experience records
RATIONALE: Social progress may result from the analysis of vivid stories about one's
personal experiences.
EXAMPLE: Clifford Beers's A Mind That Found ltself(1928).
7. Autoanalysis
PURPOSE: Autobiographical outpourings that aim at catharsis may be useful as
teaching devices or as an aid in evolving a theory of personality.
EXAMPLE: W. E. Leonard's Locomotive God (1927).
8. Historical diagnoses
PURPOSE: To shed light on the personalities of writers, artists, and other gifted
people.
EXAMPLE: Bragman's studies of Rossetti (1936); Squires's study of Dostoevsky
(1937).
9. Suwlement to psychiatric examination
PuRPOSES: Besides providing new leads for diagnosis, there may be a therapeutic
value, "initiating and helping to guide the course of treatment'' (p. 44).
10. The subject's verification and validation
PURPOSE: A "rebuttal" by a subject to another's analysis of him or her.
EXAMPLE: John Dewey (1904) responded to his expositors and critics in a series
entitled The LiiYrary of Living Philosophy.
11. Mental effects of special physical conditions

PURPOSE: Autobiographical materials may help ''to keep the influence of physical
factors in perspective" (p. 45).
12. Light on creative processes and the nature of genius
PURPOSE: To study creativity.
EXAMPLE: Willa Cather's (1927) account of her conception and writing of Death
CO'YMs for the ArchbiskcYp.
13. The psychologizing of the social sciences
PURPOSE: The application of "psychohistory" and "psychobiography'' to historical or
cultural phenomena. (These terms were generated later than Allport's 1942
monograph.)
14. The psychologizing of literature
PURPOSE: The probing of motivation by literary critics and biographers.
EXAMPLE: Robert Sears's (1979) comparison of Mark Twain's letters and his novels,
to identify periods of depression in his life.
15. Illustration
PuRPOSE: "Perhaps the commonest use of documents is to provide illustrative
material for authors who wish to exemplify some generalization already in
mind" (p. 48).
16. Induction
PURPOSE: To derive general principles from raw material or particulars.
EXAMPLE: Charlotte Buehler's Lebenspsychologie, based on 200 life histories.
17. Occupational and other types
PURPOSE: The derivation of types, or clustering cases according to similarities.
EXAMPLE: Donley and Winter's (1970) scoring of U.S. presidential inaugural
addresses in order to classify them on strength of achievement, afflliative, and
power motives.
18. Interpersonal relations
PURPOSE: "The possibility of using exchanges of letters between two persons as a
means of studying the dyadic relations of friendship, of marriage, of the parent-
child bond seems almost overlooked by social psychologists" (p. 50).
19. First step in the construction of tests and questionnaires
PURPOSE: To provide insights for generating items in standardized tests and
questionnaires.
20. Reinforcement and supplementation
PURPOSE: "Often the personal document merely falls into place as one of several
methods in a battery. It serves no other purpose than adding credibility to the
total picture developed through interviews, tests, ratings, institutional reports,
or other methods" (p. 51)
EXAMPLE: Thomas and Znaniecki's study of Polish peasants (1918).
21. Methodological objectives
PURPOSE: Social scientists may use personal documents "simply in order to find out
how they may be used to the best advantage" (p. 51). There is an allusion here
to theory development; focus is not so much on the individual or on general
laws of behavior, as it is on ''the process by which the significance of behavior
becomes known and evaluated" (p. 51).
EXAMPLE: Allport's derivation of an empirical-intuitive theory of understanding
from the reactions of students in reading Leonard's Locomotive God.
aFrom Allport (1942, Chapter 3). All quoted material is from this source.
(1984), using a similar set of documents, followed up by analyzing the

content of lonely hearts advertisements placed by homosexual and het-
erosexual men and women. Deaux and Hanna concluded that both the
advertiser's gender and his or her sexual preference were related to
both self-identified attributes and needs in a companion.
4. Another observation, based on Allport's list of purposes, is the
surprising-at least to me-lack of emphasis on theory testing in Table
1. In contrast, I would expect that contemporary psychologists, if asked
about the value of personal documents, would list as the premiere value
their clarification of theory.
For example, Swede and Tetlock (1986) note that controversy sur-
rounds many aspects of the construct of implicit personality theory.
How many basic dimensions underlie an individual's trait perceptions?
Is it appropriate to seek a dimensional representation of implicit theo-
ries of personality? Not only is there disagreement over the answers to
these questions, but also theories and models that provide a good fit
with regard to group data may not be supported by data from separate
individuals (Kim & Rosenberg, 1980). Swede and Tetlock (1986) point to
the idiographic analyses of individual respondents as a way of produc-
ing inductive generalizations on this topic.
5. Finally, I detect a shift in a direction hinted at with regard to
the accuracy of personal documents. As I have indicated, Allport was
primarily concerned with reliability and deception (whether conscious
or unconscious). But he also asked why-why deceptive? The shift has
been even more toward the process involved; for example, what were
the determinants of the process of writing the personal document or
narrating an event? Also, what are the perceptual and interpretive
processes of the social scientist who analyzes and interprets the docu-
ments (Items, 1981)? Allport was concerned that oversimplification and
arbitrary conceptualizations could limit the value of personal documents.
Now, however, there is
an acceptance ... that these filtering processes are an inherent and
meaningful part of any communication about the world. A scientist
or humanist should not seek to eliminate the filters but rather be
conscious of them and account for them as a part of the research or
creative process. (Items, 1981, p. 21)
THEORETICAL PERSPECTIVES-
ALLPORT'S AND THE PRESENT
A second contrast between Allport's work and the present deals
with theoretical perspectives. Allport himself, in Letters from Jenny
(1965), carried out one of the most visible and extensive analyses of
personal documents. Between the ages of 58 and 70, Jenny Gove Mas-
terson wrote more than 300 letters to two young friends, a married
couple who were living in a nearby Eastern college town. Her friend-
ship with the young man extended back to the time-a decade before-
when he had been a college roommate of Jenny's son, Ross. The exchange
of letters began in March 1926 and continued without interruption-an
average of a letter exchanged about every two weeks-for 11:1,.2 years,
until Jenny's death in October 1937.
Allport, after reprinting excerpts from most of the letters, analyzed
Jenny's personality from three theoretical perspectives: existential (or
phenomenological), depth (or psychoanalytic), and structural-dynamic.
Allport carried out his interpretations in the early 1960s before psy-
chology had felt the impact of dialectic thinking.
We now have other and, I believe, more appropriate models for
interpreting sets of letters (Wrightsman, 1988). Different theories of
personality seem congenial with different types of personal documents.
Autobiographies and memoirs are retrospective, and often there are
significant acts of filtering, reframing, and justifying within them; their
analyses seem most congenial with deterministic theories such as psy-
choanalysis or social-learning theory. In contrast, diaries and collections
of letters are more relevant to a dialectic theory, because they repre-
sent an ongoing, but constantly changing, production of raw material.
My reading of Jenny Masterson's letters leads me to conclude that
what is missing from Allport's analyses is a focus on the dynamic, con-
stantly changing relationship between Jenny and her son, Ross, who
was clearly the most important person or object in her worldview. Jenny's
feelings shift from trust and love to distrust and revulsion and back
again. We could employ various ways of labeling the dialectic that is
operating within Jenny: a tug between trust and distrust of Ross; pulls
between insistence on her financial independence and her reliance and
dependence on Ross and others; vacillations between realistic planning
and helpless despair. Thus can contemporary theory enrich the inter-
pretations of personal documents.
METHODOLOGIES-ALLPORT'S VERSUS THE PRESENT
Obviously the greatest difference between the status of personal

documents in Allport's time and now is not with regard to purposes,
theory testing, or theoretical perspective-it is in regard to methodol-
ogy. The growth of sophisticated statistical analyses and high-speed
computers has led to a massive shift toward more fine-grained analyses
carried out on the data generated from personal documents. Table 2
TABLE 2. Contemporary Types of Studies Using Personal Documents

1. Rosenberg and Jones (1972): The implicit personality theory of Theodore Dreiser
In a study whose methodology has served as a guide for subsequent work, Seymour
Rosenberg and Russell Jones applied content analysis and factor analysis to determine
the implicit personality theory of one person: Theodore Dreiser, the author of Sister Car-
rie and An American 'Pragedy. Dreiser was chosen because his book, A Gallery of Women,
published in 1929, contained detailed character descriptions. The book consists of 15 differ-
ent stories, each a portrayal of a different woman known by Dreiser. Thus, each person is
described in 20 to 50 continuous pages, and a variety of people are so described. By list-
ing each character mentioned in any of these stories and each trait ascribed to him or her,
Rosenberg and Jones identified 241 characters and 6761 descriptive units from the book.
Statistical analysis sought to determine what traits clustered together. (It is trite but
true to observe that these analyses could not have been completed when Allport's 1942
monograph was published.) Results indicated that traits associated with women were
quite different from those associated with men. Furthermore, the dimension of
conformity versus nonconformity emerged as an important construct in Dreiser's
descriptions of people, but conformity was not associated only with the female sex, as the
usual stereotype would have predicted.
The detailed, quantitative analysis of literary and other written materials has been
used in some of Rosenberg's more recent work and served as a model for Swede and
Tetlock's (1986) recent analysis of Henry Kissinger's implicit personality theory.
2. DeWaele and Harre (1979): Assisted autobiography as a psychological method

In a very useful, original, and detailed book chapter, DeWaele and Harre described
procedures for the creation of an "assisted autobiography," which is "really a continuous
process of negotiated autobiographical reconstruction" (1979, p. 193). Then they provide a
means for analysis of the contents of the autobiography. The chapter includes a 15-page
appendix that serves as a checklist of various aspects of the content of the life story.
Although the chapter is not relevant to advances in statistical analysis, it provides a
"fresh look" at methodology in the broad sense and reflects a humanistic perspective that
assumes that research is a negotiated cooperative interaction between the investigator
and the research subject.
3. Tetlock (1981): A comparison of preelection and postelection statements by twentieth-

century U.S. presidents.
Philip Tetlock coded public statements and addresses by presidents from McKinley
through Carter for integrative complexity. Statements made prior to election were
compared with statements made in the first month, the second year, and the third year
in office. In addition to repeated-measures analyses of variance, Tetlock used (more
conservative) quasi-F-ratios to evaluate the data, by considering some effects (time
period and paragraph sampling unit) as random, not fixed, effects. Results indicated that
except for Herbert Hoover, presidential policy statements became more complex after
assuming office.
Similarly, Suedfeld and Rank (1976) demonstrated that the long-term success of revo-
lutionary leaders was related to the cognitive complexity of their public statements. Equally
important is the content analysis of speeches and interviews of Soviet Politburo members,
to aid in understanding their personal characteristics, done by Margaret Hermann (1980).
She has also applied this content analysis to President Reagan's public statements.
4. Bennan (1985): Analysis of the diary of an older person

Harry J. Bennan of Sangamon State University, Springfield, Illinois, is developing a
methodology for examining the diaries and journals of older people, in order to
understand how they deal with the events of late adulthood. The analysis is driven by
concepts from psychoanalytic theory and Daniel Levinson's stage theory of adult
development (1978). This is a useful approach in the sense that the experiences of older
people are distinct from those of younger ages.
5. Stewart and Healy (1985): The study of adaptation to stress and life change
The explosion of interest in adult psychological development that we have witnessed
in the last decade has given renewed interest to the study of autobiographies. Daniel
Levinson and his colleagues (1977) elicited autobiographies from their subjects as part of
their investigation of stages in men's development. Abigail Stewart and Joseph Healy
have analyzed autobiographies in identifying reactions to stressful events and adaptation
to life changes.
lists a brief sampling of representative procedures in contemporary

research. It is probably irresponsible to try to pick only a few exam-
ples, and even more so, considering that I have tried here to consider
"methodology" in the broadest sense-not just data collection and anal-
ysis procedures-but the relationship of researcher to subject, the im-
beddedness of method in theory, and the use of a variety of materials
beyond the classic letters, diaries, and autobiographies.
CONCLUSION
What would Allport say if he were suddenly revived and confronted

with the breadth of use of personal documents in contemporary psy-
chology and other social sciences? He would be gracious, of course,
although I imagine he would feel regret that there was such a long
latency before his 1942 monograph had much impact. He would wel-
come the increase in number and variety of studies in the last decade.
He would find it gratifying that in 1986, the Social Science Research
Council, which sponsored his 1942 monograph, announced its plans to
appoint a Committee on Personal Testimony (Crapanzano et al., 1986).
And he would applaud the fact that the American Council of Learned
Societies was co-sponsoring this new committee, because in his time
the barriers between the social sciences and the humanities were less
formidable than they became in the 1960s and 1970s. [There are indica-
tions that more interdisciplinary cooperation is occurring now, accord-
ing to Crapanzano et al., (1986).] I think he would be amazed and a
little disturbed by the profusion of statistical methods-and especially
the cost in carrying them out. (He once told me that every piece of
research he had completed was unfunded, with the exception of the
information transmission research he did for the government during
World War II.) He might be surprised that some fundamental prob-
lems, such as how to generalize knowledge from a single case, have still
not been systematically examined. I believe he would urge us, in his
gentle way, not to forget the whole person as we apply our new and
sophisticated analyses to the investigation of personal documents.
REFERENCES
hart and Winston.
York: Social Science Research Council.
Allport, G. W. (Ed.) (1965). Letters from Jenny. New York: Harcourt Brace, and World.
Allport, G. W., Bruner, J. S., & Jandorf, E. M. (1941). Personality under social catas-
trophe: An analysis of German refugee life histories. Character and Personality,
10, 1-22.
Baldwin, A. L. (1940). The statistical analysis of the structure of a single personality.
Psychological Bulletin, 37, 518--519.
Baldwin, A. L. (1942). Personal structure analysis: A statistical method for investigating
the single personality. Journal of Abnormal and Social Psychology, 37, 163-183.
Beers, C. W. (1928). A mind that found itself(5th ed.) New York: Doubleday.
Berman, H. J. (1985, March). On the brink: Elizabeth Vining's "Being Seventy." Unpub-
lished paper presented at conference on "Humanistic Perspectives on the Aging
Enterprise in America," Center for the Study of Aging, University of Missouri,
Kansas City.
Bragman, L. J. (1936). The case of Dante Gabriel Rosetti. American Journal of Psychia-
try, 92, 1111-1122.
Buehler, C. (1935). The curve of life as studied in biographies. Journal of Applied Psy-
chology, 19, 405-409.
Cather, W. (1927). A letter from Willa Cather: A short account of how I happened to
write Death comes for the Archbishop. Commonweal, 7, 713-714.
Crapanzano, V., Ergas, Y., & Modell, J. (1986, June). Personal testimony: Narratives of
the self in the social sciences and the humanities. Items, 40(2), 25-30.
Deaux, K., & Hanna, R. (1984). Courtship in the personal column: The influence of gen-
der and sexual orientation. Sex Roles, 11, 363-375.
Dewey, J. (1904). The library of living philosophy: University of Chicago Press.
DeWaele, J.-P., & Harre, R. (1979). Autobiography as a psychological method. In G. P.
Ginsburg (Ed.), Emerging strategies in social psychological research (pp. 177-224).
New York: Wiley.
Dollard, J. (1935). Criteria for a life history. New Haven, CT: Yale University Press.
Donley, R. W., & Winter, D. G. (1970). Measuring the motives of public figures at
a distance: An exploratory study of American presidents. Behavioral Science, 15,
227-236.
Frenkel, E. (1936). Studies in biographical psychology. Character and Personality, 5,

1~5.
Frenkel-Brunswik, E. (1939). Mechanisms of self-deception. Journal of Social Psychol-
ogy, 10, 409-420.
Harrison, A. A., & Saeed, L. (1977). Let's make a deal: An analysis of revelations and
stipulations in lonely hearts advertisements. Journal of Personality and Social Psy-
chology, 35, 257-264.
Hermann, M. G. (1980). Assessing the personalities of Soviet Politburo members. Person-
ality and Social Psychology, 6, 332-352.
Items (1981, June). The uses of personal testimony. 35(1/2), 20--21.
Kiell, N. (1964). The universal experience of adolescence. New York: International Uni-
versities Press.
Kim, M. P., & Rosenberg, S. (1980). Comparison of two structural models of implicit
personality theory. Journal of Personality and Social Psychology, 38, 37~89.
Leonard, W. E. (1927). The locomotive god. New York: Century.
Rosenberg, S., & Jones, R. A. (1972). A method for investigating and representing a
person's implicit theory of personality: Theodore Drieser's view of people. Journal of
Personality and Social Psychology, 22, 372--il86.
Sears, R. R. (1979, June). Mark Twain's separation anxiety. Psychology Today, pp. 100--
104.
Smith, M. B., Bruner, J. S., & White, R. W. (1956). Opinions and personality. New York:
Wiley.
Squires, P. C. (1937). Fyodor Dostoevsky: A psychopathological sketch. Psychoanalytic
Review, 24, 36~5.
Suedfeld, P., & Rank, A. D. (1976). Revolutionary leaders: Long-term success as a func-
tion of cognitive complexity. Journal of Personality and Social Psychology, 34, 169--
178.
Swede, S. W., & Tetlock, P. E. (1986). Henry Kissinger's implicit theory of personality: A
quantitative case study. Journal of Personality, 54, 617-646.
Stewart, A. J., & Healy, J. M. (1985). Personality and adaptation to change. In R. Hogan
& W. Jones (Eds.). Perspectives on personality: Theory, measurement, and interper-
sonal dynamics (pp. 117-144). Greenwich, CT: JAI Press.
Tetlock, P. E. (1981). Pre- to post-election shifts in presidential rhetoric: Impression
management or cognitive adjustment? Journal of Personality and Social Psychol-
ogy, 41, 207-212.
Thomas, W. 1., & Znaniecki, F. (1918). The Polish peasant in Europe and America. Bos-
ton: Gorham Press.
White, R. W. (1975). Lives in progress: A study of the natural growth of personality (3rd
ed.). New York: Holt, Rinehart and Winston.
Wrightsman, L. S. (1981). Personal documents as data in conceptualizing adult personal-
ity development. Personality and Social Psychology Bulletin, 7, 367~85.
Wrightsman, L. S. (1988). Personality development in adulthood. Newbury Park, CA:
Sage.
Zawadski, B., & Lazarsfeld, P. (1935). The psychological consequences of unemployment.
Journal of Social Psychology, 6, 224-251.
CHAPTER TWELVE
Conceptions of Self and Identity

A Modern Retrospective on Allport's View
ROY F. BAUMEISTER
My particular task for this volume is to comment on the status of work

on self and identity in relation to what Allport had to say. I take as my
point of departure Allport's (1943) paper entitled "The Ego in Contem-
porary Psychology," for it serves as a clearer and more thorough state-
ment of his views on the self than does the 1937 text that was the
occasion for our symposium.
ALLPORT'S VIEWS IN RETROSPECT
In reading Allport's wise and insightful comments about the self

from nearly half a century ago, one is struck by how many of his com-
ments are still useful and germane. This is not to say, however, that
nothing has been learned in the past half-century. I wish to comment on
six of Allport's points about the self.
First, he predicted an increase in the psychological study of the
self. Writing in 1943, when the psychology of self was just getting
started, he said "we may safely predict that ego-psychology will flour-
ish increasingly" (p. 476). There is little doubt but that his prediction
has been confirmed. The decade of the 1940s saw the psychoanalytic
movement take new notice of the ego, resulting in an expanded appre-
RoY F. BAUMEISTER • Department of Psychology, Case Western Reserve University,

Cleveland, Ohio 44106.
New York, 1993.
177
178 ROY F. BAUMEISTER
ciation of its importance and capabilities. Each decade since has added
its views and its data, as not only psychology but indeed all the social
sciences and humanities became fascinated with the nature of selfhood
and identity.
Second, it is remarkable to note that Allport was willing to at-
tempt a comprehensive summary of self, listing if not integrating all
major aspects of it that psychology had found. A similar comprehensive
inventory is desperately needed today. Our theories and findings about
self have expanded all out of control, and few attempt to provide an
overview or even to fit their data into a comprehensive theory of the
self. For a great scholar like Allport to attempt such an overview in the
1940s was intellectually audacious; for anyone to attempt it today, with-
out years of work, is almost absurd. Yet in a sense it fails to do justice
to Allport's memory for us not to try. Later in this chapter, I shall
attempt to provide at least an outline of what such a comprehensive
theory of self would require today.
Third, one notes that Allport sometimes confused self and person-
ality. This confusion persists today; it is hard to draw the line. Most of
us would say that the self is not the same thing as the totality of
personality; perhaps the self should be considered part of personality.
All personality traits do not belong in the self. One's score on extraver-
sion, for example, is not a measure of self. The reason for the confusion
between self and personality has to do with the self-concept, which
does potentially include all one's traits--or at least one's beliefs about
one's traits. A recent work defined the self as "a person's mental repre-
sentation of his or her own personality" (Kihlstrom et al., 1988, p. 146).
While this definition leaves out interpersonal aspects, social identity,
the decision-making aspect of self, and perhaps a few others, it does
effectively call attention to the relation between self-concept and per-
sonality. Self includes one's ideas about one's personality.
Fourth, Allport had fairly little to say about identity. Erik Erikson
had not yet coined the term identity crisis. Writing in 1943, in the
middle of World War II, Allport did not see the Americans as obsessed
with their identity crises, with finding themselves, and so forth. Since
then, this has become an area of major focus. Identity, as a socially
constructed definition of self, has been an increasingly important fea-
ture of self research over the past several decades. In Allport's day, the
self was understood primarily in terms of the psychodynamic ego, but
later generations of scholars have come to appreciate the need to aug-
ment the notion of ego with an understanding of cultural factors, inter-
personal processes, and so forth.
Fifth, Allport emphasized the unity of self. The self is indeed expe-
CONCEPTIONS OF SELF AND IDENTITY 179
rienced as a unity. Allport spoke of the self as a unifying principle in

personality.
Modern research has lost sight of this aspect of the self to some
degree, and today's researchers might well benefit from being reminded of
Allport's discussion of this principle. Researchers today speak of vari-
ous aspects of self-concept, of the dimensionality of self-esteem, of indi-
vidual self-schemas for each trait, and other matters, as if these are
fairly independent. The focus is on the parts and aspects of the self. Too
little attention is paid to the unifying aspect of the self as a whole.
An important example concerns the notion of potentialities. Heideg-
ger's existential concept of potentiality was recently revived by Markus
and Nurius (1986) to show that people have different conceptions of
future possibilities for themselves. Calling these "possible selves," how-
ever, has misled some readers to think that it refers to a multiplicity of
selves, concealing the vitally important fact that it is in every case the
same self that is at issue. A person can imagine him- or herself in
various catastrophic fashions-as someday being a criminal, or a drug
addict, or not getting tenure-but this is quite different from imagining
someone else in those circumstances. The emotional power and force of
possible selves derives from the fact that it is oneself who is imagined
thus. In other words, it is not really a matter of different or multiple
"possible selves," but rather one's same self with different attributes,
different definitions, or different circumstances.
Probably a major reason for the modern loss of the sense of the
unity of self is the declining interest in morality, both socially and in
psychological theory. In moral action, the self participates as a unity;
the whole individual is implicated in moral action. The declining inter-
est in moral issues removes this unifying force from prevailing concep-
tions of selfhood. In its place, the new view of humanity regards the
individual as a bundle of situational responses, with minimal continuity.
Sixth, Allport noted that the most important motivation associated
with the self is the desire to maximize self-esteem; that is, to avoid loss
of esteem and if possible to increase it. The motivational importance of
self-esteem is as strong and widespread in today's psychology as it was
back then. The only difference is that today we cite different references
(and, to be sure, we have accumulated a great deal more evidence), but
it is the same effect.
Today, however, we would have to augment this motivation with at
least one other major motivation of the self, namely control (also known
as choice or personal freedom). Abundant evidence has associated the
self with control. The self seeks to gain, maintain, and exercise control.
This general principle has been shown in many contexts, ranging from
the importance of choice in cognitive dissonance (e.g., Linder et al.,

1967; on the self, see Baumeister & Tice, 1984), to reactance theory
(Brehm, 1966; also Baer et al., 1980), to systematic efforts to distort
feedback so as to preserve views of self as having control (Greenwald,
1980; Taylor & Brown, 1988). Moreover, both objective and subjective
control are sought. That is, even when the self cannot really be in
control, it wants to think it is. Hence Langer's (1975) work on the
illusion of control or Alloy and Abramson's (1979) evidence that non-
depressed people overestimate their degree of control.
ALLPORT'S INTEGRATIVE PROJECT
In 1943, Allport was willing to sketch an overview of what was

known about the ego. It is worth asking what such an outline would
look like if someone were to attempt it today.
I would suggest three major headings as necessary to cover what
is now known about the self: the natural self, the conceptual self, and
the action self. I am not proposing a new theory of self here; rather, I
am trying to say what categories would be needed to integrate and
make sense of all the knowledge that has been accumulated about the
self. At best, this is (as I suggested earlier) an outline of what a com-
prehensive theory of self might look like.
THE NATURAL SELF
The natural self includes the two inevitably or innately given as-
pects of selfbood. First, there is the physical body. Nowadays in psy-
chology we do not pay all that much attention to the body as an aspect
of self, except perhaps in developmental psychology. Developmentalists
recognize the importance of the bodily self because their attention is
repeatedly forced back to the ineluctable fact that self-knowledge be-
gins with awareness of one's own body. It is also important to keep in
mind that outside the modern, industrialized world, self has always
been equated very much with body. The psychological self, as a net-
work of inner experience and identity definition, is far from universal.
But bodies exist all over the world. Bodies are culturally universal. Self
starts with body.
The other aspect of the natural self is that little window of con-
sciousness that is sometimes called the "knower." Like the body, the
knower is cross-culturally universal. Every conscious individual has one.
It is the owner of sensations and experiences, the subjective aspect of
consciousness. Everyone (in the world) is tied to his or her perspective

on the world, the little point from which everything else is seen. This
aspect, too, has received less emphasis in modern psychology than in
the past.
The knower and the body are the basic aspects of self in that they
are universal. They comprise the natural starting point of the self,
given by our very nature, so it seems fair to call them together the
natural self. Psychology has shown little interest in these aspects of
selfbood in recent years. The natural self exists, is real, and is undeni-
ably a fundamental part of self, but for some reason it does not attract
a great deal of attention from researchers these days.
THE CONCEPTUAL SELF

Let us turn now to the conceptual self. This is the self as a con-
struct, as something made out of meaning rather than given directly by
biology and phenomenology (in contrast to the natural self). Modern
psychology has shown considerable interest in the conceptual self. Again,
one must distinguish two (overlapping) subcategories: identity and self-
concept.
Self-concept refers to the person's mental representations of the
self as well as self's attributes. It includes one's beliefs about one's
body, one's personality, and perhaps one's material possessions as well.
The evaluative dimension of self-concept is quite important: self-esteem.
Most self-concept research ends up being self-esteem research. People
have global conceptions of themselves as having a certain degree of
value and they also have evaluations of their various specific attributes.
Now, psychology has seen self-esteem research go back and forth in its
preference for focusing on global self-esteem or on attribute-specific
self-esteem. Probably both are real and important. But there is no
denying that the good-bad dimension is a core aspect of self-concept
(Greenwald et al., 1988).
Identity refers to a composite definition of self, constructed out of
several partial definitions that include social roles, personal values, iden-
tifications, and so forth. In my 1986 book on identity (Baumeister, 1986), I
suggested that identity has the following functional aspects. There is
an interpersonal aspect of identity, which includes social roles as well
as reputation, as is evident in the ''public self'' of self-presentation re-
search (e.g., Baumeister, 1982; Schlenker, 1980). Identity, after all, en-
ables you to live in the world of people, so it has a strong interpersonal
aspect.
Identity also includes a potentiality aspect. That is, identity is not
a purely static conception or definition of self in the immediate present,

but rather is known as going somewhere. Identity is goal directed.
Having an identity includes having some idea of what you are here for,
of what you wish to achieve or become.
And lastly, identity includes a structure of values and priorities.
Historically this may be somewhat new as an aspect of individual iden-
tity. In the past, values were not personal values; rather, values were
part of the common culture, held by consensus and regarded by every-
one as objective facts. Now, however, people are confronted with a
welter of possible, conflicting values, and they construct their own per-
sonal set out of them. Identity crisis begins with a reevaluation of one's
basic values. The reasonably finished set of values is fundamental for
the self's further work, including forming a concept of one's potentiality
and, more generally, making decisions throughout life. It would be wrong
to say that identity makes the choices-because making choices is a
function of our third category, namely the action self-but identity fur-
nishes a basis for making stable, coherent, consistent choices.
The study of identity crises is probably one of the biggest develop-
ments in the area of self that has occurred since Allport wrote his early
works. First there were Erik Erikson's (1950, 1968) ideas based on
clinical impressions. These immediately captured the collective imagi-
nation, and the term "identity crisis" was heard throughout our culture.
Then in the 1960s, James Marcia (e.g., 1966, 1967) pioneered a way of
studying identity crises empirically, and since then a small flood of
articles has examined the correlates of identity crisis. When I set out
to work on the book on identity, I wanted to give an account of the
process of identity crisis, but I could not find anything in the literature
that spelled it out. Research had focused on correlates of identity crisis,
but not on processes. So, with Dianne Tice and Jeremy Shapiro, I un-
dertook to construct a process model. We tried to read everything we
could get our hands on and integrate these into one overarching ac-
count of the process of identity crisis (Baumeister et al., 1985).
But we failed. We concluded that the task was inherently impossi-
ble, because there is not one basic process but rather two. These are
worth considering, because they indicate how identity can fail and how
people seek to repair it. A person can have an identity crisis when his
or her identity is not working right; that is, when the identity is inade-
quate for dealing with the behavioral issues and decisions that confront
the person. Although we like to imagine identity crises as somehow
bubbling up from inside, they seem instead to occur in response to
various life and environmental demands that reveal the inadequacy of
one's identity.
We labeled the two types of identity crisis identity deficit and identity
conflict. Identity deficit crises mean having not enough identity: There
is not enough in one's identity to make the choices that need to be
made. The identity deficit has been shown to be much more common
with males than with females, and it seems to be most common in
adolescence and then again at mid-life. It is a matter of not knowing
what one wants to be, not having personal commitments and beliefs,
lacking a long-term life plan, and so forth. The person shows mood
swings, experiments with lots of new possible identifications and ideas,
ruminates over grand issues, reflects a great deal on his or her own
actions and their possible meanings, and so forth. Erik Erikson's work
helped us to see this period of vacillation and personal turmoil as a
form of personal experimentation. The individual tries on new ideas,
traits, loyalties, and futures just as a shopper might try on new clothes
to see what fits and what elicits a favorable reaction from others.
The other is identity conflict, which means too much identity. That
is, faced with some choices or demands, the different parts of identity
make conflicting demands on the person or they prescribe contradic-
tory, incompatible courses of action. For example, the opportunities and
demands of one's chosen career may conflict with obligations to family
and children. Likewise, the immigrant wants to retain allegiance to the
old culture but also embrace the new one. Identity conflicts can occur
at any point in life; unlike identity deficits, they seem to have no partic-
ular developmental link. There is no experimentation or search for new
identifications, no trying out of new patterns. Rather, there is only the
anguished struggle for a path of compromise and often the guilt-filled
repudiation of one of the conflicting elements.
THE ACTION SELF
Lastly, let us turn to the action self. The two subheads here refer
to the agent, who makes the choices and decisions, and the motivational
aspect of self. The self is both an actor and a wanter.
The agent is the active element of self. It is what decides, initiates,
chooses. This is a familiar and vitally important aspect of self. This was
perhaps the core of Freud's original concept of the ego, and it is no less
valid and important today.
The motivational aspect includes what the self desires. Actually,
the term "motivation" tends to connote something impersonal, perhaps
unconscious, perhaps situationally induced, and so when referring to
the self it may be more appropriate to use the term "desire." The self's
main desires are the desire to maximize self-esteem and the desire to
maximize control. Deprived of esteem, people are unhappy. They are

depressed, humiliated, sad, angry, even mentally ill, and they often find
themselves unable to perform effectively at many tasks or unable to
cope with setbacks. Likewise, deprived of control, people are unhappy.
They experience stress, they get ulcers and other illnesses, they may
suffer from learned helplessness, they show reactance, and again they
may find themselves unable to perform effectively or to cope. Research
shows that people who have both high self-esteem and high feelings of
control and efficacy are an unusually happy group, far out of propor-
tion to the objective circumstances in their lives (Argyle, 1987; Camp-
bell, 1981).
Esteem and control are not the only motivations associated with
self. Swann (1987) has shown that people often seek confirmation of
their views about themselves. The self desires stability and perhaps
some form of immortality, whether through lasting fame or through the
living legacy of one's offspring.
CONCLUSION
I think that these three main parts of the self are sufficient to
cover what is currently known and thought about the self. Some of
these notions were already old hat in Allport's day, whereas others had
scarcely been studied then but have received all of their attention in
the recent decades. Still, that is where we are now in the psychology
of self.
I wish to conclude with one further impression. When I set out to
write my first book on identity, my starting point was the modern
dilemma of identity. Finding oneself, self-actualization, knowing oneself,
and all the other cliches that suggest that problems of identity are
rampant in modern life. The attempt to furnish an account of identity
today in Western culture led me into a project that kept me fascinated
for several years and taught me a great deal.
Yet, when I finished that book, I felt that I had not really solved
my original problem, which was understanding this core dilemma of
modern life. I had indeed given an account of the dilemma of identity,
but I also concluded that much of what we call identity problems is not
really a matter of identity. Rather, it is a matter of the meaning of life.
Finding oneself, and all the rest, is often not a quest for identity per se,
but rather a search for a meaningful life. The underlying malaise is not
a lack of self-knowledge but a lack of purpose and meaning in how one
interprets one's own existence. The natural sequel to my study of iden-
tity is therefore a study of life's meaning, and I have spent several

years on that project (Baumeister, 1991).
This confusion of identity issues and meaning-of-life issues is ap-
parent in psychology as well as in the popular culture, and it is not easy
to untangle them. The cause of this confusion is not mere conceptual
sloppiness. Rather, in my research on the meaning of life, I have come
to see that the confusion is almost deliberate. Modern Western culture
has left us with some severe, distinctive shortcomings and problems in
creating a meaningful life. We have come to hope and think that indi-
vidual identity contains the solutions to these problems. We want the
self to be an answer to our problems of life's meaning. As a result, we
are coming to place more and more emphasis on the self as a fundamen-
tal source of meaning, purpose, and value. The self is used to fill the
value gap, or the existential vacuum, or whatever you want to call it.
As the culture has turned away from the religious and moral certain-
ties of the past, we have tried to get the individual self to take over
many of their guiding functions. Today, the highest goals in life for
many people involve knowing oneself, cultivating one's potential, fulfill-
ing oneself, and so forth; these replace the older ideas of serving God
and country and fulfilling socially based moral obligations. The self now
has to serve as a basic value, and that is something it was not originally
cut out for.
The fascination with self that pervades our culture-and our aca-
demic psychology-is more than an idle curiosity about what is inside
the individual. The quest for self is so urgent, so important, because the
self is taking over many of the basic functions of providing us with a
meaning for life.
As I argued in my book on identity, the major trend in our history
has been to place more and more demands on the individual self. Ask-
ing the self to provide us with enough purpose and value to make life
meaningful is only our latest, and perhaps our most unfair demand. But
it is one we make in deadly earnest.
REFERENCES
Alloy, L. B., & Abramson, L. Y. (1979). Judgment of contingency in depressed and non-
depressed students: Sadder but wiser? Journal of Experimental Psychology: Gen-
eral, 108, 441-485.
Allport, G. W. (1943). The ego in contemporary psychology. Psychological Review, 50,
451-478.
Argyle, M. (1987). The psychology of happiness. London: Methuen.
Baer, R., Hinkle, S., Smith, K., & Fenton, M. (1980). Reactance as a function of actual
versus projected autonomy. JCYUrnal of Personality and Social Psychology, 38, 416-
422.
Baumeister, R. F. (1982). A self-presentational view of social phenomena. Psychological
Bulletin, 91, 3-26.
Baumeister, R. F. (1986). Identity: Cultural change and the stT'Uf}gle for self New York:
Oxford University Press.
Baumeister, R. F. (1991). Meanings of life. New York: Guilford Press.
Baumeister, R. F., & Tice, D. M. (1984). Role of self-presentation and choice in cognitive
dissonance under forced compliance: Necessary or sufficient causes? JCYUrnal of Per-
sonality and Social Psychology, 46, 5--13.
Baumeister, R. F., Shapiro, J. J., & Tice, D. M. (1985). Two kinds of identity crisis.
JCYUrnal of Personality, 53, 407-424.
Brehm, J. W. (1966). A theory of psychological reactance. New York: Academic Press.
Campbell, A. (1981). The sense of well-being in America. New York: McGraw-Hill.
Erikson, E. H. (1950). Childhood and society. New York: Norton.
Erikson, E. H. (1968). Identity: YCYUth and crisis. New York: Norton.
tory. American Psychologist, 35, 603-613.
Greenwald, A. G., Bellezza, F. S., & Banaji, M. R. (1988). Is self-esteem a central ingredi-
ent of the self-concept? Personality and Social and Psychology Bulletin, 14, 34-45.
Kihlstrom, J. F., Cantor, N., Albright, J. S., Chew, B. R., Klein, S. B., & Niedenthal, P.M.
(1988). Information processing and the study of the self. In L. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 21, pp. 145--178). San Diego, CA:
Academic Press.
Langer, E. (1975). The illusion of control. JCYUrnal of Personality and Social Psychology,
29, 253-264.
Linder, D. E., Cooper, J., & Jones, E. E. (1967). Decision freedom as a determinant of the
role of incentive magnitude in attitude change. JCYUrnal of Personality and Social
Psychology, 6, 245--254.
Marcia, J. E. (1966). Development and validation of ego-identity status. JCYUrnal of Per-
sonality and Social Psychology, 3, 551-558.
Marcia, J. E. (1967). Ego identity status: Relationship to change in self-esteem, "general
maladjustment," and authoritarianism. Journal of Personality, 35, 118-133.
Markus, H., & Nurius, P. S. (1986). Possible selves. American Psychologist, 41, 954-969.
Schlenker, B. R. (1980). Impression management. Monterey, CA: Brooks/Cole.
Swann, W. B. (1987). Identity negotiation: Where two roads meet. JCYUrnal of Personal-
Taylor, S. E., & Brown, J. (1988). Illusion and well-being: Some social psychological con-
tributions to a theory of mental health. Psychological Bulletin, 103, 193-210.
CHAPTER THIRTEEN
Current Status of the

Motive Concept
ROBERT A. EMMONS
INTRODUCTION
The period from 1968 until the early 1980s was not a prosperous one
for personality psychology. There was growing disenchantment with
the traditional individual differences paradigm and with the apparent
lack of theoretical progress being made in the field. Authors of the
Annual Review chapters on personality regularly bemoaned the state
of the field. Apparently feeling a seven-year itch, the most trenchant
critiques appeared at septimal intervals (Adelson, 1969; Rorer & Widi-
ger, 1983; Sechrest, 1976). The major reason for the dissatisfaction was
the failure of trait measures to predict specific behaviors and the fail-
ure of trait indicators to correlate appreciably with each other. The
most popular solution to the behavioral consistency controversy, the
person X -situation interaction approach, was misguided from the start.
Personologists since the time of Allport argued that situations and per-
sons are not independent of each other, that people seek out and avoid
situations on the basis of their psychological propensities. This has now
been demonstrated in several empirical studies (Diener et al., 1984;
Emmons & Diener, 1986; Emmons et al.,1986; Snyder & Ickes, 1985).
Clearly it makes no sense to partition variance to persons and situa-
ROBERT A. EMMONS • Department of Psychology, University of California, Davis, Cali-

fornia 95616.
Fifty Years of Persorwlity Psycholo[J!J, edited by Kenneth H. Craik et al. Plenum Press,
New York, 1993.
187
188 ROBERT A. EMMONS
tions when those situations are to a great degree a function of the

individuals who inhabit them. Furthermore, it has been proposed that
stability and consistency in personality may be due in large part to the
selection, evocation, and manipulation of environments congruent with
the self (Buss, 1987; Emmons & Diener, 1986).
It is my belief that the behavioral consistency controversy was
prolonged because of a failure to sufficiently take into account the role
of motivation within personological functioning. Emphasis on traits and
the goal of predictability led to an absence of interest in motivation and
the equally worthy yet perhaps more complex goals of explanation and
understanding. This was particularly disturbing as the pioneers in the
field (Allport, Murray, Stagner, Lewin, McClelland) all stressed the dy-
namic striving character of behavior-its movement toward goals,
goals that are largely idiosyncratic. Of central interest to both Allport
and Stagner was motivation. For Allport, "motivation is the go of per-
sonality, and is, therefore, our most central problem" (1937, p. 218).
According to Allport, the intentions and motivational dispositions "tell
us what sort of future a person is trying to bring about, and this is the
most important question we can ask of any mortal" (1937, p. 223). In
addition, the unity of the self, according to Allport, is reflected in goal-
directed striving. Similarly, Stagner noted that "discussing personality
without regard to dynamics (i.e., motivation) is like describing the exte-
rior of an automobile ... ignoring the characteristics of the engine"
(1937, p. 257). Both Allport and Stagner firmly believed in the idio-
graphic basis of motivation. Allport championed the notion of personal
dispositions, of which there were two types: stylistic and motivational.
Stagner offered a cultural interpretation of motivation in which he ar-
gued that the motives toward which people strive are culturally deter-
mined, but also agreed that ''to know completely the motivation of any
personality, we must study that person" (1937, p. 306).
Gordon Allport cast a very large shadow. He not only overshad-
owed Ross Stagner, but also his older brother, Floyd Allport, well-
known to social psychologists but less so in personologist circles. The
same year that Allport and Stagner were publishing their seminal texts,
1937, Floyd Allport published an article in Character and Personality
(which of course became the Journal of Personality in 1945) entitled
"Teleonomic Description in the Study of Personality." In this article,
Floyd Allport proposed that personality traits were of limited utility
for describing the personality of an individual. While some might dis-
miss this as a sibling rivalry, Allport was making an important point.
An individual's personality might be better described, according to Floyd,
in terms of what the person seems to be "trying to do" or the purpose
THE MOTIVE CONCEPT 189
or purposes that a person seems to be trying to carry out. Allport

coined the term "teleonomic trend" to describe these behavioral tend-
encies, which he claimed were more dynamic and discriminating than
trait terms. Allport also suggested that these teleonomic trends could
be used to understand apparently inconsistent behavior (Pervin, 1983).
It appears that our founding fathers were also involved in the consis-
tency controversy, an issue that continued to occupy the time of a good
many of us some 50 years later. Allport's concept of teleonomic trend
became the topic of many a doctoral dissertation for his students, some
of whom went on to achieve considerable prominence. These included
Norman Frederiksen, John Valentine, Theodore Vallance, Arnold Tan-
nenbaum, Richard Solomon, Wilbur Gregory, and Charlotte Simon.
Unfortunately, the influence of the concept did not spread far from
Syracuse University and never really caught on. A possible reason for
this may have been the cumbersome method of assessing these trends,
which required observers ratings from a large number of peers. Floyd
Allport did not believe that what an individual said about his or her
motives should be taken at face value. There is reason, however, to
believe that the concept was abandoned prematurely. With the current
emphasis on goal-directed behavior and idiographic approaches to moti-
vation (Frese & Sabini, 1985; Pervin, 1989), the time seems ripe for its
renewal. To this end, I have been developing the concept of a "personal
striving" (Emmons, 1986), a modern-day descendant of the teleonomic
trend. Personal strivings are idiographically coherent patterns of goal
strivings and represent what an individual is typically trying to do. In
other words, personal strivings refer to the typical types of goals that
a person hopes to accomplish. Each individual can be characterized by
a unique set of personal strivings. For example, a person may be "try-
ing to appear attractive to the opposite sex," "trying to be a good
listener to his or her friends," and "trying to be better than others."
Personal strivings can be thought of as superordinate abstracting qual-
ities that render a cluster of goals functionally equivalent for an indi-
vidual. A personal striving is a unifying construct; it unites what may
be phenotypically different goals or actions around a common quality or
theme. Thus, a striving can be achieved in a variety of ways and satis-
fied through any one of a number of concrete goals.
It has become fashionable to posit hierarchical action control sys-
tems (Carver & Scheier, 1981; Hyland, 1988; Vallacher & Wegner, 1987).
In such a hierarchy, personal strivings are situated between global,
diffuse motives and concrete, specific actions, and, as such, represent a
desirable yet unexplored middle ground in the hierarchy of personality
functioning. Unlike teleonomic trends, personal strivings are not re-
stricted to the behavioral domain. They may be cognitive, affective, or

behavioral in nature. Also, our interest has been in what the person is
consciously trying to do. Gordon Allport is known for suggesting that if
you want to find out something about a person, ask that person di-
rectly. So we did. We also agree with Bernie Weiner's (1986) dictum
that the royal road to the unconscious is less valuable than the dirt
road to consciousness.
The heuristic value of the personal striving concept has now been
demonstrated in a number of studies, including predicting levels of sub-
jective well-being (Emmons, 1986); the influence of conflict and ambiv-
alence on psychological and physical well-being (Emmons & King, 1988);
the relationship between self-complexity and affective reactivity
(Emmons & King, 1989); and daily life events and well-being (Emmons,
1991). A review of the personal striving literature can be found in
Emmons (1989).
OTHER CURRENT APPROACHES
Stated simply, current approaches to motivation fall into one of

two types. First is the motive concept as defined by McClelland (1951,
1985) and his students (Atkinson & Birch, 1970; McAdams, 1985, 1988a;
Winter & Stewart, 1978). These authors define a motive disposition as
a class or cluster of affectively tinged goals, or a recurrent preference
for certain experiences. A small number of social motives, namely the
"big three" of achievement, affiliation-intimacy, and power, are suffi-
cient to describe and explain behavior and experience. A more recent
exemplar of this approach is McAdams's (1985) concept of "imago," de-
fined as "an idealized and personified image of the self'' (p. 178). Im-
agoes are the central elements of a person's identity and are centered
around the themes of intimacy and power. Imagoes are broad, superor-
dinate constructs that encompass interpersonal styles, values and be-
liefs, and personal needs and motives (McAdams, 1988a).
Scoring systems for the major motives based on analysis of stories
told in response to pictures similar to those in the Thematic Appercep-
tion Test (TAT) are well-established and well-validated. It would
appear that psychometric concerns over the use of the TAT have sub-
sided. The literature on social motives continues to grow. Some of the
more impressive uses have been in the application of social motives to
predicting important life outcomes such as physical health (Jemmott,
1987; McClelland, 1989) and presidential performance (Winter, 1987).
The other major type of motivational unit being adopted by inves-
tigators is idiographic in nature. In addition to the personal striving

concept discussed earlier, several other goal units have been proposed.
Klinger (1977) developed the notion of a "current concern" out of dis-
satisfaction with the failure of the motive dispositions to predict
spontaneous thought content. A current concern is a hypothetical moti-
vational state in between the identification of a goal and either the
attainment of the goal or disengagement from it. This hypothetical state
guides a person's ongoing thoughts, emotional reactions, and behavior
during the time it is active. A similar though independently developed
concept is the personal project (Little, 1983, 1989; Palys & Little, 1983).
Rooted in Murray's concept of a serial program, personal projects are
extended sets of actions intended to achieve a personal goal. Lastly,
Cantor (Cantor & Kihlstrom, 1987; Cantor & Langston, 1989) recently
developed the concept of life task, defined as "problems that people
are currently working on" (Cantor & Kihlstrom, 1987, p. 4). These life
tasks, consensual in nature but idiographically defined, organize and
give meaning to a person's everyday activities and are especially sa-
lient during life transitions, such as marriage or graduation from col-
lege. Klinger (1989), Cantor and Zirkel (1990) and Emmons (in press)
provide an elaboration of and a critical review of the similarities and
differences among these goal units.
Recently, McClelland and his associates (McClelland et al., 1989;
Weinberger & McClelland, 1990) have distinguished two forms of moti-
vation. One is an affectively-biologically based system, termed "implicit
motives," and the other a cognitively-experientially based system, termed
"self-attributed motives." These correspond to the social motive and
idiographic goal approaches described earlier. These two systems are
believed to develop independently, to operate independently of each
other, and to predict different classes of behavior. The degree to which
they are independent, however, is disputable. Emmons and McAdams
(1991) found significant relations between personal strivings (a form of
self-attributed motives) and motive dispositions as assessed by a pic-
ture-story exercise (implicit motives). Future research should identify
conditions under which implicit and explicit motives converge or fail to.
While the late 1960s and 1970s represented a stagnant period in
personality psychology's growth, recently there has been a much needed
injection of new directions and fresh perspectives into the field. These
range from interest in biological influences on personality (Buss, 1984;
Kenrick et al., 1985) to studying the naturally occurring stream of be-
havior and private experience using innovative methods (Craik, 1986;
Singer & Kolligian, 1987) to a resurgence of interest in focusing on
persons and lives (McAdams, 1988b). No longer do we feel obligated to
cite Walter Mischel in the opening paragraph of our articles nor apolo-
gize for the sad state of affairs of the field. A renewal of interest in
motivational concepts has played a significant role in this revival. It is
appropriate that on this 50th anniversary we are returning to an idio-
graphic analysis of motivation. There has been a clear shift from the
nomothetic social motives, which dominated the motivational literature
over the past 30 years, toward more molecular, idiographic goal units,
which are tied closer to everyday naturally occurring experience. Goals
possess many desirable properties. Their hierarchical structure with
links to both higher and lower levels, the flexibility and discrimina-
tiveness yet coherence that the concept implies and its amenability to
individual differences measurement make the goal concept a highly de-
sirable unit of analysis for personality psychology.
RECOMMENDATIONS
We have seen a proliferation of concepts in recent years, and more

can be expected. This is a healthy sign, but somewhere down the road
it needs to be tempered by attempts to demarcate the boundaries be-
tween these concepts. Surely if we asked a subject to provide a sepa-
rate list of his or her personal projects, current concerns, life tasks, and
personal strivings, there will be some overlap. The degree of overlap
will partially depend on the instructional set used. Attention needs to
be directed toward clarifying three types of relationships both within
and between the hierarchical control of action:
1. The horizontal structure. What is the relationship between these
motivational units of analysis? To what degree would one's current con-
cerns, say, overlap with one's life tasks? Could the current concerns be
derived from the life tasks? Would the life tasks emerge from a cluster-
ing analysis of the current concerns? Some, perhaps most, current con-
cerns and personal projects would not achieve status as life tasks or
personal strivings, such as the examples mentioned earlier, even
though they may serve as compelling temporary guides for thought and
action. Attention needs to be directed toward specifying the time frame
and category width of these respective concepts.
2. The vertical structure. Can these idiographic units of analysis
be related to units at a more abstract level of analysis, such as the
nomothetic motives? As an example, we have developed a system for
coding personal strivings into the major motive systems, such as
achievement, affiliation-intimacy, and power. Similarly, Cantor and
Kihlstrom (1987) note that at a general level of description, life task
themes for college students center on issues of intimacy and achieve-

ment. It may very well be that at an abstract level of analysis, most
personalized goal concerns tap either agency or communion (McAdams,
1985). Although movement to a broader level of analysis violates the
idiographic nature of these constructs, it is recognized that for certain
purposes a more superordinate level of description is desirable. Sub-
stantial literatures have developed around each of the major motive
systems (Aronoff & Wilson, 1985; McClelland, 1985), so it is essential to
integrate this level of analysis with the more circumscribed idiographic
units. The Emmons and McAdams (1991) study mentioned earlier sug-
gests that this is indeed feasible.
3. Interhierarchical relationships. What is the relation of these mo-
tivational units to nonmotivational units, such as to personality traits?
One can classify acts into traits categories, as in the act-frequency ap-
proach (Buss & Craik, 1983); however, acts can also be classified in
terms of what the person is trying to do, as in the personal striving
approach (Emmons, 1989a,b). These two theoretical systems offer
different ways for classifying everyday experience and will add to our
knowledge of the relations between trait-based and motive-based
accounts of action (Alston, 1970, 1975; Buss & Craik, 1983; Emmons,
1989b).
The complexity of relationships within, between, and across per-
sonality and motivational hierarchies is likely to be considerable. But it
is a complexity that should be embraced rather than avoided. As we
celebrate the 50th anniversary of Allport's, Murray's, and Stagner's
pioneering works, let us not be discouraged by the complexity of per-
sonality. After all, there is nothing in Allport, Stagner, or Murray that
would have led us to believe that the study of personality was going to
be easy. And surely there is enough work remaining to be done to keep
us busy for at least another 50 years.
REFERENCES
Adelson, J. (1969). Personality. Annual Review of Psychology, 19, 217--252.
Allport, F. (1937). Teleonomic description in the study of personality. Character and
hart & Winston.
Alston, W. P. (1970). Toward a logical geography of personality: Traits and deeper lying
personality characteristics. In H. D. Kiefer & M. K. Munitz (Eds.), Mind, science,
and history (pp. 70--105). Albany, NY: SUNY Press.
Alston, W. P. (1975). Traits, consistency, and conceptual alternatives for personality the-
ory. Journal for the Theory of Social Behavior, 5, 17-48.
Aronoff, J., & Wilson, J. P. (1985). Personality in the social process. Hillsdale, NJ:
Erlbaum.
Atkinson, J. W., & Birch, D. (1970). The dynamics of action. New York Wiley.
Buss, D. M. (1984). Evolutionary biology and personality psychology: Toward a concep-
tion of human nature and individual differences. American Psychologist, 39, 1135--
1147.
Buss, D. M. (1987). Selection, evocation, and manipulation. Journal of Personality and
Buss, D. M., & Craik, K. H. (1983). The act-frequency approach to personality. Psycho-
logical Review, 90, 105--126.
Cantor, N., & Langston, C. (1989). Ups and downs of life tasks in a life transition. In L.
A. Pervin (Ed.), Goal concepts in personality and social psychology (pp. 127-167).
Cantor, N ., & Zirkel, S. (1990). Personality, cognition, and purposive behavior. In L. A.
Pervin (Ed.), Handbook of personality theory and research (pp. 135--164). New York:
Guilford Press.
Carver, C. S., & Scheier, M. F., (1981). Attention and self-regulation: A Control theory
approach to human behavior. New York: Springer.
Craik, K. H. (1986). Personality research methods: A historical overview. Journal of
Diener, E., Larsen, R. J., & Emmons, R. A. (1984). Person X situation interactions:
Choice of situations and congruence models of interactionism. Journal of Personal-
Emmons, R. A. (1986). Personal strivings: An approach to personality and subjective
well-being. Journal of Personality and Social Psychology, 51, 1058--1068.
Emmons, R. A. (1989a). The personal striving approach to personality. In L. A. Pervin
NJ: Erlbaum.
Emmons, R. A. (1989b). Exploring the relations between motives and traits: The case of
narcissism. In D. M. Buss & N. Cantor (Eds.), Personality psychology: Recent
trends and emerging directions (pp. 32-44). New York: Springer-Verlag.
Emmons, R. A. (1991). Personal strivings, daily life events, and physical and psychologi-
cal well-being. Journal of Personality, 59, 453-472.
Emmons, R. A. (in press). Motives and life goals. In S. Briggs, R. Hogan, & W. Jones
(Eds.), Handbook of personality psychology. Orlando, FL: Academic Press.
Emmons, R. A., & Diener, E. (1986). Situation selection as a moderator of response
consistency and stability. Journal of Personality and Social Psychology, 51, 1013-
1019.
Emmons, R. A., & King, L. A. (1988). Conflict among personal strivings: Immediate and
long-term implications for psychological and physical well-being. Journal of Person-
Emmons, R. A., & King, L. A. (1989). Personal striving differentiation and affective
reactivity. Journal of Personality and Social Psychology, 56, 478-484.
Emmons, R. A., & McAdams, D. P. (1991). Personal strivings and motive dispositions:
Exploring the links. Personality and Social Psychology Bulletin, 17, 648-654.
Emmons, R. A., Diener, E., & Larsen, R. J. (1986). Choice and avoidance of everyday
situations and affect congruence: Two models of reciprocal interactionism. Journal of
Personality and Social Psychology, 51, 815--826.
Frese, M., & Sabini, J. (1985). Goal directed behaviar: The concept of action in psychol-
ogy. Hillsdale, NJ: Erlbaum.
Hyland, M. E. (1988). Motivational control theory: An integrative framework. Journal of
Personality and Social Psychology, 55, 642-651.
Jemmott, J. B. (1987). Social motives and susceptibility to disease: Stalking individual
differences in health risks. Journal of Personality, 55, 267-298.
Kenrick, D. T., Montello, D. R., & MacFarlane, S. (1985). Personality: Social learning,
social cognition, or sociobiology? In R. Hogan & W. H. Jones (Eds.), Perspectives in
personality (Vol. 1, pp. 201-234). Greenwich, CT: JAI Press.
Klinger, E. (1977). Meaning and void: Inner experience and the incentives in people's
lives. Minneapolis: University of Minnesota Press.
Little, B. R. (1983). Personal projects: A rationale and method for investigation. Envi-
ronment and Behaviar, 15, 273-309.
Little, B. R. (1989). Personal projects analysis: Trivial pursuits, magnificent obsessions,
and the search for coherence. In D. M. Buss & N. Cantor (Eds.), Personality psy-
chology: Recent trends and emerging directions (pp. 15-31). New York: Springer-
Verlag.
McAdams, D. P. (1985). Power, intimacy, and the life stary: Personological inquires into
identity. Homewood, IL: Dorsey Press.
McAdams, D. P. (1988a). Biography, narrative, and lives: An introduction. Journal of
McAdams, D. P. (1988b). Personal needs and personal relationships. In S. Duck (Ed.),
Handbook of research on personal relationships (pp. 7-22). New York: Wiley.
McClelland, D. C. (1951). Personality. New York: Dryden.
McClelland, D. C. (1985). Human motivation. Glenview, IL: Scott, Foresman.
McClelland, D. C. (1989). Motivational factors in health and disease. American Psycholo-
gist, 44, 675-683.
McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and
implicit motives differ? Psychological Review, 96, 690-702.
Palys, T. S., & Little, B. R. (1983). Perceived life satisfaction and the organization of
personal project systems. Journal of Personality and Social Psychology, 44, 1221-
1230.
Pervin, L. A. (1983). The stasis and flow of behavior: Toward a theory of goals. In M. M.
Page (Ed.), Nebraska Symposium on Motivation (pp. 1-53). Lincoln: University of
Nebraska Press.
Pervin, L. A. (1989). Goal concepts in personality and social psychology. Hillsdale, NJ:
Erlbaum.
Rorer, L. G., & Widiger, T. A. (1983). Personality structure and assessment. Annual
Sechrest, L. (1976). Personality. Annual Review of Psychology, 27, 1-27.
Singer, J. L., & Kolligan, J. (1987). Personality: Developments in the study of private
experience. Annual Review of Psychology, 38, 533-574.
Snyder, M., & Ickes, W. (1985). Personality and social behavior. In G. Lindzey & E.
Aronson (Eds.), Handbook of social psychology (Vol. 3, pp. 881-910). Hillsdale, NJ:
Erlbaum.
Vallacher, R. R., & Wegner, D. M. (1987). What do people think they're doing: Action
identification and human behavior. Psychological Review, 94, 3-15.
Weinberger, J., & McClelland, D. C. (1990). Cognitive versus traditional motivational
models. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and

cognition (Vol. 2, pp. 562r-597). New York: Guilford Press.
Weiner, B. (1986). Attribution, emotion, and action. In R. M. Sorrentino & E. T. Higgins
(Eds.), Handbook of motivation and cognition (pp. 281-312). New York: Guilford
Press.
Winter, D. G. (1987). Leader appeal, leader performance, and the motive profile of lead-
ers and followers: A study of American presidents and elections. Journal of Person-
Winter, D. G., & Stewart, A. J. (1978). The power motive. In H. London & J. E. Exner,
Jr. (Eds.), Dimensions of personality (pp. 391-448). New York: Wiley.
CHAPTER FOURTEEN
The Ability to Judge Others from

Their Expressive Behaviors
BELLA M. DEPAULO
Allport (1937) had a very strong opinion about where to look in order
to figure out the content and structure of people's personalities: Look
at their expressive movements. That is, look not only at what people
are doing, but how they are doing it; listen not only to what they are
communicating, but also the manner in which they are communicating
it. In telling us to take these expressive movements very seriously,
Allport was not telling us to disregard what people are doing or trying
to do. In fact, he maintains that what people are trying to do is most
fundamental in revealing the nature of their traits. But still, he cau-
tioned, we should not ignore the "hows" of behavior. Sometimes the
ways that people do things are redundant with the fact that they are
doing those things. To embellish Allport's own example a little (1937,
pp. 464-465), if a group of people were to walk to Yankee Stadium
every time the Yankees had a home game, that behavior would suggest
that they were very enthusiastic about Yankee baseball. If, in addition,
one were to observe that on the way to the Stadium, they all had
bubbly faces and sprightly gaits, and that their tee shirts, hats, watches,
and tote bags were all emblazoned with the Yankee insignia, that infor-
mation would only serve to underscore the information already avail-
able from the knowledge that they attend every game. But, Allport
BELLA M. DePAULO • Department of Psychology, University of Virginia, Charlottes-

ville, Virginia 22903.
New York, 1993.
197
198 BELLA M. DEPAULO
claimed, expressive movements can do more than simply tell us the

same information in a different way. Allport believed that expressive
behavior is unconsciously determined and therefore can provide a clue
to deep-seated aspects of personality that are not always evident in the
content of behavior.
THE CONSISTENCY OF EXPRESSIVE BEHAVIORS
Allport had a very broad view of the kinds of behaviors that might
be meaningfully expressive. Some of these are very familiar to contem-
porary students of nonverbal communication, including cues such as
facial expressions, body postures, speech fluency, and vocal intensity.
Other expressive cues that interested Allport are somewhat less famil-
iar to us now-cues such as the speed with which we draw things, the
size of the check marks that we make when filling out checklists, and
the degree to which we overestimate the size of angles. Still other cues
are ones that were once of some interest to psychologists, but, with a
few exceptions, are no longer taken very seriously. These include styles
of handwriting and the pressure applied to one's pen or pencil while
writing.
Allport thought that these very diverse expressive behaviors fell
into three basic clusters of expansiveness, emphasis, and outward ten-
dency or extraversion. But if he had modern data-analytic techniques
at his disposal, his heart would be not in the specific clusters but in the
unrotated first factor. He believed that one of the fundamental truths
about expressive behaviors was that they all simultaneously expressed
the same trait. To cite one of his favorite quotes, "One and the same
spirit is manifest in all."
Allport's assumption, then, is clearly one of consistency of expres-
sive movements. He believed that different expressive behaviors were
consistent with each other and that any given expressive behavior, for
a particular individual, would be consistent across time and across situ-
ations. His expectations were not absolute, however. Instead, he cau-
tioned that we should never expect consistency to be perfect. If it
were, then we could take any one expressive behavior and learn every-
thing about personality from it. In contrast, Allport's belief was that
we need to look at the entire patterning of expressive cues and that we
should expect to find as much consistency in expressive behavior as
there is in the personality itself-not more and not less. Expressive
behavior, like personality, includes much that is consistent, but it is also
marked by conflict and contradiction.
ABILITY TO JUDGE OTHERS 199
Despite the words of caution in Allport's conclusions, the thrust of

his perspective was very optimistic with regard to the questions of
whether we can expect to be able to understand accurately other peo-
ple's feelings and traits. The promise of this perspective was not lost on
subsequent researchers, and there have been attempts to search for
consistencies across different kinds of expressive behaviors and for links
between personality and expressive behavior.
Though no one is quite ready to tie a ribbon around the results of
this research and present it to the ghost of Gordon Allport, I think it is
fair to say that some progress has been made. You can see evidence of
that progress in the refinement in the kinds of questions that are posed.
We no longer ask simply whether there is consistency nor even just
how much consistency there might be; we now ask where we should
look for this consistency-in what kinds of people, what kinds of situa-
tions, and along what sorts of dimensions? And we are all very sensi-
tized to the importance of asking the complementary question of when
not to expect any consistency at all.
Allport, in his 1937 book, bemoaned the fact that there was very
little research on the relationship between particular traits and partic-
ular expressive behaviors, and noted that methodologically, researchers
were not at all sure how to go about producing the relevant data.
Representative of the immature state of the literature at that time was
Adler's suggestion, based on no data, that one way to distinguish opti-
mists from pessimists was to observe them when they are sleeping;
pessimists, he said, would "curl themselves into the smallest possible
space and ... draw the covers over their heads" (quoted in Allport, 1937,
p. 486). Contemporary researchers would know how to substantiate (or
insubstantiate) this particular claim empirically, but have not quite
brought themselves to do so. However, researchers have documented
other stable expressive differences; for example, we know about the
loud voices of extraverts (Scherer, 1979) and the fidgety and withdrawn
behavior of the socially anxious (Leary, 1983; Schlenker & Leary, 1982).
THE DELIBERATE REGULATION

OF EXPRESSIVE BEHAVIORS
I think one very important development in the study of expressive

behavior that cannot be fully credited to Allport is the literature on the
deliberate regulation and control of expressive behaviors (see DePaulo,
1991, 1992; DePaulo et al., 1992). Although Allport was willing to admit-
though somewhat grudgingly, I think-that people might try deliber-
ately to disguise their expressive behaviors, he was willing to concede

this only for specific behaviors or for short periods of time. When it
came to what he referred to as style, or the totality and complexity of
all expressive behaviors taken together, he did not think that deliber-
ate disguise was even a possibility. In his words, "Style ... develops
gradually from within; it cannot for long be simulated or feigned" (1937,
p. 493; see also Hogan et al., 1985).
Enter Erving Goffman (1959), and a growing list of followers and
semifollowers. From an impression management perspective, many be-
haviors do not emanate purely and spontaneously from the true person-
ality within; rather, they are the product of deliberate regulation and
control. Even behaviors that appear perfectly spontaneous and natural
are not necessarily so; instead, they might be the creations of especially
smooth and skilled self-presenters. Or, it may be the case that the
expressive behaviors in question are being emitted un-self-consciously
at the moment, but only because they were carefully constructed at one
time in the past, then practiced and practiced and practiced until they
became habitual and thus nearly indistinguishable from truly spontane-
ous expressions (cf. Schlenker, 1980). So the first point about the delib-
erate regulation of expressive behaviors is that it may well occur quite
frequently.
The second point is that deliberate attempts at controlling one's
expressive behaviors can ruin expressive consistency. If, for example,
upon demolishing a much-loathed opponent at tennis, you allow your-
self just a little tiny smile, woe to the researcher who tries to find
smug mirth in every other aspect of your expressive behavior, too.
Third, attempts at deliberate regulation can also enhance expres-
sive consistency (Lippa, 1983). If a person wanted to convince you that
she was an extravert, she might deliberately try to convey extraver-
sion in every way she could think of. She might try to don an extra-
verted posture and extraverted gestures, she might speak with an
extravert's voice, and put on an extraverted face. So, consistency across
different expressive behaviors might be accentuated. Deliberate con-
trol can also increase the consistency of the link between traits and
expressive behaviors. I think that most people believe that their true
personalities, as they construe them, are immediately apparent to oth-
ers (cf. DePaulo et al., 1987). They seem to feel that there is no need to
make an effort to appear to be the way they think they really are,
because they will appear that way even if they do not try. However,
when it is really important to them that another person should be
aware of their virtuous personality traits, they might not take any
chances. The person in this situation who believes she is an extravert
and really is an extravert will take great pains to make sure that her
extraversion is abundantly clear to her partner. This is deliberate reg-
ulation, but it is regulation that strengthens the link between personal-
ity and expressiveness rather than shattering it.
A further point about deliberate attempts at expressive control is
that they are not always successful. Freud (1959, p. 94), of course, de-
lighted in warning us that if we try to keep a secret, betrayal will ooze
out of us at every pore. When the data on this issue rolled in, they
suggested that some pores are much "oozier" than others. For example,
when people are not too aroused or emotional, they tend to be very
successful at regulating their facial expressions. Research on deception
provides some interesting demonstrations of this. When the stakes for
telling a successful lie are not too high, liars are very good at using
their faces to fool their targets. In fact, in those situations, their tar-
gets might actually have a somewhat better chance of detecting the lie
if they cannot see the liar's face at all. However, as the stakes go up,
and it becomes more important to the liar to get away with the lie,
facial expressions and sometimes other nonverbal cues, too, are likely
to "leak" the information that the liar is trying to hide (DePaulo &
Kirkendol, 1989; DePaulo et al., 1985).
Deliberate attempts at regulation can fail for other reasons, too.
For example, some people are unaware of their expressive behaviors or
insensitive to their impact on others; still others realize that such be-
haviors are impactful, but are inept at controlling them. Further, some-
times the very act of trying to control expressive behaviors backfires,
and the person's behavior appears to others to be awkward, unnatural,
or overly controlled (DePaulo, et al., 1983).
The final point I want to make about the impression management
perspective on expressive behaviors is that it suggests a different way
of conceptualizing such behaviors. From this perspective, expressive
behaviors are not always unbridled expressions of a true underlying
personality. Instead, they might sometimes be better regarded as mani-
festations of social skills-skills that can perhaps be practiced and trained
(Argyle & Kendon, 1967; Friedman, 1979). From this perspective,
socially anxious individuals are not necessarily stuck with their stam-
mering, gaze avoidance, and desperate nods and smiles, and even extra-
verts can learn to calm down and shut up.
I see the impression management perspective as complementary
to, rather than competitive with, Allport's ideas about un-self-conscious
expressiveness. Certainly there are times, such as when we are caught
up in the emotion of the moment, that we are spontaneously and un-
self-consciously expressive. Further, even when we do try purposefully
to regulate our expressive behaviors, even these attempts may be

stamped with the ink of our own personal styles.
ALLPORT'S SIX QUESTIONS ABOUT THE FACE
Allport realized that much remained to be learned about expres-

sive behaviors, especially facial behaviors, and he outlined six questions
that he thought could guide the scientific study of the face. The first of
these sounds much like a quaint hypothesis from 50 years ago. Allport
asked whether "native factors in personality, such as temperament and
intelligence, are reflected in the bodily form and structure [such as "the
bony configuration of the face"]; whereas acquired traits are repre-
sented in muscular sets and changes" (1937, p. 482). The other five
questions, though, could almost serve as chapter headings in a contem-
porary textbook on the face.
The second question Allport posed was about the eyes. Allport
wondered whether "the subtleties of glance ... are especially rich in
expressive significance?" (1937, p. 482). Fifty years later, we now know
much about the flavor of that richness. We know that gaze can express
affiliation and liking and dominance and status. We know that it can be
used to gain information, to avoid giving away information, and to reg-
ulate the flow of conversation. We know that it can grab a target's
attention and arouse that person, so that she is primed to figure out
why she is being observed. Finally, I think it would please Allport to
hear that patterns of gazing have been empirically linked to gender,
culture, psychiatric status, and, of course, personality (e.g., Argyle &
Cook, 1976; Ellsworth & Langer, 1976; Ellsworth & Ludwig, 1972; Ex-
line, 1972; Fehr & Exline, 1987).
Allport's third question was, "Can patterns of facial expression ...
be analyzed into the contraction of separate muscles?" (1937, p. 482). It
took researchers about 40 of the 50 years to get to this question in a
comprehensive way, but the end results are truly elegant. In Ekman's
Facial Action Coding System (Ekman & Friesen, 1976), for example,
any facial movement can be described in terms of the separate facial
muscles whose triggering produced that movement (see also Izard, 1983).
This, of course, is just what Dr. Allport ordered.
Allport's fourth question was about smiling. "Why," he asked, "is
the smile so disarming a pattern of expression?" Allport undoubtedly
had a charming smile in mind when he posed this question, and we still
do not have a complete answer to the question of why such smiles can
be so disarmingly charming. But we do know that they are not all so
scintillating; they can be perfidious as well as polite and sociable, artful

as well as ingenuous, and miserable as well as mirthful (e.g., Brunner,
1979; Bugental et al., 1971; Ekman & Friesen, 1982; Ekman, Friesen, &
O'Sullivan, 1988; Kraut & Johnston, 1979). And, true to tradition, re-
search has uncovered stable individual differences in at least some of
these uses of the smile (e.g., Hall, 1984).
Judge for yourself whether this fifth question has a contemporary
ring to it. Allport asked, ''Why so frequently does an affective reaction
to liking or disliking a stranger precede (and sometimes preclude) ob-
jective judgment?" He goes on to note that sometimes when we have a
strong affective reaction to someone we just met, it may be because
that person is similar to some other person about whom we feel strongly.
Allport goes on to ask, "[If this is so, then] why is the affective judg-
ment swifter than the conscious recognition of similarity?"
The sixth question Allport raised takes us back more directly to
the issue with which we began-that of the ability to judge others
accurately on the basis of their expressive behaviors. In formulating
this last question, Allport referred to an impactful study conducted by
Landis in 1924. In this study, Landis tried to elicit spontaneous emo-
tional reactions in very involving and realistic ways. For example, one
of the tasks that Landis asked his subjects to perform was to slice the
head off a live rat. What he found was that when his subjects were
chopping off the rat's head, some of them looked disgusted, but others
looked rather somber, and still others actually laughed. This study, and
others showing similar results, had a devastating impact on future re-
search on nonverbal expressiveness. For all the wrong reasons, theo-
rists jumped to the inappropriate conclusion that we simply could not
expect people to be able to make accurate judgments of others based
only on their facial expressions. Allport was particularly prescient on
this issue; he thought he smelled a rat. "If," he asked, " ... patterns of
expression differ markedly from individual to individual, how does it
happen that we are able to judge other people as well as we do?"
Allport was a great believer in intuition, and in this case his intuition
told him that people can indeed in many instances make accurate judg-
ments about others on the basis of their facial expressions. The missing
link in the Landis study was the one between the situation of chopping
off a rat's head and the particular emotion experienced by each individ-
ual subject. Different subjects presumably experienced different emo-
tions or sets of emotions and had different self-presentational goals.
Years later, researchers would try to manipulate the emotions experi-
enced by subjects, so that on any given trial, they were experiencing
one of the basic emotions in a relatively pure form. When researchers
such as Izard (1971) and Ekman (1972; Ekman & Friesen, 1986) elicited
emotions such as surprise, happiness, sadness, anger, fear, disgust, and
(most recently) contempt in this very careful way, they found that the
resulting facial expressions could be recognized by persons on every
continent and in every little village into which these researchers ven-
tured. This research, of course, was on accuracy of emotion perception,
and Allport was even more interested in accuracy of personality per-
ception. But that issue, too, has resurfaced, and with a theoretical and
methodological vengeance. At this very moment, three recast papers
on accuracy of person perception are already in print in Psychological
Bulletin: David Funder's (1987), Dave Kenny and Linda Albright's
(1987), and Arie Kruglanski's (1989). Accuracy is back-and just in time
for this fiftieth anniversary celebration.
REFERENCES
Allport, G. W. (1937). Personality. New York: Holt.

Argyle, M., & Cook, M. (1976). Gaze and mutual gaze. Cambridge, England: Cambridge
University Press.
Argyle, M., & Kendon, A. (1967). The experimental analysis of social performance. In L.
Berkowitz (Ed.), Advances in experimental social psychology (Vol. 3, pp. 55-98).
New York: Academic Press.
Brunner, L. J. (1979). Smiles can be back channels. Journal of Personality and Social
Bugental, D. E., Love, L. R., & Gianetto, R. M. (1971). Perfidious feminine faces. Journal
of Personality and Social Psychology, 17, 314--318.
DePaulo, B. M. (1992). Nonverbal behavior and self-presentation. Psychological Bulletin,
111, 203-243.
DePaulo, B. M. (1991). Nonverbal behavior and self-presentation: A developmental per-
spective. In R. S. Feldman & B. Rime (Eds.), Fundamentals of nonverbal behavior
(pp. 351-397). Cambridge: Cambridge University Press.
DePaulo, B. M., Blank, A. L., & Hairfield, J. G. (1992). Expressiveness and expressive
control. Personality and Social Psychology Bulletin, 18, 276-285.
DePaulo, B. M., Kenny, D. A., Hoover, C. W., Webb, W., & Oliver, P. V. (1987). Accuracy
of person perception: Do people know what kinds of impressions they convey? Jour-
nal of Personality and Social Psychology, 52, 303-315.
DePaulo, B. M., & Kirkendol, S. E. (1989). The motivational impairment effect in the
communication of deception. In J. Yuille (Ed.), Credibility assessment (pp. 51-70).
Belgium: Kluwer Academic Publishers.
DePaulo, B. M., Lanier, K., & Davis, T. (1983). Detecting the deceit of the motivated liar.
Journal of Personality and Social Psychology, 45, 1096-1103.
DePaulo, B. M., Stone, J. I., & Lassiter, G. D. (1985). Deceiving and detecting deceit. In
B. R. Schlenker (Ed.), The self and social life (pp. 323-370). New York: McGraw-
Hill.
Ekman, P. (1972). Universals and cultural differences in facial expressions of emotion. In
J. K. Cole (Ed.), Nebraska symposium on motivation, 1971 (Vol. 19, pp. 207-283).
Lincoln: University of Nebraska Press.
Ekman, P., & Friesen, W. V. (1976). Measuring facial movement. Environmental Psychol-
ogy and Nonverbal Behavior, 1, 5~75.
Ekman, P., & Friesen, W. V. (1982). Felt, false, and miserable smiles. Journal of Nonver-
bal Behavior, 6, 238--252.
Ekman, P., & Friesen, W. V. (1986). A new pan-cultural facial expression of emotion.
Motivation and Emotion, 10, 159-168.
Ekman, P., Friesen, W. V., & O'Sullivan, M. (1988). Smiles while lying. Journal of Per-
Ellsworth, P. C., & Langer, E. J. (1976). Staring and approach: An interpretation of the
stare as a non-specific activator. Journal of Personality and Social Psychology, 33,
117-122.
Ellsworth, P. C., & Ludwig, L. M. (1972). Visual behavior in social interaction. Journal of
Communication, 22, 375-403.
Exline, R. V. (1972). Visual interaction: The glances of power and preference. In J. K.
Cole (Ed.), Nebraska symposium on motivation, 1971 (Vol. 19, pp. 163-206). Lincoln:
University of Nebraska Press.
Fehr, B.J., & Exline, R. V. (1987). Social visual interaction. In A. W. Siegman & S.
Feldstein (Eds.), Nonverbal behavior and communication (2nd ed., pp. 225--348).
Freud, S. (1959). Collected papers. New York: Basic Books.
Friedman, H. (1979). The concept of skill in nonverbal communication: Implications for
understanding social interaction. In R. Rosenthal (Ed.), Skill in nonverbal commu-
nication. Cambridge, MA: Oelgeschlager, Gunn, & Hain.
Funder, D. C. (1987). Errors and mistakes: Evaluating the accuracy of social judgment.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY: Double-
day.
Hall, J. A. (1984). Nonverbal sex differences. Baltimore, MD: Johns Hopkins University
Press.
Hogan, R., Jones, W. H., & Cheek, J. M. (1985). Socioanalytic theory. In B. R. Schlenker
(Ed.), The self and social life (pp. 175-198). New York: McGraw-Hill.
Izard, C. E. (1971). The face of emotion. New York: Appleton-Century-Crofts.
Izard, C. E. (1983). The maximally discriminative facial movement scoring system (rev.
ed.). Unpublished manuscript, University of Delaware.
Kenny, D. A., & Albright, L. (1987). Accuracy in interpersonal perception: A social rela-
tions analysis. Psychological Bulletin, 102, 390--402.
Kraut, R. E., & Johnson, R. E. (1979). Social and emotional messages of smiling: An
ethological approach. Journal of Personality and Social Psychology, 37, 1539-
1553.
Kruglanski, A. W. (1989). The psychology of being "right": The problem of accuracy in
social perception and cognition. Psychological Bulletin, 106, 395-409.
Landis, C. (1924). Studies of emotional reactions: II. General behavior and facial expres-
sion. Journal of Comparative Psychology, 4, 447-509.
Leary, M. R. (1983). Understanding social anxiety. Beverly Hills, CA: Sage.
Lippa, R. (1983). Expressive behavior. In L. Wheeler & P. Shaver (Eds.), The review of
personality and social psychology (Vol. 4, pp. 181-205). Beverly Hills, CA: Sage.
Scherer, K. R. (1979). Personality markers in speech. In K. R. Scherer & H. Giles (Eds.),
Social markers in speech (pp. 147-209). Cambridge: Cambridge University

Press.
Schlenker, B. R. (1980). Impression management: The self concept, social identity, and
interpersonal relations. Monterey, CA: Brooks/Cole.
Schlenker, B. R., & Leary, M. R. (1982). Social anxiety and self-presentation: A conceptu-
alization and model. Psychological Bulletin, 92, 641--009.
CHAPTER FIFTEEN
Judgments of Personality and

Personality Itself
DAVID C. FUNDER
My topic is accuracy in personality judgment and the kind of research

we need to do if we are interested in it. I emphasize the "if' because,
perhaps surprisingly, not everybody is interested in accuracy, not even
every psychologist who studies judgments of personality-a point I
shall return to below. But one psychologists who surely was interested
in how accurate personality judgments might be was Gordon Allport.
ALLPORT ON PERSONALITY JUDGMENT
One of our co-symposiasts has been known to refer occasionally to

the good Doctor Allport as "Saint Gordon." It is true that in some
circles Allport's classic 1937 volume has attained nearly the status of a
sacred text. So let us go directly to this esteemed book, which includes
a short chapter on "The Ability to Judge People." Allport began with a
brief overview of the topic as he saw it, a half-century ago:
From the psychologist's point of view some of the most important
problems involved in judgments of personality are the following:
(1) the nature and reliability of first impressions, (2) the chief fac-
tors involved in judging, (3) the value of interviews, (4) the ques-
DAVID C. FUNDER • Department of Psychology, University of California, Riverside,

California 92521.
New York, 1993.
207
208 DAVID C. FUNDER
tion whether ability to judge people is general or specific, (5) the

qualifications of a good judge, (6) the relative excellence of men and
women as judges, (7) the types that are best known to us, and
(8) common sources of error in judgment.* (1937, p 499)
Today, with 50 years of hindsight upon the research that followed
this statement, one might want to add or subtract a bit from this list,
and reorder some of its priorities. At least, I would. But still, five
decades after it was set down, Allport's compendium of research issues
concerning the accuracy of personality judgment remains a fair list.
Indeed, the list inspires me to a couple of observations.
First, Allport thought that the topic of interpersonal judgment be-
longed to the field of personality psychology. This viewpoint contrasts
in an interesting way with the situation today, when the topic seems
more often regarded as lying within the bailiwick of social psychology
or even more recently cognitive-social psychology.
Second, Allport thought the topic of accuracy belonged to personal-
ity psychology for a particular reason, one he probably considered too
obvious to mention. Allport assumed that there was something "real"
for judgments of personality to describe. Therefore, the study of judg-
ments of personality could not avoid being inextricably entwined with
the study of personality itself.
This is emphatically not the working assumption of social or cognitive-
social psychology nowadays. Indeed, the very idea that there might be
something out there to judge is assiduously avoided, even sidestepped,
by the vast majority of modern research on personality judgment. Judg-
ments of personality are commonly regarded as little more than inter-
esting social phenomena that have little if anything to do with any
reality beyond the realm of social judgment itself. As a result, the
study of personality judgment has been all but evicted from what All-
port would consider its rightful place within personality psychology.t
RESEARCH ON THE PROCESS OF JUDGMENT
For the most part, modern research on interpersonal judgment

takes one of two forms. The first and more dominant form is concerned
with the process of judgment. The basic strategy is to propose a model
* Ken Craik brought this passage to my attention several years ago. Since then, I have
exploited it many times to establish my own bona fides as a personality psychologist.
t Indeed, things have gotten so had that even some personality psychologists seem to think
there is something wrong with including the study of interpersonal judgments within the
field of personality (e.g., Carlson, 1984).
JUDGMENTS AND PERSONALITY ITSELF 209
of the judgmental process and then present subjects with experimental

stimuli designed to allow the assessment of the degree to which their
judgments of these stimuli follow predictions derived from the model.
This strategy is fine so far as it goes and is an excellent way to illumi-
nate the process of judgment. However, in much of this research, the
process model is given a normative status, so that when subjects'
judgments fail to match the model's prescriptions, the researcher is
emboldened to conclude not only that the model poorly describes
human judgment, but that human judgment itself is flawed. For in-
stance, Bayesian inference provides a mathematical model of how infor-
mation concerning prior probabilities should be combined when making
predictions concerning future outcomes. Many experiments have asked
subjects to make probability judgments on the basis of numerical pri-
ors, and found that subjects' models consistently deviate from Bayes'
prescriptions. These results were typically taken to reveal "errors" or
"fallacies" of judgment (e.g., the ''base rate fallacy''; see Nisbett & Ross,
1980, for many more examples).
This is not the place to reprise my critique of the error literature
(but please see Funder, 1987). For present purposes, it is sufficient to
make the point that research on judgmental error, perhaps ironically,
says nothing about judgmental accuracy, for the same reason that re-
search on errors in visual judgment, or "optical illusions," should not
lead us to conclude that people cannot see. Illusions such as the Ponzo
or Muller-Lyer effects reveal mechanisms of visual judgment that are
usually adaptive; indeed, they reveal important aspects of how accurate
visual perceptions can arise (Gregory, 1971). Only in the artificial and
rather surreal environment of the laboratory do they lead to visual
mistakes. I believe that much the same is true about many if not all of
the errors that have been demonstrated in the domain of social judg-
ment. A parsimonious explanation of why so many exist might be that
they are the result of judgmental heuristics that ordinarily produce
correct judgments in real life, even though they can lead to incorrect
judgments within laboratory environments. I realize that this is a con-
troversial opinion, but what should not be controversial is that error
research is simply not informative about the accuracy of judgment, one
way or the other.* Even though research on judgmental error can be
* Indeed, research on the judgment process, such as error research, was originally and
explicitly designed to avoid accuracy issues. The brilliant insight of Asch (1946) was that
interesting and important infonnation about the process of judgment could be obtained
in experiments in which subjects judged stimuli that were wholly artificial. As Jones
(1985) pointed out, "Asch solved the accuracy problem by bypassing it." What Jones
210 DAVID C. FUNDER
importantly informative about the process of judgment, it does not ad-

dress accuracy as Allport thought of it.
RESEARCH ON INTERJUDGE AGREEMENT
A bit closer to the mark is a second kind of research on judgments

of personality that focuses on the phenomenon of interjudge agree-
ment. After an initial burst of enthusiasm for this topic died down some
years ago (Taft, 1955), the study of agreement has enjoyed something
of a renaissance more recently (Kenny & La Voie, 1984). Sometimes
different judges of the same individual's personality agree with each
other or with the individual him- or herself in their judgments, and
sometimes they do not (e.g., Funder & Colvin, 1988; Funder & Do-
broth, 1987). This fact would seem to have obvious implications for
judgmental accuracy.
However, like researchers on judgmental process, researchers on
interjudge agreement also often seem to go out of their way to avoid
becoming entangled in accuracy issues. They have tried to avoid deal-
ing with accuracy through the use of either of two related strategies.
The first was commonly used during the first wave of agreement re-
search, which occurred from about 1930 to about 1960: Many studies
were performed in which acquaintances of the judgmental target were
asked to describe, not the personality of the target, but how the target
would describe his or her own personality. Then the target filled out
forms to describe himself or herself. The degree of congruence between
these two evaluations could of course be measured directly and was
assumed to reflect the acquaintances' degree of social sensitivity or
empathy (e.g., Gage & Cronbach, 1955; Taft, 1966).
The more recent wave of agreement research, conducted during
the 1980s, has for some reason often utilized almost exactly the oppo-
site approach: Investigators ask the target of judgment to describe, not
his or her own personality, but how he or she believes he or she will be
described by others. Then acquaintances fill out forms to describe the
meant was that by inventing a research paradigm that allowed accuracy issues to be
finessed through the use of artificial stimuli, Asch had provided a way to address other
important issues concerning the process of judgment. It seems ironic as well as unfortun-
ate that, over the years, researchers seemed to forget all this and began to make the
fundamental mistake of interpreting research on process and error as implying that
human judgment is flawed. That is not what Asch had in mind. (Incidentally, human
judgment may well be flawed. But error research is not constituted in such a way as to
find out.)
target. The degree of congruence between these two evaluations can

again be measured directly and is assumed to reflect the target per-
sons' "ability to know what kinds of impressions they convey" (DePaulo
et al., 1987, p. 311).
Both kinds of agreement, as studied in the 1930s and in the 1980s,
are surely important in their own right. It is interesting to see how
well we can predict what somebody else will say about himself or her-
self, and it would be useful to know the degree to which our impres-
sions of ourselves are held mutually by the people with whom we deal
on a daily basis. However, the study of neither sort of agreement is
equivalent to the study of judgmental accuracy. We may call our intelli-
gent acquaintance "dumb" and he may be aware of our misperception.
We may be crooked folk who manage to convince everybody else that
we are honest and be well aware of our success. The result in both
instances would be high interjudge agreement of the two sorts just
mentioned, but low accuracy.
Moreover, an exclusive concern with judgments themselves and
how well they agree can lead researchers to forget about or even avoid
thinking about what might really be out there to be judged. And that
will lead them to fail to gather further data that might actually help
determine whether or not the judgments are accurate. For instance, if
a target and acquaintance disagree about the presence of some person-
ality trait, perhaps we could fmd a way to measure directly the rele-
vant behavior of the target and settle the issue.* But as long as the
researcher is content merely to assess congruence between judgments,
for their own sake, the question will not even arise. This kind of re-
search, therefore, also often sidesteps the issue of accuracy as Allport
thought of it.
CRITERIA FOR ACCURACY
Of course, there is a good reason why all these researchers work

so hard to avoid the accuracy issue; they are not just being perverse.
The reason is that in order to study the accuracy judgments, it would
seem, you need a criterion. And what criterion can one use to evaluate
the accuracy of a judgment of personality?
* I do not underestimate the problems in gathering the sort of behavioral data that would
allow the accuracy of judgments of personality to be assessed. My point is that unless one
takes the accuracy question seriously, by focusing on more than just the relationship
between different judgments, one will not even begin the formidable task of formulating
and gathering such data.
212 DAVID C. FUNDER
To many, this question has seemed to raise an insurmountable ob-

stacle. Colleagues have advised me many times, in the most friendly
way, to drop my research on accuracy because of precisely this prob-
lem. But I maintain that finding a criterion for use in accuracy research
is not really all that hard or all that mysterious if one simply remem-
bers the lessons of a classic article you probably read in graduate school,
if not as an undergraduate: I refer to an article on "construct validity''
by Cronbach and Meehl (1955).
That landmark article, which can be read as a philosophical rebut-
tal to logical positivism and concrete operationalism, addresses the ques-
tion of how to evaluate a new personality test. If one had a new test of
"sociability," say, against what criterion could you prove it to be valid?
The answer, say Cronbach and Meehl, is that there is not one. Or, as I
sometimes like to change the emphasis, there is not one. Rather, all you
can do is correlate your new test with as much other information about
your research subjects as you can gather and hope that the pattern of
convergent relations among independent data sources will, in the end,
support the essential validity of your construct.
The task of assessing the accuracy of a personality judgment is
exactly the same. If a bunch of subjects rate each other on "sociability,"
by what criterion can we assess whether these judgments are actually
accurate? The answer you probably can see coming: There is not one;
there is not one. All you can do is gather all the information about
these subjects' behaviors that you can, including, I would hope, some
fairly concrete indices of social participation, and assess whether or not
the pattern of convergent relations among independent data sources
increases or decreases your confidence in the accuracy of the subjects'
judgments.
THE ACCURACY PROJECT
I have described elsewhere (e.g., Funder, 1989) a study of my own

that tries to study accuracy through this tactic. What we have done to
gather self-judgments and peer-judgments of the personalities of about
160 subjects. We have also administered a wide variety of personality
and ability inventories to these subjects. Finally, and perhaps most
interestingly, we have videotaped the behavior of each of these sub-
jects in three different laboratory situations. (In a further project just
beginning, we are also including reports and observations of subjects'
behavior in daily life.)
The data we have gathered allow us to assess various kinds of
agreement and disagreement between acquaintances' judgements of

personality and between acquaintances' judgments and self-judgments
(Funder & Colvin, in press). But they also allow us to assess the degree
of relationship between judgments of personality and at least some of
the actual behaviors of the persons who are judged. Our basic analytic
scheme is to examine the accuracy of personality judgment, by this
kind of criterion, as a function of four potential moderators: (1) "good
judge," the possibility that some persons are more accurate judges of
personality than are others; (2) "good target," the possibility that some
individuals are easier to judge than others; (3) "good trait," the possi-
bility that some traits are easier to judge or that some behaviors are
easier to predict than are others; and (4) "good information," the possi-
bility that the accuracy of judgment is affected by the type and amount
of information that is available to the judge.
It may not escape the notice of some readers that this is actually a
very old-fashioned kind of research. The basic design, which is to gather a
large amount of information about the personality of each subject and
then see how different kinds of information interrelate, is a strategy
that was used more than 50 years ago by Allport (1977) and also by
Murray (1938).
This fact may seem unsurprising to those cynics who believe that
research in personality tends to go around in circles. A little more
optimistically, a few years ago Jack Block (1968) suggested that at least
sometimes personality research might advance "helically," so that each
time we go around we have elevated ourselves a bit above where we
were before. However you prefer to look at it, I do believe that the
field of personality psychology could more forward just now, by moving
backward a bit and studying judgments of personality the old-fashioned
way: By taking them seriously as possible indicators of something that
might, just conceivably, really exist.
ACKNOWLEDGMENT: The author's research is supported by grant

MH42427 from the National Institute of Health.
REFERENCES
Asch, S. E. (1946). Forming impressions of personality. Joo:rool of Abnormal and Social
Block, J. (1968). Personality measurement: Overview. In D. L. Sills (Ed.), International
Encyclqpedia of the Social Sciences (Vol. 12, pp. 141-179). -New York: Macmillan
and Free Press.
214 DAVID C. FUNDER
Carlson, R. (1984). What's social about social psychology? Where's the person in person-
ality research? Joo:rnal of Persorw,lity and Social Psychology, 51, 1200-1207.
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psycho-
logical Bulletin, 52, 177-193.
of person perception: Do people know what kinds of impressions they convey? Jour-
nal of Personality and Social Psychology, 52, 303--315.
Funder, D. C. (1989). Accuracy in personality judgment and the dancing bear. In D. M.
Buss & N. Cantor (Eds.), Personality psychology: Recent trends and emerging di-
rections. New York: Springer-Verlag.
Funder, D. C., & Colvin, C. R. (1988). Friends and strangers: Acquaintanceship, agree-
ment, and the accuracy of personality judgment. Journal of Persorw,lity and Social
Funder, D. C., & Colvin, C. R. (in press). Congruence of self and others' judgments of
personality. In S. Briggs, R. Hogan, & W. Jones (Eds.), Handbook of persorw,lity
psychology. Orlando: Academic Press.
Funder, D. C., & Dobroth, K. M. (1987). Differences between traits: Properties associated
with interjudge agreement. Journal of Personality and Social Psychology, 52, 409-
418.
Gage, N. L. & Cronbach, L. J. (1955). Conceptual and methodological problems in inter-
personal perception. Psychological Review, 62, 411-422.
Gregory, R. L. (1971). Visual illusions. In R. C. Atkinson (Ed.), Contemporary Psychol-
ogy (pp. 167-177). San Francisco: Freeman.
Jones, E. E. (1985). Major developments in social psychology during the past five de-
cades. In G. Lindzey & E. Aronson (Eds.), The handbook of social psychology (3rd
ed., Vol. 1, pp. 47-107). New York: Random House.
Kenny, D. A., & LaVoie, L. (1984). The social relations model. In L. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 18, pp. 141-179). New York: Aca-
demic Press.
Nisbett, R., & Ross, L. (1980). Human inference: Strategies and shortcomings of social
judgment. New York: Prentice-Hall.
Taft, R. (1955). The ability to judge people. Psychological Bulletin, 52, 1-23.
Taft, R. (1966). Accuracy of empathic judgments of acquaintances and strangers. Psycho-
logical Bulletin, 52, 1-23.
CHAPTER SIXTEEN
Gordon Allport
Father and Critic of the Five-Factor Model
OLIVER P. JOHN AND RICHARD W. ROBINS
Over the past 50 years, Gordon Allport's views of personality, and of

personality psychology as a science, have had a guiding and pervasive
influence. In this chapter, we examine Allport's role in bringing about
one of the most significant empirical advances in the field. Allport and
Odbert's (1936) psycholexical study of English language personality de-
scriptors laid the empirical and conceptual groundwork from which the
Five-Factor Model (FFM) of personality eventually emerged. One might
therefore consider Allport one of the fathers of the FFM. Like many
fathers, however, he might not have approved wholeheartedly of his
offspring.
The FFM is an empirically derived classification of personality traits,
based on the intercorrelations among trait ratings across individuals.
At the most general (superordinate) level there are five relatively in-
dependent content domains, often labeled Extraversion (energetic,
sociable, assertive), Agreeableness (loving, pleasant, trusting), Consci-
entiousness (reliable, organized, efficient), Neuroticism (anxious, ner-
vous, worrying), and Openness (imaginative, curious, broad interests).
The history of the FFM has been reviewed by John, Angleitner, and
Ostendorf (1988), and several recent reviews have been devoted to the
OLIVER P. JOHN AND RICHARD W. ROBINS • Department of Psychology, University of

California, Berkeley, California 94704.
New York, 1993.
215
216 OLIVER P. JOHN AND RICHARD W. ROBINS
current empirical and conceptual status of this model (Digman, 1990;

Goldberg, 1990; John, 1990; McCrae & Costa, 1990).
To be sure, Allport did not anticipate this particular model, nor
would he have considered it a sufficient account of all functions and
purposes of personality description. Indeed, the FFM is not based on
idiographic methods, nor does it explicate the neuropsychic structures
Allport believed to underlie personality. Yet, although Allport was an
ardent proponent of idiographic approaches, he was not, as some have
portrayed him, an opponent of nomothetic approaches. Allport (e.g.,
1962) stated very clearly that psychology should aim to understand
both the common and the particular in human behavior; to emphasize
one to the exclusion of the other would improperly restrict the scope of
psychological science. "The psychology of personality, I have therefore
explicitly maintained, should be both nomothetic and idiographic ...
abstract dimensions have their place" (Allport, 1946, pp. 133-134).
Allport also recognized the need in the field of personality psychol-
ogy for "a satisfactory taxonomy of personality and its hierarchi-
cal structure" (Allport, 1968, p. 48). It is this need that the FFM
addresses.
ALLPORT-FATHER OF THE FIVE-FACTOR MODEL
To construct such a "satisfactory taxonomy," Allport and Odbert

(1936) followed Klages's (1926/1932) and Baumgarten's (1933) German
research and used the natural language as a starting place. The English
dictionary served as their source of personality attributes: "Each single
term specifies in some way a form of human behavior; each term is a
record of commonsense observation, inexact perhaps, but nevertheless
constituting an authentic problem for the science that has taken as its
task the purification and codification of commonsense views of human
nature." (Allport & Odbert, 1936, p. vi).
ALLPORT AND 0DBERT'S "PSYCHOLEXICAL STUDY"
Allport and Odbert (1936) selected personality-relevant terms from

the 550,000 terms in Webster's New International Dictionary. Terms
were included in the final list if they were judged to possess "the
capacity ... to distinguish the behavior of one human being from that of
another" (p. 24); thus, terms referring to common and nondistinctive
behavior were eliminated. With the addition of a few common slang
terms not (yet) included in Webster's, the final list amounted to almost
ALLPORT AND THE FIVE FACTORS 217
18,000 terms "designating distinctive forms of personal behavior. At

first this seems like a semantic nightmare. Yet, it is obvious that trait-
names bear some relation to the underlying structural units of per-
sonality, and it is our duty to discover, if we can, what this relation
is" (Allport, 1937, pp. 353-354). This task, which Allport and Odbert
(1936, p. vi) thought would keep a psychologist "at work for a life-
time," has indeed occupied personality psychologists for more than 50
years.
Allport and Odbert (1936) tried to bring some order to the seman-
tic nightmare they had created. They classified their 18,000 terms, on
conceptual grounds, into four categories. Only the first category con-
tained terms designating personal traits ("generalized and personalized
determining tendencies-consistent and stable modes of an individual's
adjustment to his environment"), such as sociable, aggressive, and in-
troverted (p. 26). The second category included temporary states,
moods, and activities, such as rejoicing, abashed, and elated. The third
category consisted of highly evaluative social and character judgments
of personal conduct and reputation, such as average, worthy, and irri-
tating. Although these terms presuppose some traits within the indi-
vidual, they do not indicate the specific attributes that give rise to an
individual's evaluation by others or by society in general. The last cat-
egory included physical characteristics, capacities and talents, terms of
doubtful relevance to personality, and terms that could not be assigned
to any of the other three categories.
Norman (1967) subsequently elaborated Allport and Odbert's ini-
tial classification and divided the domain into seven content categories:
Stable "biophysical" traits, Temporary states, Activities, Social roles,
Social effects, Evaluative terms, Anatomical and physical terms, as well
as Ambiguous and Obscure terms not intended for further consider-
ation. Allport and Odbert's and Norman's category systems illustrate
that the natural language of personality represented in dictionaries in-
cludes many different types of concepts. Individuals can be described
by their enduring traits (e.g., irascible), by the internal states they
typically experience (furious), by the physical states they endure (trem-
bling), by the activities they engage in (screaming), by the effects they
have on others (frightening), by the roles they play (murderer), and by
the general evaluations of their conduct by society (unacceptable, bad).
Moreover, individuals differ in their anatomical and morphological char-
acteristics (short) and in the personal and societal evaluations attached
to these appearance characteristics (cute).
REPLICATING ALLPORT AND 0DBERT IN GERMAN:

A PROTOTYPE MODEL
Allport and Odbert, as well as Norman, employed mutually exclu-

sive categories in their classifications and Allport and Odbert noted
that their "four-fold classification is at best only approximate and to a
certain extent arbitrary" (1936, p. 27). In addition, agreement among
the judges was not particularly high. An inspection of the classifications
quickly shows that the categories overlap and have fuzzy boundaries. This
observation led some researchers to conclude that distinctions between
classes of personality descriptors are arbitrary and should be abolished
(Allen & Potkay, 1981).
However, the "unclear cases" that fall on the boundaries between
categories create a problem only if one insists on classical definitions in
terms of necessary and sufficient attributes. Chaplin, John, and Gold-
berg (1988) presented an alternative, prototype conception where each
category is defined in terms of its clear cases rather than its bound-
aries; category membership need not be discrete but can be defined as
continuous. Chaplin et al. (1988) applied this prototype conception to
three classes of person descriptors, namely traits, states, and activities.
They found that altt•.:mgh the classification of a few descriptors was
difficult, the core of each category was distinct from each of the others
and could be differentiated by a set of conceptually derived attributes.
For example, prototypical states were seen as temporary, brief, and
externally caused; in contrast, prototypical traits were seen as stable,
long-lasting, and internally caused and needed to be observed more
frequently and across a wider range of situations than states before
they were attributed to an individual. These findings closely replicated
the classifications made by Allport and Odbert and by Norman and
generally confirmed that their initial conceptual definitions of traits and
states are widely shared.
Whereas the two previous studies had both examined American
English, Angleitner, Ostendorf, and John (1990) carried out a "psycho-
lexical study'' of the German personality vocabulary. Their study was
based on the prototype conception proposed by Chaplin et al. (1988)
and improved on the earlier studies of English in several methodologi-
cal respects. In particular, they used a continuous measure of pro-
totypicality for each descriptor in each category by employing ten
independent judges and also scored several types of reliability and va-
lidity indexes. The resulting German "personality lexicon" is much
more convenient than the unwieldy Allport and Odbert lists because
continuous prototypicality values are available for each person descrip-
tor in 13 different content categories. All terms can be listed in the

order of their membership in any of the categories, making it particu-
larly easy to select subsets of prototypical traits, states, and so on from
the total pool. In general, however, the findings in German were consis-
tent with the American ones and further demonstrated that the con-
ceptual distinctions initially made by Allport and Odbert are rooted in
a common understanding of personality.
REDUCING THE SEMANTIC NIGHTMARE: FIVE BROAD

DIMENSIONS UNDERLYING TRAIT TERMS
Allport and Odbert's four categories provided some structure but

did not eliminate the "semantic nightmare" created by the staggering
size and complexity of the four lists, which would give personality psy-
chologists bad dreams even today. A reduction and further organization
of the terms were therefore badly needed. Allport, however, felt it was
safer to err in the direction of overinclusiveness and argued that a
large number of fine-grained distinctions within each behavioral domain
was scientifically useful and necessary. For example, he considered the
more than 200 words related to politeness "a meager enough vocab-
ulary for the possible shadings and forms of polite behavior ...
'synonyms' should not be avoided; if anything they should be multi-
plied, in order to do more justice to the variety and number of those
overlapping dispositions" (Allport & Odbert, 1936, p. 34).
It is hard to fathom that Allport truly believed that 200 (!)terms
are insufficient for the scientific description of individual differences in
politeness. Although we appreciate his interest in the richness of the
unique and particular and his concern for detail, we also believe that
abstraction is necessary in science. One of the goals of scientific theo-
ries is a parsimonious representation of the most important aspects of
the subject matter. In fact, despite his advocacy of specificity in the
study of personality descriptors, Allport also anticipated the proce-
dures that eventually would lead to the discovery of the FFM:
Theoretically it would be possible to apply this ingenious method
[of factor analysis] to a complete list of trait-names, such as that
contained in this monograph. One might determine the amount of
overlap in meaning between all the terms as they are commonly
understood and employed. The investigator might then declare that
such and such trait-names are roughly synonymous and that only
one of them needs to be retained if what is desired is a vocabulary
of completely independent terms. The trait-names would be
grouped, and only a single representative would be saved for each

group. (Allport & Odbert, 1936, pp. 32-33, emphasis in original)
Even in this passage, however, Allport expressed his distrust of gener-
alization and felt compelled to remind his readers that the discovery of
a few broad dimensions is but one goal of personality research.
Inspired by the blueprint contained in the above quote and less
wary of the dangers of generalization, Cattell (1943, 1945a,b) reduced
Allport and Odbert's list of 4500 trait terms to a mere 35 trait vari-
ables, using both semantic and empirical clustering procedures (for a
review, see John et al., 1988). If Allport was too wary of abstraction,
Cattell (who eliminated more than 99% of the terms Allport had so
tenaciously defended) could have used some of Allport's fastidiousness.
In Cattell's defense, however, it should be mentioned that the small
number of variables was dictated primarily by the data analytic limita-
tions of his time, which made factor analyses of large variable sets
prohibitively costly and complex. On the basis of several oblique factor
analyses of this small set of variables, Cattell concluded that he had
identified 12 personality factors, which were eventually incorporated in
his Sixteen Personality Factors Questionnaire (Cattell et al., 1970).
Subsequent investigators, however, failed to replicate Cattell's factors
(see Digman & Takemoto-Chock, 1981; John, 1990).
At the same time as Cattell, Eysenck (1947; 1952) and Guilford
(1959; Guilford & Zimmerman, 1956) developed and promoted their own
models of personality structure, which differed from Cattell's and from
each other, both in the number and the nature of the factors. Although
Allport was discouraged by these incompatible factor models, he re-
mained convinced that "scalable dimensions are useful dimensions, and
we hope that work will continue until we reach firmer agreement con-
cerning their number and nature" (Allport, 1958, p. 252).
Twenty years later, Fiske (1978) expressed the view that little had
changed in the interim: "the empirical factors obtained by one investi-
gator are not congruent with those developed by any other researcher ...
no trend toward consensus on a standard set of conceptual dimensions,
either from a priori theorizing or from empirical analyses, is evident"
(pp. 14-15). Curiously, Fiske reached this conclusion, although he had
been the first (Fiske, 1949) to demonstrate a replicable five-factor solu-
tion in a subset of Cattell's variables. Fiske was also aware of Tupes
and Christal's finding that ''five relatively strong and recurrent factors
and nothing more of any consequence" (1961, p. 14) could be identified
in both Cattell's own data sets and in newly collected ones. And by
1963, Norman had published his now classic replication of these five
factors, which eventually became known as the "Big Five" (Goldberg,

1981)-a title chosen not to reflect their greatness but to emphasize
their extraordinary breadth and level of abstraction.
In fact, then, Fiske's appraisal of the evidence available in 1978
was entirely too negative, though symptomatic of the mood of the field
at the time. Mainstream personality psychologists either ignored the
five factors or rejected them as linguistic fictions because of their ori-
gin in the natural language. With few exceptions (e.g., Goldberg, 1976),
that perception endured, perhaps because the field was preoccupied
with issues such as the response style debate and the person-situation
debate, both of which raised considerable doubt about the feasibility
and sensibility of research on personality structure (Block, 1977).
THE EMPIRICAL BASIS OF THE FIVE FACTORS
As Block (1977) anticipated, the controversies of the late 1960s and

1970s did not produce a paradigmatic shift but led to a gradual im-
provement in the methodological quality and sophistication of the re-
search, especially in the realm of multivariate procedures (e.g., Everett,
1983). The field emerged from this period of self-criticism and doubt
with a renewed belief in the fundamental importance and scientific via-
bility of trait concepts.
RESEARCH IN THE LEXICAL TRADITION

In the early 1980s, Goldberg (1980, 1981) and Digman and Takemoto-
Chock (1981) demonstrated anew the empirical generalizability of the
Big Five factors. Similar five-factor structures were subsequently iden-
tified in sets of variables that were much broader and selected more
systematically from the dictionary than Cattell's 35 variables (Gold-
berg, 1990). Moreover, the Big Five dimensions also emerged when the
personality descriptions were made by experienced personality and
clinical psychologists (John, 1990).
The finding that experts and lay judges do not differ systemati-
cally in the factor structures of their personality descriptions may seem
surprising at first; one might expect experts' notions of personality to
differ from "folk concepts" (Tellegen & Waller, in press). However, it is
important to realize that the Big Five factors, although derived from
ratings of common language trait terms, are not folk concepts defined
as "everyday variables that ordinary people use in their daily lives to
understand, classify, and predict their own behavior and that of others"
(Gough, 1987, p. 1). Whereas each of the individual trait terms repre-
sented in the Big Five structure is a folk concept, the five factors
represent a higher level of abstraction. They were derived from inter-
correlations among trait ratings and, therefore, may not represent any
individual's folk theory. For example, the empirical discovery of the
Surgency (or Extraversion) factor implies that attributions of sociabil-
ity, energy, and dominance tend to covary across individuals, whether
or not the individual making these attributions is aware of these popu-
lation correlations. Nor does the discovery of the Big Five imply that
individual "folk" actually use these broad concepts when thinking about
and describing themselves or others. The difficulty undergraduates have
in remembering the five factors in their personality courses provides
testimony against the folk concept status of the Big Five.
RESEARCH IN THE QUESTIONNAIRE TRADITION
While the researchers in the lexical tradition were consolidating

the evidence for the Big Five, the need for an integrative framework
became more pressing among researchers who prefer to measure per-
sonality with questionnaire scales. Joint factor analyses of question-
naires developed by different investigators had shown that two broad
dimensions, Extraversion and Neuroticism, appear in one form or an-
other in most personality inventories. Beyond these "Big Two" (Wiggins,
1968), however, the various questionnaire-based models had shown few
signs of convergence. For example, Eysenck observed that "Where we
have literally hundreds of inventories, incorporating thousands of traits,
largely overlapping but also containing specific variance, each empirical
finding is strictly speaking only relevant to a specific trait .... This is
not the way to build a unified scientific discipline" (1991, p. 786).
Again, the situation began to change in the early 1980s. Costa and
McCrae had developed the NEO Personality Inventory to measure three
broad personality dimensions: Neuroticism, Extraversion, and Open-
ness (see McCrae & Costa, 1990, for a review). In 1983, however, they
realized that their NEO system closely resembled three of the Big Five
factors, but did not encompass traits in the Agreeableness and Consci-
entiousness domains. They therefore extended their model with scales
measuring Agreeableness and Conscientiousness, and demonstrated
that their five questionnaire scales converged with adjective-based
measures of the Big Five (McCrae & Costa, 1985). * Subsequent re-
*See McCrae and John (1992) for a discussion of differences between the lexically based
and the questionnaire-based factor structures.
search showed that these five factors could be recovered in various

other personality questionnaires, as well as in self-ratings on Block's
(1961) California Adult Q-Set (see McCrae & Costa, 1990).
THE EMERGING CONSENSUS AND

HEURISTIC POTENTIAL OF THE BIG FIVE
The convergence between the lexical and questionnaire approaches
led to a dramatic change in the acceptance of the five factors in the
field. With regard to their empirical status, the findings accumulated
since the early 1980s show that the five factors replicate across differ-
ent types of subjects, raters, and data sources, in both dictionary-based
and personality-questionnaire studies. Indeed, even the more skeptical
reviewers were led to conclude that "agreement among these descrip-
tive studies with respect to what are the appropriate dimensions is
impressive" (Revelle, 1987, p. 437; also Briggs, 1989; McAdams, 1992).
There seems to be an emerging consensus that the five-factor tax-
onomy "captures, at a broad level of abstraction, the commonalities
among most of the existing systems of personality det:~cription, and pro-
vides an integrative descriptive model of personality research" (John,
1990, p. 96).
The availability of a general and replicable classification of person-
ality traits and the emerging consensus it spawned precipitated an ex-
plosion of interest in the five dimensions. The heuristic and applied
value of this taxonomy is only now being realized. Researchers from
diverse areas of psychology have begun to use it as a framework to
organize and summarize their findings. For example, Barrick and Mount
(1991) reviewed the extensive literature on personality predictors of
job performance and found consistent trends in what previously had
been viewed as a morass of inconsistent findings: Regardless of the
type of job and performance outcome measures, scales related to the
conscientiousness factor predicted job performance. Scales related to
extraversion predicted performance only in jobs involving social con-
tact, such as sales; the other three factors were not systematically re-
lated to job performance.
CONCEPTUAL STATUS OF THE FACTORS:

FROM THE BIG FIVE TO THE FIVE-FACTOR MODEL
The change in the empirical status of the Big Five was accom-
panied by a shift in its conceptual status. So far, we have discussed
the discovery and the evidentiary basis of the five factors. How
should this empirical finding be interpreted? Some researchers be-
lieve that the Big Five constitute a theory of personality. For exam-
ple, Hogan and Hogan (1989) referred to the five dimensions as the
"'big five' theory" (p. 274), and Digman and Takemoto-Chock (1981)
described them as "an impressive theoretical structure" (p. 164).
Such claims are difficult to evaluate because the concept of theory,
especially in personality psychology, is so ambiguous (Chapter 7, this
volume).
Initially, the five factors were tied to the natural language of trait
description and, therefore, were properly interpreted as dimensions of
trait description or attribution (John et al., 1988), rather than as dimen-
sions of personality or traits. Strictly speaking, the Big Five structure
provides only a taxonomy of trait descriptions. Although these trait
descriptions are usually thought to reflect observable regularities in
behavior (e.g., Norman & Goldberg, 1966), the lexical tradition makes
no explicit assumptions (or claims) about the ontological status of the
trait terms or about the causal origin of the regularities to which the
trait terms refer. The Big Five structure was derived through purely
empirical and purposely atheoretical procedures; theoretical considera-
tions, such as questions about the existence and explanatory status of
traits, were deemed unimportant.
However, these questions are of considerable importance to
the Five-Factor Model. The FFM represents one interpretation of
the Big Five dimensions: Although the FFM adopts the Big Five as
its basic structural model, it also includes a number of theoretical
assumptions that do not necessarily follow either from the lexical
approach or from the empirical discovery of the five factors. The FFM
can be viewed as a particular instance of trait theory and makes the
following major claims (McCrae & Costa, 1990): Personality can be de-
scribed in terms of five broad content domains, each of which sub-
sumes several more narrowly defined and correlated subsets or
facets; individual differences in these domains are stable over long
periods of time, have a genetic basis, and derive in part from as yet
unspecified internal (e.g., physiological) mechanisms; individuals can
be described both in general terms by their scores on the five di-
mensions and in more detail by their scores on the larger number of
facets. It is now widely assumed that the FFM has the same concep-
tual status as other questionnaire-based models of personality struc-
ture advocated by trait theorists such as Eysenck (e.g., 1986),
Guilford (e.g., 1959), Cattell (e.g., Cattell et al., 1970), or Tellegen
(e.g., 1985).
In light of these differences, the FFM of personality traits should

be distinguished from the Big Five, an empirically derived taxonomy
of personality trait descriptors. Specifically, we suggest that the em-
pirically derived factors be referred to as the Big Five, to distin-
guish them from the FFM, which advances more extensive conceptual
claims.
ALLPORT-CRITIC OF THE FIVE-FACTOR MODEL
TRAIT THEORY AND PERSONALITY THEORY

The FFM is a structural model of individual differences and, like
most dimensional models of individual differences, provides an account
of personality that is primarily descriptive (rather than explanatory),
focused on variables (rather than on the individual person), and postu-
lates molar or generalized (rather than molecular) dimensions of per-
sonality. Allport, on the other hand, subscribed to a broader definition
of personality theory. He agreed with Kluckhohn and Murray (1953),
who argued that a theory of personality must address several goals: to
describe and to explain the ways in which people are (1) like all other
people (i.e., general or nomothetic laws), (2) like some other people (i.e.,
individual differences), and (3) like no other people (i.e., individual or
idiographic laws). In other words, individual differences are only one,
albeit important, part of personality. A theory of personality requires
more than a descriptive taxonomy of broadly defined dimensions of
individual differences.
Throughout his career, Allport emphasized three additional issues
a theory of personality should address: (1) the causal "neuropsychic"
structures that underlie personality, (2) the idiographic analysis of indi-
vidual lives, and (3) the richness contained in the nuances that can be
expressed by individual personality descriptors. In the remainder of
this chapter, we discuss each of these concerns, focusing on their im-
plications for an Allportian evaluation of the FFM.
DESCRIPTION AND EXPLANATION

Whether the Big Five dimensions are believed to reflect summa-
ries of behavior over time or causal mechanisms that exist within the
person depends on one's conception of personality traits. In general, we
can distinguish two conceptions of traits: (1) the summary (or categori-
cal) view, articulated by Wiggins (1974) and by Buss and Craik (1984),
and (2) the causal view espoused, for example, by Allport (1937) and
Cattell (1957). * Adherents of the summary view conceptualize traits as

descriptive summaries of an individual's past behavioral conduct and
refrain from invoking traits as causal or explanatory concepts. For ex-
ample, if Gordon has exhibited a certain frequency of fastidious acts in
the past, we can appropriately describe Gordon's behavior with the
trait label "fastidious."
The causal view holds that traits are causal entities corresponding
to as yet unknown neuropsychic structures. According to this view, to
assert that Gordon is "fastidious" is to infer an internal generative
mechanism that causes Gordon to behave in ways that we can label, in
common language terms, as fastidious. These two positions differ with
regard to the ontological status they accord the trait concept: The sum-
mary view does not regard traits as real entities, whereas the causal
view regards traits as entities that exist "in our skins" (Allport, 1968,
p. 49).
If one adopts a summary conception of traits, then the five dimen-
sions must be interpreted as descriptive, reflecting the fundamental
content domains within which observed regularities in behavior can be
described (see Hogan, 1983). For example, Hogan and Hogan (1989)
interpreted the five factors as ''broad dimensions of interpersonal eval-
uation" (p. 274), suggesting that the factors represent categories of
behavior that people find useful and convenient in describing sociocul-
turally important characteristics of others. This theoretical interpreta-
tion of the five factors is consistent with the core assumption of the
lexical approach that those individual differences that are most salient
and socially relevant in people's lives will eventually become encoded
into their language (Allport, 1937; see John et al., 1988).
In contrast, if one adopts a causal conception of traits, the five
dimensions must be based on underlying causal mechanisms, which re-
searchers should strive to discover. t This is the position Cattell (1957)
and Eysenck (1990) adopted in their own models of personality struc-
ture. Researchers in the FFM tradition, however, have not taken such
an explicitly causal stance toward the FFM and remain more circum-
spect: "Most contemporary psychologists would acknowledge that there
must be some neurophysiological or hormonal basis for personality, but
it is unlikely that we will ever fmd a single region of the brain that
*A third view (Ryle, 1949; see also Wright & Mischel, 1987; Zuroff, 1986) treats traits as
dispositions, defined as tendencies to behave in certain ways in certain kinds of situations.
Whether or not dispositions are causal concepts remains a matter of debate.
tThis causal interpretation does not imply that the five factors are isomorphic with the
underlying causal structures or that they constrain the number of these structures.
controls Neuroticism, or a neurotransmitter that accounts for Extra-

version, or a gene for Openness" (McCrae & Costa, 1990, p. 25).
Revelle (1987) has criticized the FFM for its "lack of any theoreti-
cal explanation for the how or why of these dimensions" (p. 437). Briggs
(1989) has expressed similar reservations about the atheoretical nature
of the FFM. McAdams (1992) addresses these concerns in some detail.
And Allport, no doubt, would have agreed wholeheartedly with all
three of them. We also agree that eventually a taxonomic system should
be predicated on causal and dynamic principles. However, the critics
overlook the fact that "in any science, the taxonomy precedes causal
analysis; we must analyze and classify the entities in our field of study
before we can frame meaningful theories concerning their behavior.
The astronomer classifies stars, the chemist elements, the zoologist
animals, the botanist plants; the student of individual differences must
do likewise" (Eysenck, 1991, p. 774).
Although descriptive and classificatory concerns have dominated
the work on the FFM, the five dimensions, most notably Extraversion
and Neuroticism, have been the target of various physiological and
mechanistic explanations (see Rothbart, 1989). Similarly, Block and
Block's (1980) notion of Ego Control may shed some light on the mech-
anisms underlying the Conscientiousness and Extraversion factors.
Likewise, Tellegen's (1985) interpretation of Extraversion and Neuroti-
cism as persistent dispositions toward thinking and behaving in ways
that foster positive and negative affective experiences promises to con-
nect the FFM with individual differences in affective functioning. Other
explanatory accounts may come from areas as diverse as evolutionary
biology, psychoanalytic theory, and motivational theories.
Whereas Allport emphasized the need to understand the neuropsy-
chic structures underlying traits, many personality psychologists be-
lieve that for most of their research they "need not understand the
physiological basis of traits" (McCrae & Costa, 1990, p. 25). Given the
limits of our current knowledge (see Eysenck, 1990, for a review), the
difference between descriptive and causal approaches may be more ac-
ademic than consequential. Indeed, "Allport's attempt to use traits as
explanatory terms was not especially useful. Allport's theory did little
more than describe the observed regularities in behavior. Allport's
traits may explain behavior in the sense that they identify the cause of
the behavior as an internal property of the organism ... but this expla-
nation is too vague to be satisfying or heuristically valuable" (Zuroff,
1986, p. 999).
Perhaps more useful at this juncture is the development of concep-
tual definitions of the factors. Thus far, the factors have been defined
extensionally, that is, by listing the most prototypical trait adjectives,

Q-sort items, or questionnaire scale for each factor (e.g., McCrae &
John, 1992). For example, the first factor of the Big Five can be defined
by adjectives such as talkative, assertive, active, energetic, outgoing,
outspoken, and so on (John, 1990: Table 3.2). This descriptive stage was
a necessary first step toward generally accepted factor definitions, and
some disagreements remain about the precise composition of the fac-
tors (McCrae & John, 1992).
However, such extensional definitions tell us little about the con-
ceptual basis for the particular constellation of personality traits cap-
tured by each factor. What are the features common to the traits of
each factor? One way to identify features is through detailed concep-
tual analyses of the behaviors to which the traits in each domain refer.
For example, in an analysis of traits from the Agreeableness domain,
John (1986) showed that traits related to altruism (e.g., kind, generous,
charitable) refer to behaviors that involve voluntary efforts intended to
benefit someone else without expecting reciprocity.
Such conceptual analyses lead to intensional definitions that can
provide the basis for a causal understanding of each factor. For exam-
ple, if we know that traits comprising the Agreeableness domain in-
volve the intention to benefit others, we can examine the motivational,
evolutionary, or physiological basis of such behaviors. If the FFM is to
be rooted in dynamic and psychological processes, a conceptual analysis
can help bridge the gap in our understanding between theoretically and
empirically derived models of personality structure.
NOMOTHETIC AND IDIOGRAPHIC
So far we have addressed Allport's concerns about the descriptive

and explanatory potential of individual difference dimensions. Although
Allport viewed individual differences as central to an understanding of
personality, he was also concerned with traits as they exist within the
individual, and he therefore emphasized the need for idiographic ap-
proaches in personality psychology.
The FFM is often described as a model of personality structure
(e.g., Digman, 1990). The term personality structure, however, has al-
ways had two, rather different meanings: one associated with the nom-
othetic, the other with the idiographic approach. The model of
personality structure specified by the FFM is a spatial representation
describing the major dimensions of individual differences based on the
intercorrelations among traits across individuals. Such a dimensional
model of personality structure, however, should not be confused with

the structure of personality as it exists within a particular individual:
Factors are simply a summary principle of classification of many
measures ... [they] offer scalable dimensions; that is to say, they
are common units in respect to which all personalities can be com-
pared. None of them corresponds to the cleavages that exist in any
single personality. (Allport, 1958, pp. 251-252)
In other words, nomothetic dimensions do not constitute a model
of personality structure if we mean by structure the particular configu-
ration, patterning, and dynamic organization of the individuals' total set
of characteristics (York & John, 1992). Idiographic analyses are needed
to elucidate the ways in which the five personality factors combine
within particular individuals; although not strictly idiographic, typo-
logical analyses represent a person-centered approach that can iden-
tify groups or subsets of individuals who have similar configurations of
characteristics and share the same basic personality structure (Block,
1971).
Research on the FFM has yet to examine the ways in which the
five factors combine into a coherent personality within individuals and
whether there exist groups or categories of individuals who have sim-
ilar configurations of these five characteristics. Allport believed that
each individual possesses a unique constellation of traits, some com-
mon, others individual; in his view, the functional relations among these
traits are unique for each individual. Taken as a whole, these relations
form a unified system, a personality, that is implicated in all the behav-
iors of the individual. Allport considered these unique psychological
structures and dynamics more useful for explaining behavior than the
general laws implicit in a nomothetically derived taxonomy: "Factors
fall far short of our demand for a doctrine of elements that will offer as
close an approximation as possible to the natural cleavages and individ-
ualized structural arrangements of each single life. The search must go
on" (Allport, 1937, p. 248).
Despite this pessimistic appraisal of the utility of factors, idio-
graphic analyses might benefit from employing the five content catego-
ries specified by the FFM in some of their analyses. In particular, it
seems entirely feasible (and potentially quite interesting) to apply the
five content domains to idiographic analyses of the personal accounts
contained in life history narratives (McAdams, 1989), the personal con-
structs generated in Kelly's (1955) REP test and in free self-descrip-
tions (John, 1989), or the personal projects (Little, 1989) and personal
strivings (Emmons, 1989) that individuals use to structure their lives
and work toward their goals. Although the five dimensions have not
been used in this way, their origin in the natural language makes
them ideally suited for applications in this type of person-centered
research.
MOLAR DIMENSIONS AND INDIVIDUAL TRAITS
As we have argued earlier, some of Allport's writings suggest that

he appreciated the need to reduce the semantic nightmare he created
with his list of 4500 trait terms and derive a more practicable system.
More often than not, however, he was extremely skeptical about the
prospect of seeing the rich and differentiated personality-descriptive
language reduced to a few broad dimensions. Exasperated by the ob-
servation that "the number of distinguishable qualities is so great and
the number of trait names so few," he worried about "certain statisti-
cally-minded psychologists" who "hold that the number of independent
variables in human nature is relatively small and determinable, and
they accordingly deny that each separate trait-name signifies some dis-
tinction so essential that it should not be neglected."* A factorial re-
duction of the personality vocabulary, Allport continued, is a mixed
blessing: ''Those who seek 'unique factors' would consider the accom-
plishment highly significant. Those, on the other hand, who hold a
theory of overlapping traits would not" (Allport & Odbert, 1936,
pp. 31-32).
The "theory of overlapping traits" has many adherents among con-
temporary psychologists, and the appeal of Allport's doctrine of speci-
ficity is often underestimated. Following Allportian tradition, the five
factors have been criticized for being too broad to capture all of the
variance in human personality (Briggs, 1989; Mershon & Gorsuch,
1988). For example, Briggs (1989) argued that "our success in reducing
the natural trait language to its basic dimensions should not be taken
to mean that these dimensions are optimal for description or predic-
tion ... the distinctions among constructs subsumed by any one of these
five factors are at least as important conceptually as the distinctions
among the five factors themselves" (pp. 251-252). Similarly, Allport once
described broad factors as resembling "sausage meat that has failed to
pass the pure food and health inspection" (1958, p. 251).
These objections overlook the fact that personality can be concep-
tualized at different levels of abstraction and breadth. Many trait domains
*This Allportian inclination to attend to and value the specific descriptive function that
each trait term might perform has also been highlighted by Buss and Craik (1985).
are hierarchically structured (Hampson et al., 1986). The five factors

indeed represent a rather broad level in the hierarchy of person des-
criptors (although not the highest possible level, as John et al., 1991, have
shown). The advantage of categories as broad as the five factors is their
enormous bandwidth, their disadvantage is their low fidelity: In any
hierarchical taxonomy, one loses information as one moves up hierarchical
levels.
The hierarchical level an investigator selects depends on the par-
ticular descriptive and predictive goals of the research. For example,
an idiographic analysis (e.g., a case study) of an individual life may
require numerous fme distinctions-perhaps even 200 words for polite-
ness. In contrast, for a nomothetic analysis (e.g., screening job ap-
plicants for selection interviews), a few broad dimensions might be
sufficient. At this point, the FFM is not well-specified at lower levels of
abstraction. The "facet" scales McCrae and Costa (1990) developed to
assess the next level below the five factors and Goldberg's (1990) hier-
archical analyses of trait adjectives are steps in the right direction.
Nevertheless, considerable conceptual and empirical work remains to
be done in order to achieve a more explicit and consensual specification
of the five factors at lower levels of abstraction.
Although incomplete as a fully hierarchical structure, the FFM is
already more than a classification of personality traits. It can also serve
as an integral part of construct validation because it provides us, as
Briggs (1989) notes, "with a framework for building rating scales and
omnibus inventories that will represent the domain of personality
terms broadly and systematically ... (and) allows us to locate the seem-
ingly boundless supply of new constructs and measures within a known
configuration" (p. 247). Eventually, a comprehensive hierarchical taxon-
omy will permit the placement of all personality trait constructs into an
overarching nomological network. This taxonomy, when supplemented
with general propositions about the relations among personality con-
structs and between personality constructs and behavior, can facilitate
an understanding of the structures of personality that are believed to
underlie traits.
Allport recognized the importance of such a hierarchical system:
"Ultimately, of course, our hope is to be able to reduce molar units to
molecular and, conversely, to compound molecular units into molar''
(Allport, 1958, p. 241). The development of a hierarchical taxonomy of
personality descriptive constructs should ultimately advance our under-
standing of the patterned coherence of individual personalities to which
Allport devoted his life's work.
CONCLUSIONS
We hope that these excursions into Allport's diverse and influential

writings have served to illustrate a major theme in his psychology.
Again and again, Allport emphasized that personality research must be
both descriptive and explanatory, both nomothetic and idiographic,
both molar and molecular: "No doors should be closed in the study of
personality" (Allport, 1946, pp. 133-134). At this point in its develop-
ment, the Five-Factor Model of personality traits can satisfy only the
descriptive, the nomothetic, and the molar goals articulated by Allport.
Many contemporary criticisms of the FFM address long-standing All-
portian interests in explanatory, idiographic, and molecular contextual
accounts of personality. Allport thus continues to define the broad agenda
of personality and personological research for the years to come.
ACKNOWLEDGMENTS: The preparation of this chapter was supported, in

part, by National Institute of Mental Health grants MH-49255 and MH-
43948. The second author was supported by a National Science Founda-
tion Graduate Fellowship. The support and resources provided by the
Institute of Personality and Social Research at the University of Cali-
fornia, Berkeley, are also gratefully acknowledged. We would like to
thank Kenneth H. Craik, Lewis R. Goldberg, Jeffre M. Jackson, Sarah
E. Hampson, Gerald A. Mendelsohn, Myron Rothbart, Gerard Saucier,
Auke Tellegen, Kristina Whitney, and Raymond N. Wolfe for their com-
ments and suggestions on an earlier draft.
REFERENCES
Allen, B. P., & Potkay, C. R. (1981). On the arbitrary distinction between states and
traits. Journal of Perscmality and Social Psychology, 41, 916-928.
Allport, G. W. (1937). Perscmality: A psychological interpretation. New York: Holt.
Allport, G. W. (1946). Personality psychology as science: A reply. Psychological Review,
53, 132-135.
Allport, G. W. (1958). What units shall we employ? In G. Lindzey (Ed.), Assessment of
human motives (pp. 238-260). New York: Reinhart.
Allport, G. W. (1962). The general and the unique in psychological science. Journal of
Allport, G. W. (1968). Traits revisited. In The person in psychology: Selected essays by
Gordon W. Allport (pp. ~6). Boston: Beacon Press.
Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study. Psychological
Monographs, 47, (No. 211).
Angleitner, A., Ostendorf, F., & John, 0. P. (1990). Towards a taxonomy of personality
descriptors in German: A psycho-lexical study. European Journal of Personality, 4,
89-118.
Barrick, M. R., & Mount, M. K. (1991). The Big Five dimensions and job performance: A
meta analysis. Personnel Psychology, .U, 1-26.
Baumgarten, F. (1933). Die Charaktereigenschaften. [The character traits.] In Beitraege
zur Charakter- und Persoenlichkeitsforschung (No. 1). Bern: A. Francke.
Block, J. (1961). The Q-sort method in personality assessment and psychiatric research
(reprinted 1978). Palo Alto, CA: Consulting Psychologists Press.
Block, J. (1971). Lives through time. Berkeley, CA: Bancroft Books.
Block, J. (1977). Advancing the science of personality: Paradigmatic shift or advancing
the quality of research? In D. Magnusson & N. S. Endler (Eds.), Psychology at the
crossroad: Current issues in interactional psychology (pp. 37-63). Hillsdale, NJ:
Erlbaum.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the organi-
zation of behavior. In W. A. Collings (Ed.), Minnesota symposia on child psychology
(Vol. 13, pp. 39-101). Hillsdale, NJ: Erlbaum.
Briggs, S. R. (1989). The optimal level of measurement for personality constructs. In D.
M. Buss & N. Cantor (Eds.), Personality psychology: Recent trends and emerging
directions (pp. 246-260). New York: Springer-Verlag.
Buss, D. M., & Craik, K. H. (1984). Acts, dispositions, and personality. In B. A. Maher &
W. B. Maher (Eds.), Progress in experimental personality research (Vol. 13, pp. 241-
301). New York: Academic Press.
Buss, D. M., & Craik, K. H. (1985). Why not measure that trait? Alternative criteria for
identifying important dispositions. Journal of Personality and Social Psychology,
48, 934-946.
Cattell, R. B. (1943). The description of personality: Basic traits resolved into clusters.
Journal of Abnormal and Social Psychology, 38, 476-506.
Cattell, R. B. (1945a). The description of personality: Principles and findings in a factor
analysis. American Journal of Psychology, 58, 69-90.
Cattell, R. B. (1945b). The principal trait clusters for describing personality. Psychologi-
cal Bulletin, 42, 129-161.
Cattell, R. B. (1957). Personality and motivation structure and measurement. Yonkers-
on-Hudson, NY: World.
Cattell, R. B., Eber, H. W., & Tatsuoka, M. M. (1970). Handbook for the Sixteen Person-
ality Factor Questionnaire (16PF). Champaign, IL: Institute for Personality and
Ability Testing.
Chaplin, W. F., John, 0. P., & Goldberg, L. R. (1988). Conceptions of states and traits:
Dimensional attributes with ideals as prototypes. Journal of Personality and Social
Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual
Digman, J. M., & Takemoto-Chock, N. K. (1981). Factors in the natural language of
personality: Re-analysis and recomparison of six major studies. Multivariate Behav-
ioral Research, 16, 149-170.
Emmons, R. A. (1989). Exploring relations between motives and traits: The case of
narcissism. In D. M. Buss & N. Cantor (Eds.), Personality psychology: Recent
trends and emerging directions (pp. 32-44). New York: Springer-Verlag.
Everett J. E. (1938). Factor congruence as a criterion for determining the number of
factors. Multivariate Behavioral Research, 18, 197-218.
Eysenck, H. J. (1947). Dimensions of personality. London: Routledge and Kegan Paul.
Eysenck, H. J. (1952). The scientific study of personality. London: Routledge and Kegan
Paul.
Eysenck, H. J. (1986). Models and paradigms in personality research. In A. Angleitner,

A. Furnham, & G. Van Heck (Eds.), Personality psychology in Europe: Vol. 2. Cur-
rent trends and controversies (pp. 213-223). Lisse, The Netherlands: Swets &
Zeitlinger.
Eysenck, H. J. (1990). Biological dimensions of personality. In L.A. Pervin (Ed.), Hand-
book of personality: Theory and research (pp. 244-276). New York: Guilford Press.
Eysenck, H. J. (1991). Dimensions of personality: 16, 5, or 3?-Criteria for a taxonomic
paradigm. Personality and Individual Differences, 12, 773-790.
Fiske, D. W. (1949). Consistency of the factorial structures of personality ratings from
different sources. Journal of Abnormal and Social Psychology, 44, 329-344.
Fiske, D. W. (1978). Strategies for research in personality: Observation versus interpre-
tation of behavior. San Francisco: Jossey-Bass.
Goldberg, L. R. (1976). Language and personality: Toward a taxonomy of trait-descrip-
tive terms. Istanbul Studies in Experimental Psychology, 12, 1-23.
Goldberg, L. R. (1980, May). Some ruminations about the structure of individual differ-
ences: Developing a common lexicon for the major characteristics of human per-
sonality. Paper presented at the annual convention of the Western Psychological
Association, Honolulu.
Goldberg, L. R. (1981). Language and individual differences: The search for universals in
personality lexicons. In L. Wheeler (Ed.), Review of personality and social psychol-
ogy (Vol. 2, pp. 141-165). Beverly Hills, CA: Sage.
Goldberg(L. R. (1990). An alternative "description of personality": The Big-Five factor
structure. Journal of Personality and Social Psychology, 59, 1216-1229.
Gough, H. G. (1987). Administrator's Guide to the California Psychological Inventory.
Palo Alto, CA: Consulting Psychologists Press.
Guilford, J. P. (1959). Personality. New York: McGraw-Hill.
Guilford, J. P., & Zimmerman, W. S. (1956). Fourteen dimensions of temperament. Psy-
chological Monographs, 70, (10, No. 417).
Hampson, S. E., John, 0. P., & Goldberg, L. R. (1986). Category breadth and hierarchical
structure in personality: Studies of asymmetries in judgments of trait implications.
Hogan, R. (1983). A socioanalytic theory of personality. In M. M. Page (Ed.), Nebraska
Symposium on Motivation, 1982: Personality-Current theory and research. Lin-
coln: University of Nebraska Press.
Hogan, J., & Hogan, R. (1989). How to measure employee reliability. Journal of Applied
John, 0. P. (1986). How shall a trait be called?: A feature analysis of altruism. In A.
Angleitner, A. Furnham, & G. van Heck (Eds.), Personality psychology in Europe:
Current trends and controversies (pp. 117-140). Berwyn: Swets North America.
John, 0. P. (1989). Toward a taxonomy of personality descriptors. In D. M. Buss & N.
Cantor (Eds.), Personality psychology: Recent trends and emerging directions
(pp. 261-277). New York: Springer-Verlag.
John, 0. P. (1990). The "Big Five" factor taxonomy: Dimensions of personality in the
natural language and questionnaires. In L. A. Pervin (Ed.), Handbook of personality
theory and research (pp. 66-100). New York: Guilford Press.
John, 0. P., Angleitner, A., & Ostendorf, F. (1988). The lexical approach to personality: A
historical review of trait taxonomic research. European Journal of Personality, 2,
171-203.
John, 0. P., Hampson, S. E., & Goldberg, L. R. (1991). The basic level in personality trait
hierarchies: Studies of trait use and accessibility in different contexts. Journal of

Persorw,lity and Social Psychology, 60, 348-361.
Kelly, G. A. (1955). The psychology ofpersorw,l constructs (Vols. 1, 2). New York: Norton.
Klages, L. (1932). The science of character. London: George Allen & Unwin. (Original
Kluckhohn, C. M. & Murray, H. A. (1953). Persorw,lity in nature, society, and culture.
New York: Knopf.
Little, B. R. (1989). Personal projects analysis: Trivial pursuits, magnificent obsessions,
and the search for coherence. In D. M. Buss and N. Cantor (Eds.), Personality
psychology: Recent trends and emerging directions (pp. 15-31). New York: Springer-
Verlag.
McAdams, D. P. (1989). The development of a narrative identity. In D. M. Buss and N.
Cantor (Eds.), Personality psychology: Recent trends and emerging directions
(pp. 160--174). New York: Springer-Verlag.
McAdams, D. (1992). The Five-Factor Model in personality: A critical appraisal. Journal
of Persorw,lity, 60, 329-361.
McCrae, R. R., & Costa, P. T. (1985). Updating Norman's adequate taxonomy: Intelli-
gence and personality dimensions in natural language and in questionnaires. Journal
of Persorw,lity and Social Psychology, 49, 710-721.
McCrae, R. R., & Costa, P. T. (1990). Persorw,lity in adulthood. New York: Guilford.
McCrae, R. R., & John, 0. P. (1992). An introduction to the Five-Factor Model and its
applications. Journal of Personality, 60, 175-215.
Mershon, B., & Gorsuch, R. L. (1988). Number of factors in the personality sphere: Does
increase in factors increase predictability of real-life criteria? Journal of Personality
and Social Psychology, 55, 675-680.
Norman, W. T. (1967). 2,800 personality trait descriptors: Normative operating character-
istics for a university population. Ann Arbor: Department of Psychology, Univer-
sity of Michigan.
Norman, W. T., & Goldberg, L. R. (1966). Raters, ratees, and randomness in personality
structure. Journal of Personality and Social Psychology, 4, 681-691.
Revelle, W. (1987). Personality and motivation: Sources of inefficiency in cognitive per-
formance. Journal of Research in Personality, 21, 436-452.
Rothbart, M. K. (1989). Temperament and development. In G. A. Kohnstamm, E. Bates,
& M. K. Rothbart (Eds.), Handbook of temperament in childhood (pp. 188-247).
Chichester, England: Wiley.
Ryle, G. (1949). The concept of mind. New York: Barnes & Noble.
Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing
anxiety, with an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.),
Anxiety and the anxiety disorders (pp. 681-716). Hillsdale, NJ: Erlbaum.
Tellegen, A., & Waller, N. G. (in press). Exploring personality through test construction:
Development of the Multidimensional Personality Questionnaire. In S. R. Briggs &
J. M. Cheek (Eds.), Persorw,lity measures: Development and evaluation (Vol. 1).
Greenwich, CN: JAI Press.
Tupes, E. C., & Christal, R. C. (1961). Recurrent personality factors based on trait
ratings (Tech. Rep. No. ASD-TR-61-97). Lackland Air Force Base, TX: U.S. Air
Force.
Wiggins, J. S. (1968). Personality structure. Annual Review of Psychology, 19, 293-350.
Wiggins, J. S. (1974). In defense of traits. Unpublished manuscript, University of British
Columbia, Vancouver, Canada.
Wright, J. C., & Mischel, W. (1987). A conditional analysis of dispositional constructs: The
local predictability of social behavior. Journal of Personality and Social Psychology,

53, 1159--1177.
York, K. L., & John, 0. P. (1992). The four faces of Eve: A typological analysis of wo-
men's personality at midlife. Journal of Personality and Social Psychology, 63, 494-
508.
CHAPTER SEVENTEEN
To Predict Some of the People

More of the Time
Individual Traits and the Prediction of Behavior
PETER BORKENAU
Walter Mischel's (1968) book, Personality and Assessment, was ex-

tremely important for the development of personality theory and re-
search in the last two decades. Mischel pointed to the modest
cross-situational consistency of behavior that he interpreted as a sign
of situational specificity. Moreover, he pointed to the low validity of
personality measures in predicting isolated acts of subjects. What he
did not mention, however, was that Gordon Allport (1937) had already
opposed similar arguments by Hartshorne and May (1928) some de-
cades earlier. Whereas Mischel (1968) referred to Hartshorne and
May's study at length, he did not mention Allport's counterarguments.
It was mainly Allport who had imaginatively defended the concept of
traits against the objections of those who proposed to analyze individ-
ual differences at the lower level of habits instead of traits.
Mischel's (1968) theses were first responded to by some theoretical
papers. Bowers (1973), for instance, criticized situationism and sug-
gested interactionism as an alternative, whereas Wachtel (1973)
suggested that personality may be more important for the choice of sit-
uations than for the responses that are elicited from subjects in experi-
PETER BoRKENAU • Department of Psychology, University of Bielefeld, W-4800

Bielefeld 1, Germany.
New York, 1993.
237
238 PETER BORKENAU
mental settings. Among the first empirical studies designed with Mischel's
critique in mind were Jaccard's (1974) on single-act and multiple-act
criteria and Bern and Allen's (1974) on individual differences in cross-
situational consistency. Bern and Allen (1974) referred to Allport's work
at length. The present chapter explores how their approach worked
empirically and the extent to which it embodied Allport's ideas.
ALLPORT'S CRITIQUE OF HARTSHORNE AND MAY
Hartshorne and May (1928) examined the consistency of personal-

ity among schoolchildren. In an ambitious investigation of moral char-
acter, they arranged situations in which the honesty of subjects could
be observed. Of major importance, in the present context, was their
finding that the average correlation among their various tests of hon-
esty was .23. Hartshorne and May must somehow have felt the ambigu-
ity of this finding. Usually, they emphasized the specificity of behavior,
arguing, for instance, "that honesty or dishonesty is not a unified char-
acter trait in children of the ages studied, but a series of specific re-
sponses to specific situations" (Hartshorne et al., 1929, p. 243). At other
places, however, they admitted some support for a more global concept
of honesty as a trait:
Just as one test is an insufficient and unreliable measure in the
case of intelligence, so one test of deception is quite incapable of
measuring a subject's tendency to deceive. That is, we cannot pre-
dict from what a pupil does on one test what he will do on another.
If we use ten tests of classroom deception, however, we can safely
predict what a subject will do on the average whenever ten similar
situations are presented. (Hartshorne & May, 1928, p. 135)
Thus Hartshorne and May obtained findings in support of the
aggregation principle, implying that behavior is situationally specific
at the level of single activities but consistent at the dispositional level.
This view has more recently been emphasized by Epstein (1979;
Epstein & O'Brien, 1985). Other readers of Hartshorne and May, how-
ever (e.g., Mischel, 1968), cited their study as unequivocal support for a
situationist stance. This conclusion is not unambiguously supported by
Hartshorne and May's data, but it is in agreement with their general
claim that alleged traits, such as deception, helpfulness, cooperative-
ness, persistence, and self-control, are "groups of specific habits rather
than general traits" (Hartshorne et al., 1930, p. 1).
By contrast, Allport (1937) emphasized that individuals differ not
only in the degree to which they show common traits, but also in terms
INDIVIDUAL TRAITS 239
of which traits are even relevant. "Strictly speaking, no two persons

ever have precisely the same trait. Though each of two men may be
aggressive (or esthetic), the style and range of the aggression (or es-
theticism) in each case is noticeably different" (Allport, 1937, p. 297).
Allport's definition of personality traits makes allowance for this no-
tion. He defines a trait as a "generalized and focalized neuropsychic
system (peculiar to the individual), with the capacity to render many
stimuli functionally equivalent, and to initiate and guide consistent
(equivalent) forms of adaptive and expressive behavior" (Allport, 1937,
p. 295). Thus each individual may have his idiosyncratic equivalence
classes of situations and behaviors. Consequently, Allport criticized the
use of factor analysis (Allport, 1961, p. 331), a technique that presumes
that all dimensions are equally useful for describing each individual and
that individual differences are appropriately accounted for by different
factor scores on the same set of underlying dimensions (see Eysenck &
Eysenck, 1980, for this view). Taking Allport's perspective, the cross-
situational consistency coefficients reported by Hartshorne and May
are not particularly meaningful. "The low correlations between the
tests employed prove only that children are not consistent in the same
way, not that they are inconsistent with themselves" (Allport, 1937,
p. 250; italics in the original). By this he meant that a particular behav-
ior may serve various traits, the traits being more stable than cross-
situational correlations indicate:
It may be that child A steals pennies because he has a consistent
personal trait of bravado based upon his admiration for the gang-
sters he reads about in the tabloids and sees on the screen; child B
steals because he has a persistent interest in tools and mechanics
that drives him to buy more equipment than he can honestly afford;
child C, suffering from a gnawing feeling of social inferiority, steals
pennies to purchase candy to buy his way into favor with his play-
mates. Child D does not steal pennies, but he lies about his cheat-
ing, not because he has a general trait of dishonesty, but because he
has a general trait of timidity (fear of consequences); child E lies
because he is afraid of hurting the feelings of the teacher whom he
adores; child F lies because he is greedy for praise. Each of these
children behaved as he did toward these tests, not because he had
specific habits,but because he had some deep-lying and characteris-
tic trait. All that the Character Education Inquiry discovered was
that the particular trait of honesty, as defmed in the usual ethical
terms and tested in various conventional situations, was not one of
which the children possessed constant individual degrees, especially
in the face of perhaps a stronger tendency of each child to express
some trait other than honesty through the behavior of lying and
240 PETER BORKENAU
stealing. The children did not all have the same trait, but they had
nevertheless their own traits. (Allport, 1937, p. 251-252; italics in
the original)
Thus Allport suggested that it is not situational factors that ac-
count for the low cross-situational consistency of particular behaviors.
Rather, in this passage he is claiming that persons are consistent with
themselves and implying that such consistency is what enables oth-
ers-acquaintances and psychologists alike-to predict their behaviors
from knowledge of their individual dispositions. Cross-situational con-
sistency estimates, however, presume that all subjects do possess ex-
actly the same trait, albeit to different degrees.
Furthermore, Allport points out that there may be qualitatively
important mismatches between subjects' actual traits and the trait con-
jectured by the investigator; these will, of course, tend to reduce the
apparent consistency of personality.
The error of probing for consistency in the wrong place (and failing
to find it, pronouncing in favor of specificity) has been likened by G.
B. Watson to the absurdity of asking whether a person using the
public library has a trait causing him to take out only books with
red or with blue covers. Of course he hasn't. If only the bindings
were studied, no consistency should be expected. But if the subject-
matter of the chosen books was investigated, well-organized traits
of interest would appear. (Allport, 1937, p. 256; italics in the original)
And concerning the problem of misinterpreting individual traits as
situational specificity, he stated: "Statistical methods are ordinarily ap-
plied only to those variables to which all people may be ordered. If
many people do not happen to fit the variable then the illusion of spec-
ificity results" (Allport, 1937, p. 256).
BEM AND ALLEN'S ELABORATION OF ALLPORT'S IDEAS
Bern and Allen (1974) acknowledged that previous studies had

found cross-situational consistency to be discouragingly low. They real-
ized that this pattern of findings might be due in part to a mismatch-
for some subjects at least-between the trait under investigation and
the trait actually operating to determine the behavior of interest. Thus,
despite the low cross-situational consistency coefficients obtained in
psychological research, lay persons believe that individuals are consis-
tent because human intuition operates on idiographic rather than on
nomothetic assumptions. Individuals appear as consistent because
there is no implicit assumption that all individuals, albeit to a different
extent, should have the same traits. Rather, trait words are usually
selected from the huge vocabulary of thousands of trait-descriptive
terms (Allport & Odbert, 1936) such that they fit the specific equiva-
lence classes of the individual being described. For instance, we distin-
guish between persons who can be counted on to keep their promises
and persons who are habitually punctual. But if a person has one of
these traits and not the other, we do not speak of him as inconsistently
dependable. We do not first impose a trait term and then modify it by
describing the instances that fail to fall into that equivalence class.
Rather, we attempt first to organize the person's behaviors into ratio-
nal sets and label them afterward.
All this is a paraphrase of Allport's ideas, as Bern and Allen readily
admitted. However, Bern and Allen also suggested a procedure to over-
come the main shortcoming of Allport's idiographic approach; its inabil-
ity to yield testable hypotheses. This is probably the main reason why
Allport's ideas, although documenting an ingenious insight, did not in-
spire a considerable research program (Allport, 1966). Research psy-
chologists are simply not interested in investigating the personalities of
particular ordinary individuals. Rather, they search for general princi-
ples. The main contribution of Bern and Allen's (1974) article, therefore,
was its conceptualizing and initiation of a research program rooted in
Allport's (1937) ideas. The kernel of this program was: "Separate those
individuals who are cross-situationally consistent on the trait dimension
and throw the others out .... we believe that the rewards for this small
idiographic commitment can even be paid in the sacred coin of the
realm: bigger correlation coefficients" (Bern & Allen, 1974, p. 512).
Bern and Allen (1974) tested their prediction for two personality
traits, friendliness and conscientiousness. Subjects completed an
86-item personality inventory, rated themselves on two traits (e.g.,
"In general, how friendly and outgoing are you?"), and indicated their
cross-situational variability for these traits (e.g., "How much do you
vary from one situation to another in how friendly and outgoing you
are?"). Ratings of friendliness and conscientiousness by parents and
peers and several behavior observations served as criteria for cross-
situational consistency. Two measures of trait variability were used:
(1) the self-rated situational variability and (2) a so-called ipsatized
variance index (lVI), which was the ratio of the individual's variance on
the multiple-item scale of the trait and the variance of all responses to
the 86-item inventory. The effects of self-reported variability were as-
sessed by: (1) correlations among ratings of various judges (self, par-
ents, peers, etc.), (2) correlations among observed behaviors, and
(3) correlations between judges' ratings and observed behaviors. It
242 PETER BORKENAU
turned out that the response to the "How much do you vary" item for
friendliness, but not the corresponding lVI, predicted individual differ-
ences in consistency for friendliness. In contrast, the lVI but not the
response to the "How much do you vary" item for conscientiousness
predicted individual differences in consistency for conscientiousness.
Bern and Allen's (1974) study was intriguing and widely cited. There
are more than 500 citations listed in the Social Science Citation Index
up to 1991. Most of the citations welcomed this fresh approach to the
study of personality. The more critical comments, however, referred to
two main problems: the one theoretical and the other one empirical-
whether Bern and Allen's study was really idiographic and whether the
phenomenon could be replicated. I will discuss these two issues in turn.
WAS BEM AND ALLEN'S STUDY TRULY IDIOGRAPHIC?
The answer to the first question is clearly negative, as already

admitted by Bern (1983). According to Allport (1937), ''idiographic" re-
fers to the unique organization of traits within individuals, whereas
Bern and Allen intended to distinguish between persons to whom a
trait is more or less applicable. Though not idiographic in the strict
sense, however, Bern and Allen's approach does offer a way of testing
one of Allport's ideas.
Allport distinguished between individual traits and common traits:
"Common traits are those aspects of personality in respect to which
rrwst mature people within a given culture can be compared" (Allport,
1937, p. 300, italics added). Common traits cannot provide a full de-
scription of a person's dispositions; but they are important because
they permit individuals to be compared on common scales. Moreover,
Allport emphasized the distinction between persons who can be com-
pared on a common trait and others who cannot. With regard to hon-
esty, for instance, he wrote:
There are honest people, dishonest people, and atypical people-
honest in most respects, but not always capable of resisting temp-
tation .... Honesty is either a general characteristic or a set of
specific habits ... depending also upon your method, and upon the
particular individual you happen to be studying. (Allport, 1937, p. 255)
Thus it may be that Bern and Allen identified those subjects who
were "typical" with regard to the dimensions under study. They had
then derived testable hypotheses from one of Gordon Allport's ideas. To
state, however, that their study was an instance of the classic modera-
tor variable strategy instead of bearing on the idiographic-nomothetic
distinction (Mischel & Peake, 1982, footnote 8) confuses the theoretical

side with the methodological side of the issue. The moderator variable
strategy is a particular statistical method of data analysis that may be
used to investigate an unlimited number of psychological theories and
hypotheses (Saunders, 1956; Tellegen et al., 1982).
One avoidable shortcoming of Bern and Allen's study, however, was
the choice of their moderator variable. They sought to identify those
subjects who were cross-situationally consistent by asking subjects
how much they varied from one situation to another in how friendly or
conscientious they were. By operationalizing consistency in this man-
ner, Bern and Allen appear to have been using three concepts-trait
applicability, cross-situational consistency, and lack of situational vari-
ability-interchangeably, thus blurring some important distinctions. A
trait may be inapplicable for describing a person, not so much because
it does not fit the equivalence class of situations for an individual, but
rather because it does not fit the person's equivalence class of behav-
iors. Bern and Allen provide a good illustration by their example of a
student so dedicated to study that he has time for little else and is
therefore negligent in most other areas of his life. This student would
more appropriately be classified as achievement oriented rather than
as conscientious, independent of the situational variability of his
achievement-related activities. This is because the equivalence class of
behaviors referred to by the term "conscientious" refers not only to
achievement-related behavior, but also to behaviors like keeping one's
home in order, being meticulous about grooming, and so on. Thus trait
applicability and cross-situational consistency are related but different
concepts.
It is also necessary to consider cross-situational consistency as sep-
arate from situational variability. Magnusson (1976) noted the distinc-
tion between absolute consistency and relative consistency. Absolute
consistency implies that situations make no difference at all, whereas
relative consistency implies that the average behavior of a subject sam-
ple may change from one situation to another, with the rank order of
subjects presumably remaining the same across situations. For in-
stance, most persons are likely to be more anxious while giving a
speech to a large audience than while talking to a friend. The mean
anxiety level is therefore different in the two situations. It may well
be, however, that the individuals who are more anxious while talking to
a friend are the same ones as those who are especially anxious when
giving a speech to a large audience. Now, consider that subjects are
asked how much their trait varies from one situation to another. It is
likely that they consider their absolute consistency, whereas Bern and
244 PETER BORKENAU
Allen intended to assess their relative consistency. I submit that rela-

tive consistency is difficult to ask for without long-winded explanations
of statistical concepts. This difficulty, however, is easily bypassed by
asking subjects the more accurate question of how applicable, relevant,
or important a trait is to describe their personality. Trait applicability
was asked for by Borkenau and Amelang (1985), trait relevance was
asked for by Zuckerman and associates (Zuckerman et al., 1988, 1989)
and trait importance was asked for by Cheek (1982). These and other
recent studies of the variables that influence self x peer agreement are
attempts to replicate and to extend the results of Bern and Allen's
(1974) study.
THE REPLICABILITY OF BEM AND ALLEN'S FINDINGS
TRAIT RELEVANCE OR TRAIT EXTREMITY?
Studies of trait variability and trait importance as moderators face

the methodological problem that the less-variable subjects, or those for
whom the trait is more important, are those who are either extremely
low or extremely high on that trait. Variability and importance are
therefore confounded with trait extremity (Paunonen, 1988; Rushton et
al., 1981; Stones & Burt, 1978). More extreme values on both poles of a
dimension, however, imply larger standard deviations and, other things
being equal, higher correlations. Higher correlations for the less vari-
able group, or for those subjects for whom a trait is more important,
can therefore simply reflect the biased sampling of more extreme cases.
This was a major argument advanced by critics of Bern and Allen's
study (Chaplin & Goldberg, 1984; Paunonen & Jackson, 1985; Rushton
et al., 1981).
Two techniques have been used to overcome this problem. Bern
and Allen (1974) conducted separate median splits for subjects at each
of the seven points of the trait rating scale. Thus subjects who had a
moderate trait rating for friendliness needed a higher self-rated vari-
ability than subjects with an extreme rating on the friendliness dimen-
sion in order to be assigned to the more variable subgroup. Thus
extremity and variability were unconfounded by adjusting the distribu-
tions of trait ratings within the more- and the less-variable subgroups.
The other approach to the problem is to conduct regression analyses or
moderated regression analyses (Borkenau & Amelang, 1985; Tellegen et
al., 1982) instead of computing correlations. In contrast to correlations,
regression coefficients do not increase with the variance of the vari-

ables (Hanushek & Jackson, 1977).
WHO ARE THE TRAIT-CONSISTENT SUBJECTS?
Bern and Allen (1974) reported correlations among trait ratings,

correlations among behavior observations, and correlations between trait
ratings and behavior observations. Moreover, they compared the in-
traindividual variances of the friendliness and conscientiousness meas-
ures for the variable and unvariable subjects. Thus there is a problem
as to whether these measures of actual consistency may all be used
interchangeably. Mischel and Peake (1982) claimed that it was the cor-
relations among behavior observations only that indicated cross-situational
consistency of behavior. This is appropriate as far as cross-situational
consistency is concerned. But what measures are appropriate to iden-
tify those persons who are "typical" for a trait according to Allport?
It is likely to be the correlations among ratings by various knowl-
edgeable informants that best indicate the applicability of a trait. Low
correlations (or high intraindividual variances) among judges are likely
to indicate that the informants formed different impressions of the tar-
gets' behavior. Consider once more the totally dedicated student who
finds no time to keep his apartment in order. He may be rated by some
informants as extremely conscientious and by others as not at all con-
scientious, depending on which aspect of conscientiousness the judges
choose to focus on. Because the target is inconsistently conscientious,
judges' ratings of his conscientiousness are unlikely to agree. The same
argument applies to the equivalence class of situations. If subjects are
more conscientious than others in some situations but less conscien-
tious in other situations, various judges who observe the targets in
different settings will probably form different impressions.
By contrast, correlations among behavior observations face the
problem that targets must have the trait as operationalized by the
researcher in order that high cross-situational consistency coefficients
are obtained (Allport, 1937). Therefore, if the correlations among be-
havior observations are not lower for subjects with a low self-rated
trait applicability, this may indicate: (1) that subjects are not able to
provide valid judgments of whether their equivalence classes are ap-
propriately described by the trait under study; (2) that they regard the
trait as undescriptive of their behavior in general, although it is de-
scriptive of the particular behavior being investigated by the experi-
menter; or (3) that they regard the trait as irrelevant in terms of their
everyday life situations, despite the fact that it is descriptive of their
246 PETER BORKENAU
behavior in the situations being investigated. Thus there are numerous

reasons why correlations among behavior observations may be difficult
to predict from ratings of trait applicability. It is therefore not surpris-
ing that two of the three published studies using behavior observations
(Chaplin & Goldberg, 1984; Mischel & Peake, 1982) failed to replicate
Bern and Allen's findings. The one study that produced positive results
(Zanna et al., 1980) used attitudes instead of traits.
WAS THE SEARCH FOR MODERATORS SUCCESSFUL?
A meta-analysis of the studies that investigated trait variability,

trait applicability, trait importance, or trait relevance as moderators of
self-peer agreement has recently been published by Zuckerman et al.
(1989). The meta-analysis shows a reliable moderator effect in the
expected direction. This reflects the fact that there are studies with
higher rater agreement for cross-situationally stable subjects (Bern &
Allen, 1974; Kenrick and Stringfield, 1980; Mischel & Peake, 1982) or
subjects for whom a trait is applicable (Borkenau & Amelang, 1985),
important (Cheek, 1982), or relevant (Zuckerman et al., 1988, 1989), and
studies that do not produce this finding (Chaplin & Goldberg, 1984;
Paunonen & Jackson, 1985). By contrast, there are no published studies
that show reliably higher correlations for inconsistent subjects. More-
over, self-rated applicability of a trait to oneself, self-rated trait import-
ance, and self-rated trait relevance are equally efficient moderators as
the self-rated cross-situational variability of behavior. Thus, although
trait appropriateness is preferable on theoretical grounds (Amelang &
Borkenau, 1986; Zuckerman et al., 1989), there is no evidence that sug-
gests the superiority of this variable as a moderator. How can this
finding be explained?
One possibility is that subjects do not distinguish between the
variability of a trait and its applicability to oneself, its importance, and
its relevance. Correlations among self-ratings of variability, applica-
bility, importance, and relevance rarely surpass .20 (Burke et al.,
1984; Paunonen, 1988; Zuckerman et al., 1988, 1989); this finding led
Zuckerman and co-workers (1989, p. 283) to conclude that trait rele-
vance and trait consistency are relatively independent of one another.
But both are so unreliable that a significant correlation between them
is not to be expected. Borkenau (1981) checked the stability of self-
rated variability that was measured at six-month intervals and found
that these stability coefficients did not differ from zero. In contrast, the
average six-month stability of the trait ratings was .30. Borkenau
(1981) and Chaplin and Goldberg (1984) found low or zero correlations
between self-rated variability for a trait and the ipsatized variance

index for the same trait. Thus low correlations among measures of trait
variability and related concepts are difficult to interpret. The low reli-
ability of measures of trait variability raises important questions about
the wisdom of continuing to study it.
Zuckerman et al. (1988, 1989) also compared rating measures and
ranking measures of trait relevance, whereby the latter required a
ranking of several traits for their variability or relevance. Rating mea-
sures confound trait x person interactions with systematic individual
differences in trait relevance. They do not only identify the traits that
are of particular importance for each person, but also those persons
who generally believe that traits are useful to describe their personal-
ity. In contrast, ranking measures do not reflect systematic individual
differences in trait relevance, as the sum of all ranks is the same for all
subjects. Zuckerman et al. (1989) found that ranking measures of mod-
erators produced greater effects than rating measures. Moreover,
Borkenau and Amelang (1985) found moderator effects when using an
adjective generation technique to identify trait-relevant subjects. Thus,
if the considerable technical problems are overcome, the moderator
variable approach may be useful to somewhat increase the prediction of
ratings for some of the people.
CONCLUSIONS
The Bern and Allen approach should be evaluated in two respects:

(1) whether it really helped to predict some of the people more of the
time and (2) whether it revitalized Allport's ideas. As far as prediction
is concerned, it has not been clarified in more than a decade's time
whether the phenomenon first reported by Bern and Allen is reliable or
not. Some authors (Chaplin & Goldberg, 1984; Paunonen & Jackson,
1985) found that the phenomenon is not replicable, whereas others,
including myself, found confirming evidence. Even if there is such a
phenomenon, however, it is highly elusive.
And what about the revitalization of Allport's ideas? Unfortu-
nately, Gordon Allport died seven years before Bern and Allen (1974)
published their study. It would have been interesting to hear his re-
actions. I wonder whether he would have applauded this approach.
Allport argued against the routine use of statistical methods, and he
emphasized that personality is an organized whole. He tried to under-
stand the individuality of the single person. By contrast, Bern and Allen
suggested an approach that led to exactly those mass investigations
248 PETER BORKENAU
that Allport considered unlikely to prove fruitful in understanding per-

sonality. I conjecture that Gordon Allport, if still alive, would be more
likely to share Lawrence Pervin's (1985) view on the state of personal-
ity research; that is, that one learns more about people from personal
contact than from reading the personality literature. Thus, although
Bern and Allen's work is one of the major offshoots during the last
decades of Allport's theorizing, I think Allport himself would have re-
garded it as a bastard child.
REFERENCES
Allport, G. W. (1937). Personality: A psyclwlogical interpretation. New York: Holt.
Allport, G. W. (1961). Pattern and grawth in personality. New York: Holt.
Allport, G. W. (1966). Traits revisited. American Psyclwlogist, 21, 1-10.
Allport, G. W., & Odbert, H.S. (1936). Trait-names: A psycholexical study. Psyclwlogical
Monographs, 47 (1, No. 211).
Amelang, M., & Borkenau, P. (1986). The trait concept: Current theoretical considera-
tions, empirical facts, and implications for personality inventory construction. In A.
Angleitner & J. S. Wiggins (Eds.), Personality assessment via questionnaires: Cur-
rent issues in theory and measurement (pp. 7-34). Berlin: Springer.
Bern, D. J. (1983). Further deja vu in the search for cross-situational consistency: A reply
to Mischel and Peake. Psychological Review, 90, 390-393.
Bern, D. J., & Allen, A. (1974). On predicting some of the people some of the time:
The search for cross-situational consistencies in behavior. Psychological Review, 81,
506-520.
Borkenau, P. (1981). Intraindividuelle Variabilitat und differentielle Vorhersagbarkeit
[lntraindividual variability and differential prediction]. Unpublished doctoral disser-
tation, University of Heidelberg, Germany.
Borkenau, P., & Amelang, M. (1985). Individuelle Angemessenheit von Eigen-
schaftskonstrukten als Moderatorvariable fiir die Ubereinstimmung zwischen
Selbst- und Bekanntenratings [Individual appropriateness of traits as a moderator of
self-peer agreement]. Diagnostica, 31, 105-118.
Bowers, K. S. (1973). Situationism in psychology: An analysis and a critique. Psyclwlogi-
cal Review, 80, 307-336.
Burke, P. A., Kraut, R., & Dworkin, R. H. (1984). Traits, consistency, and self-schemata:
What do our methods measure? Journal of Personality and Social Psychology, 47,
568-579.
Chaplin, W. F., & Goldberg L. R. (1984). A failure to replicate the Bern and Allen study
on individual differences in cross-situational consistency. Journal of Personality and
Cheek, J. (1982). Aggregation, moderator variables, and the validity of personality tests.
Epstein, S. (1979). The stability of behavior: I. On predicting most of the people much of
the time. Journal of Personality and Social Psychology, 37, 1097-1126.
Epstein, S., & O'Brien, E. J. (1985). The person-situation debate in historical and current
perspective. Psychological Bulletin, 98, 513-537.
Eysenck, M. W., & Eysenck, H. J. (1980). Mischel and the concept of personality. British
Jaurnal of Psychology, 71, 191-204.
Hanushek, E. A., & Jackson, J. E. (1977). Statistical methods for social scientists. New
Hartshorne, H., & May, M.A. (1928). Studies in the nature of character: Vol. 1. Studies
in deceit. New York: Macmillan.
Hartshorne, H., May, M. A., & Maller, J. B. (1929). Studies in the nature of character:
Vol. 2. Studies in service and self-control. New York: Macmillan.
Hartshorne, H., May, M. A., & Shuttleworth, J. B. (1930). Studies in the nature of char-
acter: Vol. 3. Studies in the organization of character. New York: Macmillan.
Jaccard, J. J. (1974). Predicting social behavior from personality traits. Journal of Re-
search in Personality, 7, 358--367.
Kenrick, D. T., & Stringfield, D. 0. (1980). Personality traits and the eye of the beholder:
Crossing some traditional philosophical boundaries in the search for consistency in
all of the people. Psychological Review, 87, 88-104.
Magnusson, D. (1976). The person and the situation in an interactional model of behavior.
Scandinavian Jaurnal of Psychology, 17, 253-271.
Mischel, W., & Peake, P. K. (1982). Beyond deja vu in the search for cross-situational
consistency. Psychological Review, 89, 730-755.
Paunonen, S. V. (1988). Trait relevance and the differential predictability of behavior.
Jaurnal of Personality, 56, 599--619.
Paunonen, S. V., & Jackson, D. N. (1985). Idiographic measurement strategies for person-
ality and prediction: Some unredeemed promissory notes. Psychological Review, 92,
486-511.
Pervin, L. A. (1985). Personality: Current controversies, issues, and directions. Annual
Rushton, J. P., Jackson, D. N., & Paunonen, S. V. (1981). Personality: Nomothetic or
idiographic? A response to Kenrick and Stringfield. Psychological Review, 88, 582-
589.
Saunders, D. R. (1956). Moderator variables in prediction. Educational and PsY,chologi-
cal Measurement, 16, 209-222.
Stones, M. J., & Burt, G. (1978). Quasi-statistical inference in rating behavior. Journal of
Research in Personality, 12, 381-389.
Tellegen, A., Kamp, J., & Watson, D. (1982). Recognizing individual differences in predic-
tive structure. Psychological Review, 89, 95-105.
Wachtel, P. L. (1973). Psychodynamics, behavior therapy, and the implacable experi-
menter: An inquiry into the consistency of personality. Journal of Abnormal Psy-
chology 82, 324-334.
Zanna, M. P., Olson, J. M., & Fazio, R. H. (1980). Attitude-behavior consistency-An
individual difference perspective. Journal of Personality and Social Psychology, 38,
432-440.
Zuckerman, M., Koestner, R., DeBoy, T., Garcia, T., Maresca, B. C., & Satoris, J. M.
(1988). To predict some of the people some of the time: A reexamination of the
moderator variable approach in personality theory. Journal of Personality and So-
cial Psychology, 54, 1006-1019.
Zuckerman, M., Bernieri, F., Koestner, R., & Rosenthal, R. (1989). To predict some of the
people some of the time: In search of moderators. Journal of Personality and Social
CHAPTER EIGHTEEN
The Scientific Credibility of

Commonsense Psychology
GARTH J. 0. FLETCHER
By the early 1980s, the idea that ordinary human social cognition is
rational or scientific in character was under serious attack in personal-
ity and social psychology. Personality psychology was in the throes of
the debate concerning whether behavior was consistent across situa-
tions and, although the jury was still out on the issue, powerful voices
and persuasive data suggested that the concept of personality traits
was a quaint shibboleth of commonsense psychology. As Nisbett and
Ross (1980) put it, "the personality theorists' (and the layperson's) con-
viction that there are strong cross-situational consistencies in behavior
may be seen as merely another instance of theory-driven covariation
assessments operating in the face of contrary evidence" (p. 112).
In social cognition circles, a flood of research on errors and biases
in social judgment apparently demonstrated that laypeople were sub-
ject to an extraordinary range of invidious social judgment biases-
laypeople were purported to underestimate the causal role of situa
tional determinants of behavior and overestimate the role of personal
determinants (the so-called fundamental attribution error), to be poor
statisticians, to be unduly influenced by prior theories while underuti-
lizing data, and much more (for reviews, see Markus & Zajonc, 1985;
Nisbett & Ross, 1980; Ross, 1977). As Fiske and Taylor (1984) con-
GARTH J. 0. FLETCHER • Department of Psychology, University of Canterbury,

Christchurch 1, New Zealand.
New York, 1993.
251
252 GARTH J. 0. FLETCHER
eluded in their textbook, Social Cognition: "Instead of a naive scientist

entering the environment in search of the truth, we find the rather
unflattering picture of a charlatan trying to make the data come out
in a manner most advantageous to his or her already held theories"
(p. 88).
This issue and related arguments impact on personality psychology
and theory in various ways. First, the degree of rationality inherent in
everyday social cognition is an important question in relation to general
personality theories. Second, much current personality research and
theory is derived from the personality theory embedded in common-
sense psychology (see Chapter 19, this volume). Consider, for example,
the development of the big five personality structure as articulated by
McCrae & Costa (1985): extroversion, neuroticism, conscientiousness,
agreeableness, and openness. The factorial analyses that produce these
factors are based on ratings of personal dispositions drawn from the
English language. Hence, the resultant personality structure is essen-
tially an explication of the general personality theory inherent in our
language and in laypeople's personality schemata. To put the matter
bluntly, if commonsense psychology goes down the gurgler, then so do
large portions of contemporary personality psychology.
Assessments of the rationality or scientific credibility of common-
sense social cognition depend on the normative model it is compared
with. The first section of this chapter will draw on some contemporary
philosophy of science and sketch some central elements in scientific
cognition. I shall then use this analysis as a framework to analyze and
evaluate recent social psychological and personality research in terms
of the rationality of ordinary social cognition and in relation to the
scientific credibility of our commonsense psychological theories that un-
dergird everyday social judgment.
A REALIST THEORY OF PSYCHOLOGICAL SCIENCE
The idea of a realist approach to psychology that transcends a

narrow empiricism is not new. For example, Allport (1966) recommended
that we adopt what he termed "heuristic realism": He writes
Galloping empiricism ... dashes forth like a headless horseman. It
has no rational objective; uses no rational method other than math-
ematical; reaches no rational conclusion. It lets the discordant data
sing for themselves. By contrast, heuristic realism says ... the area
we carve out should be rationally conceived, tested by rational
methods, and the findings should be rationally interpreted. (p. 40).
SCIENTIFIC CREDIBILITY OF COMMONSENSE PSYCHOLOGY 253
Like Allport, I believe that a variant of scientific realism offers the

most appropriate and viable scientific model for psychology. Scientific
realism is a wide-ranging theory of science; thus, I will consider only a
few central components of this approach that are most relevant to the
later discussion*: truth as a scientific aim, the relation between theory
and data, and a generative concept of causality.
TRUTH AS A SCIENTIFIC AIM
According to scientific realism, scientific theories routinely postu-

late unobserved theoretical entities to explain phenomena; for example,
cognitive mechanisms and personality constructs in psychology or sub-
atomic particles in physics. The nub of a realist approach is that these
deep-structural theories are intended to adequately refer to and rep-
resent a world, some of which is independent of human cognition.
Hence, one of the pivotal aims of science is to construct true theories
about that world. However, the drive toward "truth" is conceptualized
here as a guiding ideal or idealized horizon (McMullin, 1983), which can
be conceived of as a highly valued, though unobtainable, goal of scien-
tific enquiry.
THE RELATION BETWEEN THEORY AND DATA
The hypothetico-deductive model is probably the most commonly

accepted normative model in personality and social psychology. Accord-
ing to this model, observational claims are deducted from theories, then
tested. A variant of this model proposed by Popper (1959), which still
has a considerable following in psychology, posits that such tests can
falsify but not verify theories. Yet the hypothetico-deductive model and
Popper's falsificationist version have been subject to compelling attacks
from philosophers over the last three decades and retains few adher-
ents in contemporary philosophy of science.
Probably the major reason for the demise of hypothetico-deductivism
in philosophy of science is simply because it does not fit the way sci-
ence appears to work. For example, this model implies that scientists
should drop theories like hotcakes when faced with disconfirming data.
In fact, the numerous analyses carried out by philosophers of science
examining historically important episodes of scientific change in the
physical sciences show that this is not what happens to all (e.g., see
*For philosophical accounts of scientific realism see Bhaskar (1978), Hooker (1987), and
McMullin (1984); for more psychologically oriented treatments, see Howard (1985),
Manicas and Secord (1983), and Greenwood (1989).
Feyerabend, 1975; Kuhn, 1970). The same point holds good for social
psychology and personality research, which is full of controversies in
which apparently disconfirming data have failed to convince the pro-
ponents of the original theories that their hypotheses or theories were
false; examples would include the debate between self-perception theory
and cognitive dissonance theory, the role of consensus information in
social judgment (see Ross & Fletcher, 1985), and whether trait-related
behavior is consistent across situations (see Chapter 17, this volume).
However, what to the orthodox empiricist appears to be a blatant
disregard for evidence can instead be construed as perfectly legitimate
strategies for dealing with recalcitrant data, including positing method-
ological deficiencies in the relevant research, reinterpreting the data so
as to render them consistent with one's preferred theory, or simply
modifying the original theory.
The thrust of my argument can be summed up in the aphorism
that theory is underdetermined by data. This point implies that there is
no algorithm that connects data to theory selection. Instead, theory
evaluation within a scientific community is always a judgment call that
proceeds in a relatively disheveled fashion involving a complex and
subtle interplay between conceptual and empirical concerns.
The loose relation between data and theory perhaps explains why
dissent and argument are permanent features of psychology, and
indeed all science. However, scientific disciplines, again including psy-
chology, can hardly be described as irredeemably inchoate: contro-
versies are resolved, consensus is often attained, and progress is
made. What explains the existence of this consensual aspect of psychol-
ogy? My short answer is in terms of the rational structure of science,
which provides an interlocking set of normative values, rules, and aims
that regulate theory appraisal and development. These values include
some important epistemic criteria used for theory evaluation, apart
from the consistency between theory and data or predictive accuracy
(Fletcher & Haig, 1990; Howard, 1985).
Consider, for example, the following epistemic criteria proposed by
Fletcher and Haig (1990): Explanatory depth involves the postulation
of one or more underlying causal mechanisms to account for the surface
phenomena. A theory that manages to integrate and explain hitherto
disparate items of knowledge possesses the valuable property of unify-
ing power. A theory whose content is not marred by logical inconsisten-
cies has internal coherence while a theory that is consistent with other
entrenched or accepted theoretical knowledge has the virtue of exter-
nal coherence. Fertility refers to the ability of a theory to stimulate
further fruitful lines of research or generate novel and powerful exten-
sions to our knowledge base. The notion of simplicity has proved diffi-
cult to formulate; nevertheless, scientists are frequently attracted by
theoretical elegance rather than cumbersome complexity. Finally, the
successful application of a mature theory counts as an important epi-
stemic value in a science concerned to give us useful knowledge.
The first criterion (that of explanatory depth) is especially relevant
to scientific realism and embraces a particular theory or concept of
causality known as a generative concept. To conclude this section, I
turn to a brief examination of this concept.
A GENERATIVE CONCEPT OF CAUSALITY

A generative concept of causality proposes that causal relations
between events, or between events and dispositional properties, con-
tain some sort of necessary connection or bond that is embedded in the
nature of things. A generative account can usefully be contrasted with
an approach, usually identified with David Hume, that construes cau-
sality purely in terms of regularity. A regularity theory of causality
interprets a causal relation as one in which the events are regularly
conjoined in space and time, with the cause preceding the effect. The
major problem with this approach is that it becomes difficult to concep-
tually distinguish between bona fide cause-effect relations and coinci-
dental noncausal relations that may be perfectly regular (e.g., night
following day). A generative theory solves this problem.
A generative causal account has at least two other important fea-
tures from my perspective. First, this approach is consistent with
realism's demand for producing deep-level explanatory theories that
represent hypothetical generative mechanisms. Second, it happily ac-
commodates current cognitive, social psychological, and personality ap-
proaches that postulate cognitive-structural models as being causally
implicated in the generation of cognition, affect, or behavior.
It is instructive to evaluate the rationality of the layperson's thinking
and also the scientific credibility of the layperson's social psychological-
personality theories using our thumbnail sketch of a realist account of
scientific cognition as a normative model. I will attempt these tasks next.
EVALUATING THE LAYPERSON'S SOCIAL COGNITION:

SCIENTIST OR SIMPLETON?
As noted previously, science, according to a realist account, is cen-

trally concerned with the truth value of its theories. But can the same
be said for laypeople in their everyday social cognition? As indicated in

the introduction, powerful voices within social psychology have an-
swered no! One challenge to the notion that laypeople are seriously
interested in truth emanates from the view that humans are more in-
terested in retaining a flattering and positive self-concept than in truth.
Indeed, there is a large body of evidence that suggests normal human
cognition is characterized by unrealistically positive views of the self,
exaggerated perceptions of personal control, and unrealistic optimism
(for reviews, see Greenwald, 1980; Taylor & Brown, 1988).
However, this self-serving picture of human cognition is not neces-
sarily antithetical to the layperson's employment of the more dispas-
sionate scientific values of accuracy and logic in the pursuit of truth.
The work of Trope (1979, 1980) and others (Strube et al., 1986) has
shown that subjects will, at times, prefer tasks that provide good diag-
nostic information concerning their own level of ability, even when such
tasks have the potential of providing ego-threatening feedback.
In short, both cognitive-rational and motivational models of social
cognition are correct; people are both rational and rationalizers-at times
concerned with explanation, at times with justification, at other times
both, and at still other times with neither. Viewed in this way, the
proper question becomes to what extent and under what conditions are
lay social judgments determined by self-serving motivational factors.
A second major claim to emanate from mainstream social cogni-
tion, and one that apparently lays waste the scientific credibility of
ordinary cognition, is that laypeople are simply unwilling to modify or
abandon beliefs, hypotheses, or theories in the face of disconfirmatory
evidence (Fiske & Taylor, 1984; Nisbett & Ross, 1980). However, the
above postulate is derived from research and theorizing that largely
embodies adherence to the sort of clear-cut algorithmic connection be-
tween theory and data that I previously challenged. To recap, I argued
that there was no such clear-cut relation between theory and data, and
that it was not uncommon for scientists to retain theories in the face of
disconfirmatory evidence. Science is inherently conservative in terms
of theory change and for obvious reasons. If we revamped or aban-
doned our theories strictly and immediately according to the vicissi-
tudes of our data, we would indulge in theory change with unsettling
regularity. Moreover, the standard situation in science is to have evi-
dence that both confirms and disconfirms our preferred theory. The
layperson's conservatism, on this view, should not necessarily be la-
beled as unscientific.
Once again, the correct question to ask is to what extent the
layperson ignores data that run against extant theories or hypotheses.
In fact, the evidence concerning this issue shows that the layperson's
hypotheses or theories are usually influenced by data-the claim that
subjects' performance is less than optimal is based on the usual finding
that subjects underutilize the data, according to some normative model
of inference such as Bayes's theorem or to the standard Fisherian anal-
ysis using significance tests (for reviews substantiating this point, see
Einhorn & Hogarth, 1981; Higgins & Bargh, 1987; Ross & Fletcher,
1985). One can view the glass as half full or half empty and be im-
pressed or dismayed at the layperson's attention to predictive accuracy.
A second pivotal point is that if we dispense with a standard
hypothetico-deductive framework and accept the idea that there are
several epistemic criteria available for theory evaluation, then this can
alter the standard interpretation of the layperson's performance in re-
search tasks. Why? Because psychologists typically do not consider all
the epistemic values that their subjects may quite reasonably be con-
sidering. Take, for example, the study by Lepper, Ross, and Lau (1986),
concerned with belief perseverance, in which two groups of high school
subjects performed very well or very poorly at solving novel problems
under the guidance of either thoroughly effective or ineffective instruc-
tional films. In spite of being made aware at a later time of the impact
of the instructional films, subjects continued to draw unwarranted in-
ferences (to use the author's words) concerning their own levels of
ability that were based on their original performances.
But consider the plausible possibility that subjects would have
tried to explain their initial performance and sought to integrate this
new information with their extant knowledge. If we consult our prior
list of epistemic values, this implies that subjects were utilizing values
of explanatory depth, unifying power, and internal and external co-
herence. In turn, this bolstering of the initial ability attributions may
have rendered these judgments less susceptible to sudden revision
when information was received that, from the experimenter's perspec-
tive, apparently decisively invalidated the subjects' original ability at-
tributions. It is all too easy to place subjects in research contexts that
produce inaccurate subject inferences, but this does not entail that such
judgments are necessarily irrational or unwarranted.
Finally, let me briefly consider the overall research literature on
errors and biases in social judgment. Several authors have drawn sim-
ilar conclusions based on literature reviews of this research (Fletcher
& Haig, 1990; Klayman & Ha, 1987); namely, under unfavorable condi-
tions that promote a casual, automatic, or data-driven style of informa-
tion processing, laypeople will typically rely on fallback heuristics or
easily used rules of thumb. On one hand, these heuristics are often
reliable, adaptive, and effective devices; on the other, they produce

characteristic biases or errors under certain conditions (Nisbett &
Ross, 1980). However, under more friendly processing conditions, that
provide useful cues or promote more in-depth information processing,
these default heuristics tend to be corrected or discarded, hence reduc-
ing resultant biases and errors. These favorable processing conditions
include adequate processing time (Fletcher et al., 1990a), a light mem-
ory load (Gilbert et al., 1988), having the problem framed in a familiar
context (Griggs & Cox, 1983), having cues that help in strategy choice
(Hinz et al., 1988), possessing adequate knowledge or sufficient schema
complexity to process the information (Fletcher et al., 1986), and being
sufficiently motivated to carry out an in-depth analysis (Tetlock, 1988).
Take as an example the celebrated bias known variously as the
correspondence bias or the fundamental attribution error: the tendency
to underestimate the causal role of situational determinants of behavior
and overestimate the causal role of the internal determinants (Ross,
1977). Recent studies have suggested that under conditions that en-
courage an in-depth and careful processing of the stimulus materials,
compared to a superficial and casual analysis, correspondence bias will
decrease (Fletcher et al., 1990; Gilbert et al., 1988; Tetlock, 1985).
Individual differences form another, though not commonly re-
searched, class of conditions that may influence social inference biases.
Fletcher and his colleagues have examined the role that the complexity
of attributional schemata has in relation to the level of expertise exhib-
ited in social judgment, using a recently developed scale termed the
Attributional Complexity Scale (Fletcher et al., 1986). This 28-item
scale includes seven attributional subconstructs that vary along a simple-
complex dimension, including the level of motivation to explain human
behavior, the tendency to indulge in metacognitive attributional think-
ing, the tendency to infer complex causal explanations that are both
internal and external, the tendency to infer causes from the distant
past, and so on. Initial results have confirmed that this scale possesses
good internal reliability, convergent and discriminant validity, concur-
rent validity, and predictive validity (Brookings & Brown, 1988;
Fletcher et al., 1986; Flett et al., 1989).
Using the Attributional Complexity Scale, Fletcher and his col-
leagues, and others, have found that subjects who possess complex
attributional schemata produce more accurate trait and attitude judg-
ments than do those with simple schemata, but that such an advantage
appears to be manifested most strongly under conditions that encour-
age in-depth information processing that is goal driven (Devine, 1989;
Fletcher et al., 1988, 1990, 1992). In a reaction-time study Fletcher et
al. (1992) also found that attributionally complex subjects displayed a

clear-cut tendency to control the amount of processing time according
to the difficulty level of the causal problem; in contrast, attributionally
simple subjects did not control their processing time at all according to
the difficulty level of the causal problem. Truly, some laypeople appear
to be better naive (social) scientists than others.
CONCLUSION
To conclude this section, I think it is apparent that to view the

layperson as a lazy thinker, uninterested in truth, and replete with
biases and shortcomings in his or her social thinking is at best an exag-
gerated half-truth. However, the extent to which one judges the ration-
ality of social judgment, or even the propriety of labeling a mode of
thought as a "bias" or an "error," will be determined, in part, by the
normative framework adopted. If the normative framework of science
advanced here is accepted, then the layperson's social cognition, un-
der certain conditions, looks remarkably similar to that of our proto-
typical scientist.
FOLK PERSONALITY-SOCIAL PSYCHOLOGY THEORIES:

TOUCHSTONE OR CROCK?
Introductions to social psychology or personality texts often seek

to distance these disciplines from common sense. A commonly adopted
strategy to demonstrate the relative paucity of common sense is to
contrast commonsense maxims that appear contradictory, such as
"birds of a feather flock together" and "opposites attract," and proceed
to argue that research is needed to go beyond such crude and unhelpful
formulations. In contrast to the standard textbook litany, I will argue
that substantial elements of personality and social psychological theo-
rizing are clearly derived from commonsense psychological theory.
Of course, most personality and social psychologists happily admit
to their interest in commonsense psychological theory, insofar as that
theory is causally implicated in the generation of behavior. The sche-
mata and cognitive processes that people use in making judgments and
interacting with others are part and parcel of "naive" commonsense
theory; hence, common sense is, in part, the subject matter of personal-
ity and social psychology. Moreover, personality and social psychologi-
cal theories also routinely borrow concepts and principles from common
sense in building scientific theories; such theories abound with com-
260 GARTHJ.O.FLETCHER
monsense concepts such as attitude, intention, trait, need, drive, belief,

emotion, goal, and so forth. Of course, these concepts are typically re-
fined and more precisely defined in their new theoretical homes than
according to their colloquial usages, but the concepts are usually en-
sconced in theoretical frameworks that are quintessential "common
sense"-what Dennett (1987) calls the "intentional stance." The "inten-
tional stance" involves treating humans as rational systems that have
certain goals; we can predict and explain behavior by assuming that
humans will attempt to achieve those goals, in virtue of various mental
states such as beliefs, aspirations, knowledge, attitudes, and so forth.
Empirical examinations that have been made of the similarity be-
tween psychological and commonsense theories support my contention
concerning the close links between professional and commonsense psy-
chology (for a review, see Furnham, 1989). For example, substantial
overlap has been found between lay conceptions and psychologists' the-
ories concerning the introversion-extroversion personality dimension
(Semin & Krahe, 1987), intelligence (Sternberg et al., 1981) delinquency
(Furnham & Henderson, 1983), and depression (Rippere, 1977).
Although there may be debate concerning the nature and degree of
convergence between the theories of the layperson and those of the
personality and social psychologist, there is relatively little dispute
over the claims that there is considerable overlap between the two
spheres. Fierce debate, however, has waxed over the questions of
whether commonsense psychological theory is true and whether it is
doomed to be replaced by theories with sterner scientific credentials.
The answers to these latter questions are not necessarily related to the
scientific status of everyday social cognition. It is possible that ordi-
nary folk rely on muddled or fallacious theories because of their limited
access to the powerful experimental and research techniques that are
available to the professional research psychologist, rather than because
they are other than immaculately rational and suitably curious about
human behavior. Nonetheless, conclusions concerning the silliness of
commonsense psychological theory are relevant to an overall assess-
ment of everyday cognition.
Over the last decade, there has been considerable debate among
philosophers of science concerning the feasibility or status of what is
referred to as "folk psychology'' (see Fletcher & Haig, 1990). The argu-
ments have resolved around the status of ''beliefs" or other cognitive
attributions in commonsense psychology. For example, one of the most
vigorous attacks on folk psychology theory has been developed by
Churchland (1984, 1985) who claims that folk psychology is destined for
extinction on the grounds that our commonsense psychological frame-
work is a false and radically misleading conception of the causes of

human behavior and the nature of cognitive activity. Common sense, it
is argued, will be replaced in due course by a mature neuroscience.
Although Churchland produces little empirical evidence for his nega-
tive assessment of folk psychology, he does present a variety of argu-
ments that I have not the space here to evaluate. However, his central
contention is
... that neuroscience is unlikely to find "sentences in the head" or
anything else that answers to the structure of individual beliefs and
desires. On the strength of this shared assumption, I am willing to
infer that folk psychology is false, and that its ontology is chimeri-
cal. Beliefs and desires are of a piece with phlogiston, caloric, and
the alchemal essences. (Churchland, 1988, p. 508)
One major problem with Churchland's position is that folk psychol-
ogy is narrowly represented as a cognitive theory. In fact, it seems
obvious that common sense also contains a sophisticated and subtle
theory of personality and social behavior. This personality theory is
typically conceived of by psychologists as consisting of schemata that
embrace a large set of personality dispositions that are organized to
represent a general personality structure. Factor-analytic studies sug-
gest that such personality schemata can be described in terms of gen-
eral dimensions such as social desirability and intellectual desirability
(e.g., Rosenberg et al., 1968). Certain traits are thus seen as naturally
co-occurring; for example, skillful and intelligent, clumsy and naive,
humorless and unsociable, and helpful and sincere.
Now a huge number of personality dispositions appear to refer to
the behavioral as well as the mental spheres. For example, attributions
of warmth, introversion, or aggressiveness imply certain loosely de-
fined classes of behaviors, as well as some sort of mental component
such as emotional events (e.g., introverted people feel more uncomfort-
able in company). In addition, many dispositions contain an interper-
sonal component; for example, dispositions such as warm, sensitive,
abrupt, and so forth assume an interactional context. In short, folk
psychology is closer to a personality and social psychological theory
rather than simply a theory of mind. Moreover, a considerable amount
of research attention has been allocated by personality and social psy-
chologists to the question of whether this theory is accurate or reason-
able, and the results suggest that commonsense theory may have
considerably more going for it than suggested by Churchland's damn-
ing indictment.
Several influential reviews of the literature in the 1960s (e.g.,
Mischel, 1968) suggested that there was relatively low consistency of

behavior across situations or between self-rated dispositions and indi-
vidual behaviors (usually resulting in correlations lower than .3). How-
ever, it has become clear from the recent flurry of research and critical
argument that when behavior is aggregated, rather than treated as
single instances, the resultant consistency correlations are much higher
(up to .7) (see Kenrick & Funder, 1988, for a recent review). This level
of consistency is all that is required as a basis for folk psychology,
given that many commonsense, dispositional concepts are loosely de-
fined in terms of rough families of behaviors (e.g., sociable, assertive,
unfriendly).
Another line of attack on the validity of folk personality theory has
argued that the pattern of correlations produced when subjects rate
other people on a set of personality scales is a product of the semantic
relations between the traits rather than the empirical relations be-
tween the traits-a thesis known as the semantic distortion hypothesis
(see, for example, D'Andrade, 1974; Shweder, 1975). The evidence pro-
duced for this claim includes demonstrations that subjects' ratings of
the semantic meanings of trait terms produces a factor structure sim-
ilar to that derived from ratings of real people.
However, against the semantic distortion hypothesis, two recent
ingenious studies have shown that when semantic effects are con-
trolled, the pattern of correlations among the trait ratings is affected
very little (DeSoto et al., 1985; Weiss & Mendelsohn, 1986). Moreover,
several studies have shown that, under certain conditions, there is
good agreement between self-personality judgments and judgments of
peer raters on particular traits and also reliable interjudge agree-
ment concerning the individual being rated (for a recent review, see
Funder, 1987). As Kenrick and Funder (1988) have pointed out,
these results cannot be explained in terms of the semantic distortion
hypothesis.
DISPOSITIONS AS CAUSAL EXPLANATIONS
It is clear from casual observation and analysis of the everyday

attribution of dispositions, and also recent research (Asch & Zukier,
1984; Hastie, 1984), that our commonsense model of dispositions is, in
part, a deep-structural model in that many dispositions are perceived
as causes for other traits or specific behaviors. This characterization of
commonsense personality theory is consistent with the proposition that
the concept of causality implicit in everyday cognition is generally
closer to a generative rather than a regularity theory of causality. The
question of what status to accord dispositional or trait concepts in aca-

demic psychology is a considerably more contested issue that I shall
now discuss.
The major competing views on dispositions are conveniently cap-
tured by Zuroff (1986) in his recent discussion of whether Allport was
a trait theorist:
There are three basic positions on the reality of traits. As is well
known, Allport held that traits are real, causal entities that corre-
spond to as yet unknown neurophysiological structures. An oppos-
ing view is that traits are purely descriptive; they summarize a
person's past behavior, but they have no real existence and are
certainly not causal entities (Buss & Craik, 1983; Wiggins, 1974). A
third possibility is that traits are dispositional concepts (Ryle,
1949)..... the dispositional view of traits ... is that they describe a
tendency to perform a certain class of acts when the individual is
placed in a certain class of situations. Dispositions are distinct from
summaries because they do not imply anything about the actual
occurrence of behavior; in the absence of the eliciting stimuli, even
a strong disposition will not be manifest in the stream of behavior.
Although there are some philosophers who view dispositions as
causal concepts (Armstrong, 1969; Hirschberg, 1978), they are gen-
erally not considered to provide causal explanations, and, of course,
they are not entities. (p. 996)
One major difficulty, both with personality theory and in the de-
bate surrounding this issue, is the disagreement concerning what "dis-
positions" and "traits" are taken to include. Some psychologists treat
these two concepts (traits and dispositions) as equivalent, some do not.
Some theorists think all internal dispositions, such as abilities, needs,
beliefs, and attitudes, should be excluded from the study of personality;
others include a broad range of personal dispositions including those
just listed. Fletcher (1984) and Newman and Uleman (1989) have sug-
gested that some classes of disposition qualify for causal status while
others do not. For example, many dispositions seem to refer to pat-
terns of observable behavior and perhaps best fit the descriptive sum-
mary approach: for example, talkative, punctual, and untidy. Other
dispositions appear to be attributions to the "mind" and to clearly rep-
resent potential causes for behavior or mental events: for example,
beliefs, abilities, and states of knowledge. Still another enormous class
of dispositions appear to fall into both camps containing both a causal,
usually mental, component and a set of behaviors that are regularly
evinced: for example, tolerant, warm, independent, stubborn, confident,
and insecure.
The attempt to build a personality theory operating purely with a

summary-descriptive definition of dispositions, such as Buss and
Craik's (1983) act frequency approach, cannot handle cases where dis-
positions are rarely or never manifested. This approach also ignores or
postpones the all-important task of building explanatory theories (see
Block's [1989] devastating critique of the act frequency approach). This
criticism also applies to the remaining approach, that Zuroff (1986)
terms the dispositional account of traits. In this view, ascribing a dispo-
sition is tantamount to a lawlike statement that states a person is liable
to behave in characteristic ways under certain conditions. The major
problem with this approach is again that it lacks explanatory power; it
merely describes empirical regularities and fails to deal with the under-
lying cognitive or personological constructs that can cause and hence
explain such regularities.
The causal-explanatory approach to at least a subset of disposi-
tions, is, of course, the option favored by realists. For the realist ac-
count, to ascribe a disposition is to ascribe a property that enables the
person to behave in a particular way under appropriate conditions.
Hence, dispositions are causes, and the enabling relation is a causal
relation. In consequence, realists, like Allport, assert what Zuroff and
others deny; namely, that dispositions can function as genuine causal
explanations.
One common objection to interpreting dispositions as causes is that
they are simply promissory notes for causal explanations. Zuroff (1986),
for example, notes that "An explanatory system that makes use of trait
concepts must embed the trait terms in a process theory that redeems
the promissory notes if it is to generate new findings and deeper un-
derstandings" (pp. 999-1000). We certainly accept the notion that dispo-
sitional explanations tend to operate as placeholders for later more
insightful and complex causal accounts. However, unlike promissory
notes, dispositional explanations in psychology cannot usually be ex-
changed for fuller explanations on demand. Patience may be rewarded,
however. The lesson to be learned from the development of good scien-
tific theories in other fields is that dispositional explanations often re-
quire prolonged cultivation for small yield. For example, 25 years
passed before the Sutton-Boveri hypothesis of chromosomal gene loca-
tion provided the first independent test of Mendel's theory of the gene.
Indeed, it is my view that recent work that has examined traits or
individual differences from a social cognitive perspective is currently
cashing in some of these promissory notes (e.g., Fletcher & Fincham,
1991; Higgins & Bargh, 1987; Showers & Cantor, 1985).
CAVEATS AND CONCLUSIONS
One theme running through this chapter has been that common
sense may be more subtle and sophisticated than is often appreciated
and that much of the research and theorizing (of both philosophers and
psychologists) may be vitiated by their often cavalier specification of
the content and structure of our folk psychology theory. Being anxious
to avoid falling into the same trap, I would stress there are some im-
portant and obvious differences between commonsense theory and the
corpus of "scientific" psychological theories.
Commonsense psychological theories, to a large extent, consist of
tacit knowledge, whereas "scientific" theories are laid out in compara-
tively explicit detail. In addition, commonsense psychological theories
seem to have a wider range of uses and aims than scientific theories
and consist of a more amorphous, flexible and sprawling set of concepts
and models than most psychological theories.
Finally, I have argued for the scientific credibility of commonsense
psychology and have suggested that under certain conditions laypeople
are considerably more rational in their social judgments than suggested
by mainstream social cognition. This does not mean, however, that per-
sonality and social psychologists should exploit commonsense psychol-
ogy in an unreflexive fashion or that folk psychology necessarily
represents the correct psychological theory (as, of course, could be said
of any theory in science). To treat common sense as a resource for
theory building is a fine strategy, but to sanctify it is quite another
matter-caveat emptor!
REFERENCES
Allport, G. W. (1966). Traits revisited. American Psychologist, 21, 1-10.

Annstrong, D. M. (1969). Dispositions are causes. Analysis, 30, 28-.'36.
Asch, S., & Zukier, H. (1984). Thinking about persons. Journal of Personality and Social
Psychology, 46, 1230-1240.
Bhaskar, R. (1978). A realist theury of science (2nd ed.). Sussex: Harvester Press.
Block, J. (1989). Critique of the act frequency approach to personality. Journal of Per-
Brookings, J. B., & Brown, C. E. (1988, April). Dimensionality of the attributional com-
plexity scale. Paper presented at the meeting of the Midwestern Psychological
Association.
Buss, D. M., & Craik, K. H. (1983). The act frequency approach to personality. Psycho-
Churchland, P. M. (1984). Matter and consciousness. Cambridge, MA: MIT Press.
Churchland, P. M. (1985). Reduction, qualia, and the direct introspection of brain states.
The Journal of Philosophy, 82, 8-28.
Churchland, P. M. (1988). The ontological status of intentional states: Nailing folk psy-
chology to its perch. Behavioral and Brain Sciences, 11, 507-508.
D'Andrade, R. G. (1974). Memory and the assessment of behavior. In H. M. Blalock (Ed.),
Measurement in the social sciences. Chicago: Aldine-Atherton.
Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press.
Desoto, C. B., Hamilton, M. M., & Taylor, R. B. (1985). Words, .people, and implicit
personality theory. Social Cognition, 3, 369--382.
Devine, P. (1989). Overattribution effect: The role of confidence and attributional com-
plexity. Social Psychology Quarterly, 52, 149-158.
Einhorn, H. J., & Hogarth, R. M. (1981). Behavioral decision theory: Processes of judg-
ment and choice. Annual Review of Psychology, 32, 53--88.
Feyerabend, P. K. (1975). Against method. London: New Left Books.
Fiske, S. T., & TaylorS. E. (1984). Social cognition. Reading, MA: Addison-Wesley.
Fletcher, G. J. 0. (1984). Psychology and common sense. American Psychologist, 39,
203-213.
Fletcher, G. J. 0., & Fincham, F. D. (1991). Attribution processes in close relationships.
In G. J. 0. Fletcher & F. D. Fincham (Eds.), Cognition in close relationships (pp. 7-
35). Hillsdale, NJ: Erlbaum.
Fletcher, G. J. 0., & Haig, B. (1992). The layperson as "naive scientist": An appropriate
model for social psychology? Unpublished manuscript, University of Canterbury,
Christchurch, New Zealand.
Fletcher, G. J. 0., Grigg, F., & Bull, V. (1988). The organization and accuracy of personal-
ity impressions: Neophytes versus experts in trait attribution. New Zealand Jour-
nal of Psychology, 17, 68-77.
Fletcher, G. J. 0., Reeder, G. D., & Bull, V. (1990). Bias and accuracy in attitude attribu-
tion: The role of attributional complexity. Journal of Experimental Social Psychol-
ogy, 28, 275-288.
Fletcher, G. J. 0., Rhodes, G., Rosanowski, J., & Lange, C. (1992). Accuracy and speed of
causal processing: Experts versus novices in social judgment. Journal of Experi-
mental Social Psychology, 28, 320--338.
Flett, G. L., Pliner, P., & Blankstein, K. R. (1989). Depression and components of attribu-
tional complexity. Journal of Personality and Social Psychology, 56, 757-764.
Psychological Bulletin, 101, 75--90.
Furnham, A. F. (1989). Lay theories: Everyday understanding of problems in the social
sciences. Oxford: Pergamon Press.
Furnham, A., & Henderson, M. (1983). Lay theories of delinquency. European Journal of
Gilbert, D. T., Pelham, B. W., & Krull, D. S. (1988). On cognitive busyness: When person
perceivers meet persons perceived. Journal of Personality and Social Psychology,
54, 733-741.
tory. American Psychologist, 35, 60~18.
Greenwood, J. D. (1989). Explanation and experiment in social psychological science.
New York: Springer Verlag.
Griggs, R. A., & Cox, J. R. (1983). The elusive thematic-materials effect in Wason's
selection task. Quarterly Journal of Experimental Psychology, 35, 519-533.
Hastie, R. (1984). Causes and effects of causal attribution. Journal of Personality and
Hinz, V. B., 'lindale, J. E., Nagao, D. H., Davis, J. H., & Robertson, B. A. (1988). The
influence of the accuracy of individuating information on the base rate information in
probability judgment. Journal of Experimental Social Psychology, 24, 127-145.
Hirschberg, N. (1978). A correct treatment of traits. In H. London (Ed.), Personality: A
new look at meta-theories (pp. 41Hi8). New York: Halstead Press.
Hooker, C. A. (1987). A realistic theory of science. New York: State University of New
York Press.
Howard, G. S. (1985). The role of values in the science of psychology. American Psychol-
ogist, 40, 255-265.
Kenrick, D. T., & Funder, D. C. (1988). Profiting from controversy: Lessons from the
person-situation debate. American Psychologist, 43, 23-34.
Klayman, J., & Ha, Y. W. (1987). Confirmation, disconfirmation, and information in hy-
pothesis testing. Psychological Review, 94, 211-228.
Kuhn, T. S. (1970). The structure of scientific revolutions (2nd rev. ed.). Chicago: Univer-
sity of Chicago Press.
Lepper, M. R., Ross, L., & Lau, R. R. (1986). Persistence in inaccurate beliefs about the
self: Perseverance effects in the classroom. Journal of Personality and Social Psy-
chology, 50, 482-491.
Manicas, P. T., & Secord, P. F. (1983). Implications for psychology of the new philosophy
of science. American Psychologist, 38, 399-413.
Markus, H., & Zajonc, R. B. (1985). The cognitive perspective in social psychology. In G.
Lindzey & E. Aronson (Eds.), The handbook of social psychology (3rd ed., pp. 137-
230). New York: Random House.
McCrae, R. R., & Costa, P. T. (1985). Updating Norman's "adequate taxonomy": Intelli-
gence and personality dimensions in natural language and in questionnaires. Journal
of Personality and Social Psychology, 49, 710-721.
McMullin, E. (1983). Values in science. In P. D. Asquith & T. Nickles (Eds.), Proceedings
of the 1982 Philosophy of Science Association (Vol. 2, pp. 3-23). East Lansing, MI:
Philosophy of Science Association.
McMullin, E. (1984). A case for scientific realism. In J. Leplin (Ed.), Scientific realism
(pp. 8-40). Los Angeles: University of California Press.
Newman, L. S., & Uleman, J. S. (1989). Spontaneous trait inference. In J. S. Uleman & J.
A. Bargh (Eds.), Unintended thought (pp. 155--188). New York: Guilford Press.
Nisbett, R. E., & Ross, L. (1980). Human inference: St~tegies and shortcomings of
social judgment. Englewood Cliffs, NJ: Prentice-Hall.
Popper, K. (1959). The logic of scientific discovery. London: Hutchinson. (Original work
published 1935)
Rippere, V. (1977). Commonsense beliefs about depression and antidepressive behaviour:
A study of social consensus. Behaviour Research and Therapy, 15, 465--470.
Rosenberg, S. C., Nelson, C., & Vivekananthan, P. S. (1968). A multidimensional ap-
proach to the structure of personality impressions. Journal of Personality and So-
cial Psychology, 16, 619--626.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attri-
bution process. In L. Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 10, pp. 173-220). New York: Academic Press.
Ross, M., & Fletcher, G. J. 0. (1985). Attribution and social perception. In G. Lindzey &
E. Aronson (Eds.), The handbook of social psychology (3rd ed., pp. 73-122). New
York: Random House.
Ryle, G. (1949). The concept of mind. New York: Barnes & Noble.
Semin, G. R., & Krahe, B. (1987). Lay conceptions of personality: Eliciting tiers of a
scientific conception of personality. European Journal of Social Psychology, 17,

199-209.
Showers, C., & Cantor, N. (1985). Social cognition: A look at motivated strategies. An-
nual Review of Psychology, 36, 275--305.
Shweder, R. A. (1975). How relevant is an individual difference theory of personality?
Sternberg, R. J., Conway, B. E., Ketron, J. L., & Bernstein, M. (1981). People's concep-
tion of intelligence. Journal of Personality and Social Psychology, 41, 37-05.
Strube, M. J., Lott, C. L., Le-Xuan-Hy, G. M., Oxenberg, J., & Deichmann, A. K. (1986).
Self-evaluation of abilities: Accurate self-assessment versus biased self-enhancement.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological per-
spective on mental health. Psychological Bulletin, 103, 193-210.
Tetlock, P. (1985). Accountability: A social check on the fundamental attribution error.
Social Psychology Quarterly, 48, 227-236.
Trope, Y. (1979). Uncertainty-reducing properties of achievement tasks. Journal of Per-
Trope, Y. (1980). Self-assessment and task performance. Journal of Personality and
Weiss, D. S., & Mendelsohn, G. A. (1986). An empirical demonstration of the implausibil-
ity of the semantic similarity explanation of how trait ratings are made and what
they mean. Journal of Personality and Social Psychology, 50, 595-601.
Wiggins, J. S. (1974). In defense of traits. Invited address at the Ninth Annual Sympo-
sium on Recent Developments in the Use of the MMPI, Los Angeles.
CHAPTER NINETEEN
A Commonsense Approach to
Personality Measurement
RAYMOND N. WOLFE
The sun, the moon, the clouds, and the stars provided ancient astrolo-
gers with clues to people's characters and destinies. Other oracles re-
lied on other sources of information: The hydromancer read the airs,
winds, and waters; the necromancer consulted with the dead; the
haruspex interpreted patterns of lightning or the entrails of a sacrificed
animal; the alchemist read shapes formed by a thin stream of molten
lead poured into a bowl of water. These soothsayers were regarded as
unenlightened by their rivals and descendants, who offered more per-
sonalized pronouncements based on elements of the client's own phys-
ical structure: The chiromancer read the lines on one's palm; the
physiognomist, facial features; the phrenologist, skull conformation.
More recently, psychologists have sought to discern people's character
by interpreting records of their self-expression, such as samples of
handwriting, projective test responses, and nonverbal behavior. Today's
methods of personality measurement arose from this venerable tradi-
tion. Many of them still contain remnants of the occult (the word comes
from the Latin occultus, past participle of a verb meaning to cover up,
to hide) that need to be dispelled.
When personality measurement began to emerge as a subdiscipline
of psychology in the 1920s and 1930s, much of it took place in psychiat-
RAYMOND N. WOLFE • Department of Psychology, State University of New York,

Geneseo, New York 14454-1401.
New York, 1993.
269
270 RAYMOND N. WOLFE
ric institutions. As Szasz (1968, 1977) and others have pointed out, an
adversarial relationship exists between patients and clinicians in such
settings; assessment in this context is primarily a ritual adjunct to
pejorative labeling.* It therefore behooved the early clinical psycholo-
gists to swathe this part of their craft in mystery. They did so by
developing instruments that were bewildering to laypersons. Projec-
tive tests were well suited to this purpose and soon became a basic tool
of the clinical trade.
Objective inventories, some of them consisting of hundreds of
statements presented for endorsement, tended to be similarly bewil-
dering. The person taking such tests could not be very sure of what
was being measured or how; this was thought to be a desirable state of
affairs from the psychologist's standpoint. Confronted by a great many
statements diverse in content, and ignorant of how one's endorsements
were to be scored, weighted, combined, and mapped onto a personality
profile, the test taker would be hard put to dissimulate systematically
in responding. Even so, psychologists thought that further safeguards
against strategic self-presentation were necessary; procedures in-
tended to detect lying, self-contradiction, and random responding were
devised and incorporated into many scales.
Adversarial logic was not confined to the psychiatric hospital and
the mental hygiene clinic. Psychologists in university and industrial
settings mistrusted their subjects too. Many of the instruments devel-
oped for use with college samples were constructed in the same way as
clinical tests, with subscales designed to spot deliberate or undeliberate
fakery on the part of respondents. Scales that assessed only a single
construct contained buffer items to partly conceal the purpose of the
test by misleading the test taker about what was presumably being
measured. It was taken for granted that an appreciable proportion of
subjects either could not or would not give truthful reports about
themselves. t
Partly because of its origins in clinical psychology, then, early per-
sonality measurement resembled espionage. It was a cat-and-mouse-
like endeavor, with the psychologist seeking clever ways to penetrate
the subject's supposedly elaborate and shifting psychic defenses. The
question always in the background (and often at the focus of attention,
as in the extensive work on response styles) was How can students of
*Anderson (1981) describes the polity's need for experts who are willing to assume the
responsibility of identifying surplus individuals and populations. Szasz (1977, chap. 14)
describes the distinction between technical and ceremonial uses of assessment.
t For an interesting discussion of deception in modern psychology, see Scheibe (1978).
COMMONSENSE APPROACH TO PERSONALITY MEASUREMENT 271
personality obtain trustworthy self-reports from their subjects? Sus-

tained attacks on the problem gained little ground. Their outcomes
compel two conclusions: There is no solving it, and in seeking to solve
it, one is likely to mystify oneself. Szasz (1984) provides a coda: "But
should we not rather rejoice that men who seek to deceive others, even
with the best of intentions, in the end only deceive themselves? So long
as that remains true, there is hope for mankind" (p. 169).
The institutional machinery of science is designed to facilitate self-
correction; its workings are evident in a trend toward increasing straight-
forwardness in personality measurement that seems to have begun in
the 1930s. Gordon Allport had a hand in it. His teachings, together
with efforts of many other psychologists (e.g., Campbell & Fiske's in-
troduction in 1959 of multitrait-multimethod analysis) and the
parapposite development of electronic data processing led to some salu-
tary advances.
Allport urged psychologists to rely on common sense in trying to
understand personality (Allport, 1937) and in devising and using assess-
ment procedures (1961, p. 399). He advocated multimethod measure-
ment wherever feasible, but tended also to favor direct over indirect
methods (e.g., objective over projective techniques) (Allport, 1953). Fi-
nally, he contributed objective measures of his own (Allport & Allport,
1928; Allport & Ross, 1967; Allport et al., 1960).
Objective personality assessment started to emphasize direct meas-
ures in the 1960s. Instruments appeared in which every item was face
valid. Investigators began to act on the assumption that if you want to
find out something about someone, the best way to do it is to ask him,
as clearly and directly as possible, and take his answer at face value.
From the standpoint of the earlier personality testers, a psychologist
would be naive to proceed this way.
But there is much in favor of such a commonsense approach. It is
conservative; it does not presume that the psychologist can know more
about a person than the person himself can know or will tell. It is
parsimonious in that it minimizes the number of inferential steps be-
tween the response and the psychologist's inference about where the
subject stands on a particular trait dimension. It is demystifying: The
psychologist need not masquerade as a wizard; the subject can partici-
pate as a reporter about himself (in Jessor's phrase, "an ethnographer
of his own life") rather than as a fugitive adversary. And with regard to
technical adequacy, tests based on commonsense assumptions compare
favorably with tests based on other rationales (Ashton & Goldberg,
1973; Scott & Johnson, 1972).
The steps in developing an instrument via the commonsense ap-
proach were specified by Jackson (1970). Outcomes of my own studies

convince me that, in certain circumstances, the sequence he prescribes
is more efficient than any other. In this chapter, I define the circum-
stances in which Jackson's method is superior and present embellish-
ments of it, along with some minor qualms.
The would-be test builder is confronted by a long list of challenges,
many of which are difficult to meet. To ignore any of them, or to fail to
meet any of them, reduces the probability that the measure will be
able to hold up under multitrait-multimethod analysis. But with certain
subject populations and under certain conditions, the method de-
scribed here is the best available. The present prescription may be
helpful also to users of personality tests; to select wisely from an
array of inventories, prospective users need to know the hurdles that
test builders ought to have negotiated en route to their final lists
of items.
The prescription for commonsense measurement consists of an-
swers to three broad questions:
1. Under what conditions is it most reasonable to assume that sub-
jects will give truthful reports about themselves in responses to
personality test items?
2. Given these conditions, how can one devise a suitable set of
prospective items?
3. Given such a list of items, what criteria can one apply so as to
retain the best and eliminate the poorest?
VERIDICALITY
The assumption that subjects will describe themselves accurately

on personality tests is most defensible when there are grounds for
believing that subjects are
1. able to read and understand the instructions and content of each
item;
2. motivated throughout the test to give honest and accurate
responses;
3. free from confusion about what the scale as a whole is intended
to measure; and
4. free from confusion about the meaning of each item.
The importance of these conditions can hardly be overemphasized. If
the subject is illiterate, not motivated to reply truthfully, or confused
about what it is he is being expected to report about himself, his re-
sponses are worse ,than useless in the hands of a psychologist who

erroneously believes them to be useful. To the extent that any of the
conditions is not fulfilled, it is extremely difficult to know what, if any-
thing, a response signifies.
The literacy condition implies some obvious limits. Objective per-
sonality tests are ill-suited for subjects below the age of 10, for sub-
jects whose intelligence is below the normal range, for poorly educated
adults-that is, anyone whose reading ability is suspect; their applica-
bility with elderly people is also questionable:
Currently available personality tests have not proven reliable with
older adults. Test items are likely to have been written for and
normed on younger and better-educated persons. Older persons
may also fatigue more easily, may not like being asked the sorts of
questions often included in personality tests, and may be more in-
hibited and subject to response sets. (Hogan et al., 1985, p. 43)
Further delimiters are implied by the second condition, motivation.
Willingness to participate and to give veridical self-reports may be low,
for example, among subjects in total and near-total institutions,
where inmates and staff are likely to be adversaries. Even in ostensi-
bly ideal institutional settings, such as university research centers or
psychology departments, subjects may take a dim view of the assess-
ment enterprise. Their attitudes can be adversely affected by informa-
tion from sources over which the individual investigator has little
control: the mass media, the campus grapevine, the subjects' own
educational experiences, the ambiance created by recent and current
research events.
It is also likely that the investigator's local reputation can influence
a potential subjects' expectancies concerning participation in a particu-
lar project. Ordinary ethical conduct is essential. Research promises
must be kept. One must be direct and honest with subjects in all phases
of a study, from initial recruiting through the final providing of feed-
back in the form of group results. Confidentiality of results must be
maintained during and after a study. Beyond their importance with
regard to subjects' rights, these rules are an integral part of the fabric
of straightforward assessment.
Another part of the fabric is the test materials themselves. In
creating these, the investigator must take pains to assure that they will
not mislead the subject either by design or through carelessness. The
test builder has to labor over general and specific instructions until
they are as simple and unmistakable as can be. Instructions should also
include an invitation to the subject to signal the examiner if further
clarification of purpose, items, materials, or anything else about the

testing situation is desired.
In the body of the questionnaire, all items intended to measure a
particular trait should be grouped together* and headed by a designa-
tion of the trait name (e.g., "The items on this page are intended to
measure how energetic you are" or "Items 8 through 17 pertain to
satisfaction with life"). Although earlier psychologists would have re-
coiled from such clarity and directness (it invites dissimulation by the
subject), such headings turn out to be valuable because they serve to
reduce misunderstandings on the part of the subject.t
Response format should be as simple and unmistakable as possible.
A yes-no or true-false format is ideal; however, the Likert format pro-
vides a range that is psychometrically advantageous in the early stages
of test construction. So here, a compromise is recommended: Likert
format throughout item analysis and up to the point at which the in-
strument has exhibited a satisfactory array of multitrait-multimethod
relationships and then it may be feasible to switch to a simpler format.
Instructions and the items themselves should consist of words that
are easy to read, arranged in short, simple, self-referential statements
(Amelang and Borkenau [1986, pp. 27-28]) cite several empirical stud-
ies showing that items having these features tend to be more valid
than items lacking them. The meaning of each item must be as unequiv-
ocal as human ingenuity can make it. And this meaning should corre-
spond in an obvious way to the trait-name heading under which the
item appears. Where these conditions are fulfilled, it becomes possible
to begin to exploit ordinary language and common sense in the meas-
urement of personality.
* Interspersal of items measuring various traits has been a common practice in the con-
struction of global inventories. By itself, this does not amount to a serious defect because
sets of dependable items tend to yield dependable variables whether they are inter-
spersed or not; such sets of items tend to cohere predictability, more or less regardless
of the field of items in which they are embedded. Interspersion is thus to some extent
defensible; but because it is a potential source of confusion to the test taker, the common-
sense approach forbids it.
t Although "transparency'' of items has been regarded as a defect (e.g., Epstein, 1984,
p. 312), it is necessary and desirable from a commonsense standpoint. Aiming to enlist the
subject as a reporter of his own characteristics and seeking to achieve this end by
facilitating accurate self-report, straightforward assessment seeks to maximize clarity
and directness without insulting the respondent's intelligence. The work of Knowles
(1988) demonstrates that the responses of college students, at least, become more reliable
as it becomes clearer to the subject exactly what it is that he's being asked to report
about himself; this implies that higher-quality responses are obtained when the trait
being measured is known than when it is unspecified or concealed or disguised.
DEVISING ITEMS
PRELIMINARY ASSUMPTIONS
The commonsense approach makes no distinction between traits

and behavioral dispositions. It simply presumes that people differ in
terms of the traits they possess. It is compatible with most definitions
of dispositional constructs, including the causal, act frequency, and
conditional views described by Wright and Mischel (1987, p. 1160). It
is premised on the beliefs that there are individual differences in
the strength of people's dispositions to act in certain ways; that these
are measurable and worth measuring; that in circumstances conducive
to veridicality an efficient way to do this is to ask appropriate self-
referential questions grouped together in scales; and that such scales
should be devised by the rational method. Finally, it assumes that a
scale must measure one and only one construct [Briggs and Cheek
(1986), and Wolfe and Kasmer (1988), illustrate the confusions that can
arise when psychologists use scales that measure more than one con-
struct] and that a relatively small number of items will suffice for a
given construct.
A personality scale is one way of operationalizing a construct that
is embedded in a theory. The theory identifies the construct (along with
other constructs); the scale quantifies it. It is therefore necessary to
start with a theory or a minitheory that specifies two or more con-
structs and the supposed relationships between them. This accords the
theory its warranted logical priority and at the same time compels the
test builder to start thinking in terms of convergent and discriminant
validity at the very outset.*
To devise good items within these constraints is an exacting task.
Sustained attempts to meet this challenge sometimes produce the (per-
haps illusory) impression that one's understanding of the construct and
theory at issue has been sharpened by the effort (although, at a later
stage, subjects' responses can, and must be permitted to, have a de-
cisive role in this respect) (Jackson, 1970, p. 75). It seems plausible,
though, that one may acquire greater awareness of a construct's mean-
ing by trying repeatedly to distinguish between its "core" and its "cor-
relates" [or, in Wright and Mischel's terms, central and peripheral acts,
respectively (1987, p. 1163)]. This is one of the aspects of scale con-
*These points echo two of the essential principles of scale development laid down by
Jackson (1970, p. 63).
struction in which Campbell and Fiske's (1959) ideal-refining theory

and measurement at the same time-can be realized.
FROM THE THEORY TO THE ITEM
The more fully a theory specifies its constructs' relationships with

behavior and with each other, the easier it becomes to devise items.
The construct of religiosity within Jessor and Jessor's (1977) theory of
problem behavior, for example, is quite well specified in terms of its
behavioral referents and its presumed direction of correlation with
other constructs. Given this rich nomological net, one can readily gen-
erate prospective religiosity items. The construct of private self-
consciousness described by Fenigstein, Scheier, and Buss (1975), for
example, is poor by comparison, with regard to both behavioral refer-
ents and connections with other constructs. It is consequently difficult
to think of items likely to be sensitive to individual differences in
private self-consciousness and not something else. Data described by
Angleitner, John, and Lohr (1986) show that those dimensions of a con-
struct that are most fully explicated in the theory tend to be most
heavily represented among test items (pp. 67-76); this should be true
by design alone, but is also probably due in part to ease of generat-
ing items.
Angleitner and co-workers identify six categories of item content:
characteristic activities; attribution of traits to oneself; wishes, in-
terests, and preferences; biographical facts; attitudes and beliefs; and
others' reactions to oneself. These can be thought of as the reportable
facets or dimensions of a construct-the ways it may manifest itself
tangibly enough for the person to recognize it and therefore be able to
give an accurate self-description regarding it. Proportions of item con-
tent in the finished scale ought to correspond approximately to the
theory's specifications as to how the construct is supposed to manifest
itself. Item writers should therefore apportion their efforts accordingly
(even though it often turns out that the number of items created in a
given content category is a poor predictor of the number ultimately
retained). The number of items initially devised will depend on many
variables apart from richness of theoretical detail: desired length of the
finished scale, number of item writers, their familiarity with the con-
struct and theory, and their steadfastness.*
*Item writing can be done as a convivial activity or both solitarily and in groups. Brain-
storming sessions can facilitate this phase of the work; when they are part of the process,
productivity is doubtless affected by group dynamics as well.
Each item is describable in terms of surface characteristics and

meaning characteristics (Angleitner et al., 1986, pp. 80-93). Surface
characteristics can be determined objectively: They are length (number
of words or letters, number of clauses), complexity (person, voice,
tense, mood, negation), and sentence type (grammatical case). Meaning
characteristics are not so readily ascertained, but can be rated reliably
by trained judges: They are ambiguity, comprehensibility, abstractness,
self-reference, and desirability of endorsement.
The prescription for straightforward assessment is clear for all
features of the item itself. Shortness is prized. Word count should be
held to perhaps 15 at most. Qualifying words and phrases should be
held to a minimum. The longer an item is, the more complex it is likely
to be; the more complex it is, the more alternative interpretations it
permits; the more interpretations it permits, the more interpretations
subjects will give it; the more interpretations subjects give it, the less
psychometrically useful it becomes.
Simplicity and directness are also prized (Scott & Johnson, 1972).
Items should be stated in the first person, the active voice, the present
tense (except, of course, for biographical facts), and the indicative
mood. Negatives are to be avoided; they are confusing (Holden &
Fekken, 1990; Matlin, 1983, p. 142), and, as Fletcher (1984) argues,
"For mental dispositions, ... positive instances of behavior should as-
sume a much more potent diagnostic role than negative instances"
(p. 210) [the findings of Ahlawat (1985)* and Funder and Dobroth
(1987) support Fletcher's position].
Declarative and interrogative sentences seem to be equally suit-
able, although the respondent's task becomes more difficult when both
are included in the same scale (response format has to be compatible
with sentence type; complications arise when format varies within a
scale). Some rules of thumb for writing simple items: Avoid commas,
connectives, negatives, and subjunctives.
Angleitner and co-workers present criteria for evaluation of mean-
ing characteristics, along with results of preliminary analyses of 1051
items. The five meaning characteristics turn out to be weakly related
to one another, although some pairs yield significant correlations. Am-
biguity, for example, correlates -.22 with comprehensibility and .26
with abstractness. Some surface characteristics are also associated with
certain elements of meaning (e.g., item length correlates approximately
-.60 with comprehensibility).
*I am indebted to Michelle Picioccio for bringing Ahlawat's excellent research to my

attention.
After items are written and edited into presentable form with re-
spect to their surface characteristics, their meaning characteristics
should be evaluated independently by judges using Angleitner and co-
workers criteria, or some variation thereof. This step can identify
statements that need to be altered because they are equivocal or other-
wise suspect. Improvements in clarity cost less, of course, if they can
be achieved early in the sequence. The wisdom of eliminating prospec-
tive items at this stage is questionable, although some deselection here
may be justified if a great many items are available. (Where possible,
one ought to avoid imposing unnecessarily on subjects.) Because psy-
chometric criteria provide the most defensible grounds for item elimi-
nation, however, most decisions to retain or not are best left until a
later stage. ltemmetric facts are associated with and in a sense under-
lie psychometric facts (for documentation, see Angleitner et al., 1986;
Holden et al., 1985) but should not be used in lieu of them. A depend-
able psychometric technology for item selection exists; test builders
can and should rely on it.
Still, the importance of itemmetric analysis as a first step can
hardly be disputed. Straightforward assessment demands that a test
item must not bewilder the subject; careful examination of an item's
surface and meaning characteristics provides a good safeguard against
such bewilderment.
FROM ITEMMETRICS TO PSYCHOMETRICS
When each of the face valid items designed to measure a particular

construct has been tailored to the point where its surface and meaning
characteristics are satisfactory, the test builder will be eager to admin-
ister the first draft of the new scale to subjects. But should the test
builder be seeking to devise measures of other constructs as well (a
useful strategy not only in the interests of theory development but also
because it more or less compels one to focus on discriminant validity
right from the outset), it may be more efficient to delay data collection
until all prospective items in all the scales have satisfied itemmetric
requirements.
A large sample-at least 200-of the subject population the test is
intended for must then be recruited and tested under good physical
(adequate space, seating, lighting, and ventilation; comfortable temper-
ature; freedom from external distractors, etc.) and psychological (free-
dom from coercion, from time constraints, from concerns about
confidentiality, etc.) conditions.
The distribution of responses to each item is first examined for
central tendency and dispersion. Any item that yields either a sharply
peaked distribution or an asymmetrical distribution is at once suspect;
nevertheless, all items should be retained for further analysis. It is
quite likely that items showing little dispersion of response and/or ap-
preciable skewness will fare poorly in other analyses. Items exhibiting
either of these features can be suspected of eliciting style-determined
rather than content-determined responses (Jackson, 1970, p. 73) and
will not correlate highly with total scale score (Angleitner et al.,
1986, p. 96).
Estimates of homogeneity are calculated next. These include inter-
item and item-total correlations, Kuder-Richardson reliability, and coef-
ficient alpha. To justify the tester's procedure of adding together the
subject's responses to different items, it is necessary that these re-
sponses exhibit a high level of homogeneity across items: 1.00 is the
ideal. Classical measurement theory specifies interchangeability of
items as a necessary condition for additivity. In practice, however, it
turns out that the content of pairs of items that intercorrelate above .5
is often markedly redundant. Students of personality should therefore
be willing to accept sets of items with interitem correlations in the
range of .2 to .4, which "would seem to offer an acceptable balance
between bandwidth and fidelity" (Briggs & Cheek, 1986, p. 115). Al-
though there is room for debate as to what constitutes an optimal level
of correlations among items, it is obvious that negative interitem corre-
lations or very low positive interitem correlations raise major questions
about the meaning of a total score that is the sum of responses to such
diverse items.
One can increase Kuder-Richardson reliability and the value of co-
efficient alpha by lengthening the scale with parallel items (i.e. items
that merely restate the content of items that already cohere). Some
psychologists may have resorted to this expedient in order to be able
to report higher reliability values.
Can the desired level of internal consistency be obtained through
more legitimate efforts? For a scale to be properly content-saturated
(Jackson, 1970), all of its items have to show purity by exhibiting high
loadings on the factor the scale is supposed to measure and zero load-
ings on factors the scale is not supposed to measure; in other words,
every item must pass the tests of convergent and discriminant validity.
Jackson's (1967) development of the Personality Research Form follows
the steps dictated by this rationale. He started with a fairly detailed
theory-Murray's (1938) theory of psychogenic needs-and sought to
develop a separate scale for each of the 20 needs; these constitute an
excellent base from which to establish the necessary discriminant valid-
ity. More than 100 candidate items were written for each scale. After
sustained efforts at refinement, he arrived at finished scales, most of
which exhibited both types of validity.
I would like to think that useful instruments can be devised with-
out undertaking so Herculean a task. However, there is no getting
around the fact that several scales need to be developed simultaneously
(or, alternatively, that other established scales have to be used along
with those being developed) if discriminant validity is to be considered
systematically from the beginning.
Test builders with aims more modest than Jackson's can avail
themselves of many techniques for sifting prospective items. Beyond
interitem correlations and other well-known estimates of internal con-
sistency stands a large array of factor analytic methods, many of which
offer the kind of information needed in deciding to retain, eliminate, or
alter prospective items. These should be exploited, but it needs to be
done thoughtfully. The danger is that in using techniques so efficient,
so streamlined, one may forget that it is necessary to touch base from
time to time with fresh batches of empirical reality in the form of data
from new and different groups of subjects (with N=200 or more in each
group to assure stability of the correlation matrix).
Having calculated coefficient alpha for a set or subset of more than
enough prospective items, for example, one may winnow through the
set by recalculating alpha repeatedly, each time with a different sus-
pect item removed. If the value of alpha goes up when a given item is
removed, that item is a good candidate for discarding. When all the
items thus identified are discarded, coefficient alpha for the remaining
items is acceptably high. Another example: One has an abundance of
items that could conceivably form a set of clusters compatible with
theory. One then goes fishing for this particular structure through suc-
cessive approximation, conducting one factor analysis after another and
eliminating one or more items at each step. After several such analy-
ses, one may arrive at the desired structure. In both of these examples,
the investigator capitalizes on idiosyncracies of the data set at hand,
thereby decreasing the likelihood of successful cross-validation.
It is, of course, absolutely necessary that stable factor structures
be demonstrated sooner or later. This is always a difficult challenge,
but it is somewhat easier to meet when test builders are following the
commonsense procedure. After two or three careful revisions followed
by new data collections, the items and scales start to become known
quantities. Their action becomes fairly predictable; the likelihood of
unpleasant surprises (e.g., shrinkage on cross-validation, occurrence of
unexpected or uninterpretable factor structures) diminishes rapidly.
As Jackson says, "Successive attempts must be made to approach

a variety of optimal properties" (1970, p. 65). At the item level, these
properties include all the elements of itemmetric adequacy, plus con-
vergence with the other items, discriminant validity with respect to
items known to measure other constructs that are theoretically unre-
lated to the construct at issue (and preferably a large number of such
other constructs, the more the better), and freedom from the influence
of desirability. Most psychometricians would also insist on adding tem-
poral stability to this already rather daunting list.
FROM THE ITEM TO THE SCALE
The next question is, given a set of items each of which exhibits all
of these optimal properties, can they be assembled into a scale that
faithfully represents the construct? Jackson (1970) invites attention to
content bias as a potential problem: It is quite possible that items in-
tended to measure one or another facet or manifestation of the con-
struct will have been eliminated disproportionately. In such a case, the
test builder must backtrack, perhaps to the stage of preparing brand-
new items, and work again through the itemmetric-psychometric se-
quence. At some point, the test builder is loath to retrace these steps
one more time and decides to present the scale as "finished" (all this
means is that the test builder has chosen-to use Kafka's classic oxy-
moron-to temporarily abandon the effort to improve it further). The
finished scale, then, may misrepresent the construct in some important
way or ways. It is in this respect that the psychological adage "The
measure is the theory" is importantly true (see Rocklin & Revelle,
1981, for an example).
Degree of correspondence between scale and construct is the cen-
tral question of content validity. The answer to it is usually arrived at
through inspection and informed opinion; statistical analysis of re-
sponses is less decisive on this matter than on the other elements of a
scale's adequacy. With consensus to the effect that the scale does fairly
represent the construct, this requirement is met. If consensus is not
sought, or is sought but not obtained, either of two scenarios is likely
to ensue. In the first, multitrait-multimethod analyses call attention to
the disparity and lead to attempts at revising the scale in order to
improve the match.
In the second, the measure can threaten to supplant or become the
theory. This is particularly likely when the test builder and the theorist
are the same person. Given the considerable investment of effort that
test construction entails, the psychologist may well be motivated to
make a "satisficing" compromise in judging the match between scale

and construct, and this conclusion will probably favor the scale at the
expense of the theory. In presenting the scale, then, the psychologist
may be motivated to play down or ignore altogether the elements of
mismatch. Consequences of the second scenario are harmful indeed:
Confusion is sown among students; prospective users of the scale are
misled. Thus measurement, which ought to be the servant of theory,
can come to dictate it even before the data are in.*
SCALE BREADTH VERSUS SCALE LENGTH
Constructs differ in breadth (Hampson et al., 1986) as well as with

regard to the richness of the nomological net in which they are embed-
ded. As breadth and/or number of specified relationships within the net
increase, so does the difficulty of devising a scale that can adequately
represent all facets of the construct without using a very large number
of items. The trade-off between length and breadth is the focus of an
edifying debate between Burisch (1984a,b, 1985, 1986) and Paunonen
and Jackson (Paunonen, 1984; Paunonen & Jackson, 1985).
Burisch argues for shortness. He demonstrated that length is not
necessarily associated with validity and makes a convincing case for the
cost-benefit advantages of scales consisting of less than ten items.
Paunonen and Jackson contend that aggregating across larger numbers
of items (say, 16 or 20 per scale) produces a more reliable measure and
one that better represents the construct's various facets. A longer
scale, they claim, should therefore be capable of predicting a greater
variety of criterion behaviors and might fare better in cross-validation.
In the end, however, Paunonen and Jackson concede that "As observed
by Burisch ... a judiciously abbreviated test can for practical purposes
be as useful as the full measure in certain instances" (1985, p. 341,
italics in original).
Burisch's position does seem radical in that it tends to divorce
the instrument from the theory. Psychologists should, he says, present
*Later on, as empirical results begin to accumulate, the psychologist may be tempted to
let data lead the theory more powerfully than they ought to. This temptation needs to be
resisted because, like the method of resolving mismatch in the second scenario, it is
conducive to a sterile state of affairs in which methodology takes complete precedence
over theory. If this imbalance of power is permitted to prevail, future historians of
psychology will find it easy to sum up twentieth-century developments in personality
research: "Advances occurred in the area of measurement, as psychologists progressed
from reifying traits to reifying trait-method combinations."
subjects with easily understood trait labels or undisguised content-

oriented statements rather than with restatements of the theoretical
definition of the construct. It is more necessary that subjects give accu-
rate reports about themselves than that the scale represent all facets
of a construct. The scale's fidelity to the construct is beside the point if
subjects are unable or unwilling to give truthful responses to it:
That is why it is not wise to keep them busy for too long, have
them read lengthy instructions, or require them to strain their
brains on word meanings different from everyday usage. This trap
avoided, it is also important not to fool ourselves regarding the
quality of the resulting data. Given the coarse grain of the language
we converse in, we should stop expecting major breakthroughs
from polishing verbal instruments. (Burisch, 1984n, pp. 96-97)
Several recent reports address the question: At what point does
one begin to encounter diminishing returns with regard to scale length?
Along with Burisch's demonstration that the number of items in the
Hamburger Depressions-Skala can be drastically reduced, there have
been many comparable efforts to shorten other scales (Cacioppo et al.,
1984; Hensley & Waggenspach, 1986; Leary, 1983; Perri & Wolfgang,
1988; Strube, 1986; Tomaka et al., 1987; Wolfe et al., 1984). Results are
essentially the same in all cases. When items are retained on the basis
of an internal consistency standard, it is possible to reduce scale length
by 50% or more at little apparent expense in terms of reliability or
validity (despite the range attenuation disadvantage that necessarily
accompanies reduction in number of items). Thus the early returns are
promising; when abbreviated scales come into more extensive use and
are used to predict a greater variety of criteria, they could compare
favorably with their longer counterparts.
Burish's proposal that a small number of items may prove suffi-
cient for measuring certain personality constructs appears to be defen-
sible. The probable cost of relying on small sets of items-though
some predictive capacity is apt to be lost-may be tolerable when it is
weighed against the advantages of reduced demands on subjects' pa-
tience and cooperativeness. Although aggregation across items indis-
putably offers important advantages (Epstein, 1984), there appear to
be some construct-item combinations for which it is less necessary.
Studies in my laboratory (Meyers et al., 1986; Tomaka et al., 1987;
Wolfe, 1986) suggest that the case for short scales is most persuasive in
precisely those circumstances that are most compatible with the com-
monsense rationale; namely, where the construct being measured is
already familiar to the respondent, is capable of being asked about in
very plain language, and has observable behavioral referents. It also

seems that items instructing subjects to give what amounts to a self-
report of behaviors aggregated across occasions or situations-what
Wright and Mischel (1987) call summary constructs-need not be so
numerous as other kinds of items.
There are risks as well as benefits associated with the use of very
long and very short scales. If investigators are aware of the trade-offs
involved, they can consider these in view of the circumstances and aims
of their research in deciding which scales to use for personality assess-
ment. Whereas short scales might confer few advantages in work with
college students, for example, they could well be decisive in eliciting
cooperation from samples of older adults (Hogan et al., 1985).
CONSTRAINTS ON THE KINDS OF CONSTRUCTS

THAT CAN BE MEASURED
During the past two decades, students of personality have become

increasingly comfortable with a ''big five" set of higher-order factors or
broad traits: surgency, agreeableness, conscientiousness, emotional sta-
bility, and culture. These emerge so dependably that they can be said
to fulfill the requirement of critical multiplism (Houts et al., 1986). At
last, we are starting to get somewhere!
Why, one may ask, is it these traits and not others that show up
time and again? Why are self-deception, self-consciousness, and self-
monitoring, for example, not on the list? From the commonsense stand-
point, there are three good reasons. The first is that each of the "big
five" can be translated directly into everyday language: surgency=
energeticness; agreeableness and conscientiousness need no translat-
ing; emotional stability=calmness or serenity; and culture=bookishness.
Second, each translation is meaningful to the average layperson. Third,
each translation has some relatively unequivocal behavioral referents,
which makes it possible for people to describe themselves and their
acquaintances more or less accurately on each dimension. McCrae and
Costa (1987) have documented the capacity of these broad traits, and
most of the several facets or dimensions of each, to hold up very well
under multitrait-multimethod examination.
The constructs in the alternate list-self-deception and so on-
have no simple counterparts in ordinary language, tend to be more
difficult for the average layperson to comprehend, and have few obvi-
ous or unequivocal behavioral referents. For these reasons, they can be
expected to resist measurement technology; it is difficult to imagine
that these constructs could yield the required pattern of correlations in

a multitrait-multimethod matrix (which is not to say that all the dispo-
sitions that matter are subsumed by the "big five").
Results of a study of 142 college students by Meyers and co-workers
(1986) demonstrate that convergence between scale scores, self-ratings,
and peer ratings varies directly with the construct's observability (and
perhaps with how easy it is for the laypersons to comprehend). Here
are the convergent correlations found by Meyers and colleagues be-
tween scale scores and the average of ratings by between three and six
acquaintances: religiosity, .74; alcohol use, .73; physical ability, .60; so-
ciability, .56; boredom susceptibility, .25; concern for appropriateness, .24.
This is a pattern with clear implications for an individual-differences
approach to personality. It shows that those traits that have unmistak-
able behavioral referents are most likely to exhibit the convergence
required in support of the claim that a construct has real existence or
can be dependably measured. The correlations for boredom susceptibil-
ity and concern for appropriateness come very close to being nonsignif-
icant (a one-tailed test is required); although these constructs are
comprehensible to respondents, they lack the unequivocal behavioral
referents that the other constructs possess. By examining scales of
different lengths and using different numbers of rating items for their
constructs, Meyers and co-workers are also able to show that the num-
ber of items needed for adequate measurement of a trait varies in-
versely with observability of trait-related behaviors and, as Gorsuch
(1984) has noted, with the trait's familiarity to respondents.
It follows that there are limits on the kinds of constructs we can
hope to measure successfully. The more readily understandable a trait
or disposition is to the literate layperson, the greater the likelihood
that a useful scale can be devised to assess it. The psychometric excel-
lence of Chapman and Chapman's (1987) measure of handedness, for
example, is attributable partly to the fact that everyone is familiar with
the construct of handedness. But observability remains the sine qua
non (Amelang & Borkenau, 1986; Funder & Dobroth, 1987).
Adequate measurement requires the demonstration that at least
two traits, when operationalized by two maximally dissimilar methods,
exhibit the expected pattern of convergent and discriminant coeffi-
cients. In most cases, ratings by acquaintances or other knowledgeable
informants will comprise one of the methods against which self-report
instruments are validated. Where ratings are used, both subjects and
raters must be laypersons if the results are to be taken as representa-
tive of what goes on in everyday life (this is another reason why the
psychologist should stick to ordinary language when devising con-
structs and instruments). The implication, of course, is that valid as-

sessment of individual differences in personality can be achieved only,
or at least most handily, for those dispositions the layperson is familiar
with. This state of affairs not only restricts the range of traits being
studied, but in effect reduces trait theory to the status of a merely
descriptive endeavor and one limited to cataloguing what, in a sense,
nearly everybody already knows anyway at that.
Put this way, the constraints are perhaps overstated [cf., Wright
and Mischel (1987, p. 1160) for a slightly more optimistic picture], but
the fact remains that psychometric adequacy has its price; we must
"acknowledge limits in order to proceed with power and confidence"
(Gould, 1987, p. 16).
Without psychometrically adequate measures, we have only specu-
lation and no decisive ways to tell the idle from the valuable. A measuring
technique is needed; the one offered here appears most promising. To
adopt it, one must also accept the necessary constraints, which, after
all, do not rule out the possibility of robust and informative empirical
work.
CONCLUSIONS
The trend toward a commonsense approach to personality theory

and assessment seems to have gotten its start in the writings of Gordon
Allport. In the 1960s, psychologists increased their reliance on person-
ality tests consisting largely or entirely of face-valid items. In 1970,
Jackson presented a sequential method for developing such tests. Dur-
ing the past two decades, students of personality have tended more and
more to follow commonsense assumptions in measurement, and many
tests have been devised in accordance with Jackson's prescription. This
trend has enhanced the quality of instrumentation, and improved in-
struments have in turn led to several replicable demonstrations of con-
sistent patterns of individual differences in personality traits.
Future psychologists will doubtless remain aware of the value of a
commonsense approach. Whether they will subscribe to Burisch's view
that properly abbreviated scales are in the end more advantageous
than full-length scales remains to be seen. I believe the commonsense
approach, as embodied in Jackson's 1970 prescription, can ultimately be
compatible with Burisch's proposal (despite the slightly ironic fact that
Jackson remains one of the stoutest opponents of this proposal).
At present, psychologists are favoring shortness of scales and
items. In his 1986 book, for example, Buss describes 37 relatively new
personality scales by numerous authors and lists their items. They av-
erage eight items in length and for the most part are straightforward,
with simple response formats and short, face-valid, clearly stated items.
It is encouraging to see this set of instruments; such an array could
probably not have been found 20 or 30 years ago, much less presented
and recommended for use.
For scientific-as opposed to ceremonial-purposes, the common-
sense approach tends to be superior to other techniques of measurement.
From a practical standpoint, it minimizes demands on the test taker,
and thus, in principle at least, should enlarge the lists of populations
and subjects who will provide usable protocols in studies of personality.
It has some limitations as well. It tends to be sheerly descriptive;
its advantages over other methods of assessment are probably minimal
for traits that lack observable behavioral referents (and perhaps for traits
that are not familiar to laypersons); its efficiency is questionable in
certain applied settings and with subjects whose reading skills are poor.
The study of personality had progressed to the point where ad-
versarial logic is no longer the order of the day. And psychologists can
now avoid deceiving themselves in at least two important respects:
mistaking method variance for trait variance and being content to
know that a scale works (i.e., exhibits predictive validity) without
knowing why it works. Multitrait-multimethod analysis enables psy-
chologists to tell the one kind of variance from the other. Devising and
using scales that measure one and only one construct (as Briggs &
Cheek, 1986, urge) helps to counteract psychologists' willingness to be
satisfied when a total score on a factorially complex instrument pro-
duces a significant validity coefficient, despite the fact that they have
no idea as to whether a particular subset of items is actually doing the
explanatory work.
Not long ago, many psychologists were disillusioned about person-
ality measurement (Jackson, 1970, pp. 61-62). Today, its prospects look
good; although we are by no means all the way out of the woods, there
has been some real progress. The commonsense approach has played an
important part in it and now needs to be exploited more fully.
REFERENCES
Ahlawat, K. S. (1985). On the negative valence items in self-report. Journal of Social

chiatry, 25, 107-119.
Allport, G. W., & Allport, F. H. (1928). The A-S reaction study. Boston: Houghton
Mifflin.
Allport, G. W. (1961). Pattern and Growth in Personality. New York: Holt, Rinehart and
Winston.
Allport, G. W., & Ross, J. M. (1967). Personal religious orientation and prejudice. Journal
of Personality and Social Psychology, 5, 432-443.
Allport, G. W., Vernon, P. E., & Lindzey, G. (1960). Manual for the study of values (3rd
ed.). Boston: Houghton Mifflin.
Amelang, M., & Borkenau, P. (1986). The trait concept: Current theoretical considera-
tions, empirical facts, and implications for personality inventory construction. In A.
Angleitner & J. S. Wiggins (Eds.), Personality assessment via questionnaires: Cur-
rent theory and measurement (pp. 7-31). New York: Springer-Verlag.
Anderson, M. L. (1981, August). Mental Illness and mental retardation. Legitimization
for surplus pfYJYUlations. Paper presented at the meeting of the American Psycho-
logical Association, Los Angeles.
Angleitner, A., John, 0. P., & Lohr, F. J. (1986). It's what you ask and how you ask it: An
itemmetric analysis of personality questionnaires. In A. Angleitner & J. S. Wiggins
(Eds.), Personality assessment via questionnaires: Current issues in theory and
measurement (pp. 61-107). New York: Springer-Verlag.
Ashton, S. G., & Goldberg, L. R. (1973). In response to Jackson's challenge: The compar-
ative validity of personality scales constructed by the external (empirical) strategy
and scales developed by experts, novices, and laymen. Journal of Research in Per-
sonality, 7, 1-20.
Briggs, S. R., & Cheek, J. M. (1986). The role of factor analysis in the development and
evaluation of personality scales. Journal of Personality, 54, 106-148.
Burisch, M. (1984a). Approaches to personality inventory construction: A comparison of
merits. American Psychologist, 59, 214-227.
Burisch, M. (1984b). You don't always get what you pay for: Measuring depression with
short and simple versus long and sophisticated scales. Journal of Research in Per-
sonality, 18, 81-98.
Burisch, M. (1985). I wish it were true: Confessions of a secret deductivist. Journal of
Burisch, M. (1986). Methods of personality inventory development-A comparative anal-
ysis. In A. Angleitner & J. S. Wiggins (Eds.), Personality assessment via question-
naires: Current issues in theory and measurement (pp. 109-120). New York:
Springer-Verlag.
Buss, A. 91986). Social behavior and personality. Hillsdale, NJ: Erlbaum.
Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for
cognition. Journal of Personality Assessment, 48, 306-307.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validity by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Chapman, L. J., & Chapman, J. P. (1987). The measurement of handedness. Brain and
Cognition, 6, 175-183.
Epstein, S. (1984). A procedural note on the measurement of broad dispositions. Journal
of Personality, 52, 318-325.
Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self-conscious-
ness: Assessment and theory. Consulting and Clinical Psychology, 43, 522-527.
Fletcher, G. J. 0. (1984). Psychology and common sense. American Psychologist, 39,
203-213.
Funder, D. C., & Dobroth, K. M. (1987). Differences between traits: Properties associated
with interjudge agreement. Journal of Personality and Social Psychology, 52,

409-418.
Gorsuch, R. L. (1984). Measurement: The boon and bane of investigating religion. Amer-
ican Psychologist, 39, 228--236.
Gould, S. J. (1987). This view of life: Justice Scalia's misunderstanding. Natural History,
96, 14-21.
Hampson, S. E., John, 0. P., & Goldberg, L. R. (1986). Category breadth and hierarchical
structure in personality: Studies of asymmetries in judgments of trait implications.
Hensley, W. E., & Waggenspach, B. M. (1986). A brief scale of role-playing. Journal of
Hogan, R., Carpenter, B. N., Briggs, S. R., & Hansson, R. 0. (1985). Personality assess-
ment and personnel selection. In H. J. Bernardin & D. A. Bownas (Eds.), Personal-
ity assessment in organizations (pp. 21-52). New York: Praeger.
Holden, R. R., & Fekken, G. C. (1990). Structural psychopathological test item character-
istics and validity. Psychological Assessment: A Journal of Consulting and Clinical
Holden, R. R., Fekken, G. C., & Jackson, D. N. (1985). Structured personality test item
characteristics and validity. Journal of Research in Personality, 19, 386-394.
Houts, A. C., Cook, T. D., & Shadish, W. R., Jr. (1986). The person--situation debate: A
critical multiplist perspective. Journal of Personality, 54, 52-105.
Irwin, M. E., Lundeen, E. J., & Knowles, E. S. (1988, April). Effects of retest on the
factor structure of two personality tests. Paper presented at the annual meeting of
the Southwestern Psychological Association, Tulsa.
Jackson, D. N. (1967). Manual for the Personality Research Form. Goshen, NY: Re-
search Psychologists Press.
Jackson, D. N. (1970). A sequential system for personality scale development. In C. D.
Spielberger (Ed.), Current topics in clinical and community psychology (Vol. 2,
pp. 61-96). New York: Academic Press.
Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A
lon!}itudinal study of youth. New York: Academic Press.
Knowles, E. S. (1988). Item context effects on personality scales: Measuring changes the
measure. Journal of Personality and Social Psychology, 55, 312-320.
Leary, M. R. (1983). A brief version of the Fear of Negative Evaluation Scale. Personal-
ity and Social Psychology Bulletin, 9, 371~75.
Matlin, M. (1983). Cognition. New York: Holt, Rinehart, & Winston.
McCrae, R. R., & Costa, P. T., Jr. (1987). Validation of the five-factor model of personality
across instruments and observers. Journal of Personality and Social Psychology,
52, 81-90.
Meyers, M., Malorzo, L., & Wolfe, R. (1986, April). MTMM validation of concern for
appropriateness. Paper presented at the annual meeting of the Eastern Psychologi-
cal Association, New York City.
Murray, H. A. (1938). Explorations in Personality. New York: Oxford University Press.
Paunonen, S. V. (1984). Optimizing the validity of personality assessments: The import-
ance of aggregation and item content. Journal of Research in Personality, 18,
411-431.
Paunonen, S. V., & Jackson, D. N. (1985). The validity of formal and informal personality
assessments. Journal of Research in Personality, 19, 331~42.
Perri, M., & Wolfgang, A. P. (1988). A modified measure of need for cognition. Psycholog-
ical Reports, 62, 955-957.
Rocklin, T., & Revelle, W. (1981). The measurement of extraversion: A comparison of the
Eysenck Personality Inventory and the Eysenck Personality Questionnaire. British
Juurrw,l of Social Psycholoflll, 20, 279--284.
Scheibe, K. E. (1978). The psychologist's advantage and its nullification: Limits of human
predictability. American Psychologist, 83, 869--881.
Scott, W. A., & Johnson, R. C. (1972). Comparative validities of direct and indirect
personality tests. Juurrw,l of Consulting arul Clinical Psychology, 88, 301~18.
Strube, M. J. (1986). An analysis of the Self-Handicapping Scale. Basic arul Applied
Social Psycholoflll, 7, 211-224.
Szasz, T. S. (1968). Science and public policy: The crime of involuntary mental hospital-
ization. Medical Opinion and Review, 4, 24-35.
Szasz, T. S. (1977). The manufacture of madness. New York: Harper.
Szasz, T. S. (1984). The therapeutic state: Psychiatry in the mirror of current events.
Buffalo, NY: Prometheus Books.
Tomaka, J. W., Lennox, R. D., & Wolfe, R. N. (1987, April). Social avoidance arul dis-
tress: Construct evaluation arul measurement. Poster presented at the annual meet-
ing of the Eastern Psychological Association, Arlington, VA.
Wolfe, R. N. (1986). Correlates of scores on the Morningness-Eveningness Question-
naire. Unpublished raw data.
Wolfe, R.N., & Kasmer, J. A. (1988).'JYpe vs. trait: Extraversion, impulsivity, sociability,
and preferences for cooperation and competitive activities. Juurrw,l of Personality
arul Social Psychology, 54, 864-871.
Wolfe, R. N., Lennox, R. D., Welch, L. K., & Cutler, B. L. (1984, April). A shortened
version of the Fear of Negative Evaluation Scale. Paper presented at the annual
meeting of the Eastern Psychological Association, Baltimore.
Wright, J. C., & Mischel, W. (1987). A conditional approach to dispositional constructs:
The local predictability of social behavior. Juurrw,l of Personality and Social Psy-
chology, 58, 1159--1177.
PART FIVE
EPILOGUE
CHAPTER TWENTY
An Optimistic Forecast
ROBERT HOGAN
This book celebrates 50 years of personality psychology as a formal

discipline, the birth of which was announced by the near-simultaneous
publication in 1937 of Gordon Allport's and Ross Stagner's textbooks.
Allport and Stagner thought they were witnessing (if not engendering)
the beginning of a new scientific enterprise. Fifty years later, it might
be useful to ask how well we are doing.
Although most people on the street think the subject matter of
personality psychology-the nature of human nature-is the core of
psychology, academic psychology has a more ambivalent view of the
discipline. 'l\vo examples of this ambivalence can be quickly cited. First,
personality is not a prestigious topic in academic psychology: Federal
funding remains scarce and personality psychologists are not invited to
join the National Academy of Sciences. Second, personality psycholo-
gists make up a small fraction of all psychologists and only about 10%
of Division 8 (Personality and Social Psychology) of the American Psychol-
ogical Association. Despite the popular appeal of personality psychology,
it remains something of an acquired taste within academia.
Although personality psychology is not faring as well as some
might like after 50 plus years as a discipline, it is nonetheless doing
much better than it was a few years ago. The response set controversy,
which flared up in the mid-1950s (and which its early proponents still
keep alive), was internecine warfare at its worst. The controversy ex-
ROBERT HOGAN • Department of Psychology, University of Tulsa, Tulsa, Oklahoma

74104.
New York, 1993.
293
294 ROBERT HOGAN
hausted the participants, alienated many nonparticipants, and left the

discipline open to attack by unsympathetic critics (advocates of the
response set notion argued that the key methodology of the discipline-
personality measurement-was deeply flawed and research based on
that methodology was uninterpretable).
The response set controversy was followed immediately by a mas-
sive behaviorist critique of personality psychology. This critique argued
that, not only was the measurement base confounded by artifact (i.e.,
response sets), but the principle assumption of the discipline-that there
is a stable core to personality-was empirically unjustified. This led to
the so-called "person-situation" debate, which further diverted atten-
tion from substantive issues and further eroded interest in the field.
Since the mid-1980s, however, there has been a resurgence of in-
terest in personality psychology, and I can point to eight trends that
are quite encouraging.
First, except for Jones (1990) and Ross and Nisbet (1991), the crit-
ics are now largely silent. Block (1965) and Rorer (1965) effectively
ended the response set controversy, and Kenrick and Funder (1988)
announced the end of the person-situation debate. Not only are the
critics relatively silent, but there is a new enthusiasm for personal-
ity psychology in the air, and this book is one expression of that
enthusiasm.
Second, Allport was consistently critical of nomothetic sciences as
a way to study individuals. He would, therefore, be delighted to note
that the interpretive model (in which individuals are compared with
themselves rather than with group norms) has achieved considerable
legitimacy in modern personality psychology, as seen in the writings of
Bruner (1986), McAdams (1985), Sarbin (1986), and Runyan (1990).
Third, Allport and Odbert's (1936) psycholexical study of English
language trait words provided the textual basis for the development of
the Five Factor Model (FFM) (Digman, 1989). The FFM is a well-
replicated taxonomy of personality descriptors and provides a very useful
springboard for future research as well as a methodology for classify-
ing and understanding the noisome myriad of idiosyncratic personality
measures that are introduced into the literature each year. The FFM is
a major contribution to the scientific study of personality.
Fourth, for reasons that may be more personal than intellectual,
experimental social psychology has been hostile to the notion of individ-
ual differences for some years. All that now seems to have changed;
a generation of younger social psychologists has discovered the reality
of individual differences and has begun to incorporate individual dif-
ferences measures and concepts systematically into their writing
AN OPTIMISTIC FORECAST 295
(Baumeister & Tice, 1988; Cantor & Kihlstrom, 1987; DePaulo et al.,
1987; Funder, 1991; Ickes, 1982; Kenrick, 1989; Snyder, 1981).
Fifth, it is encouraging to note the emergence of a group of young
personality psychologists in the United States who are seeking to solve
some of the fundamental problems identified in the original Allport and
Stagner texts (e.g., Buss, 1991; Cheek, 1982; John, 1990; McCrae, 1990;
Moskowitz, 1986).
Sixth, Allport's prescriptions for personality research provide the
framework for ambitious programs of ongoing work in England (e.g.,
Emler, 1990; Fumham, 1992) and in Germany (e.g., Angleitner et al.,
1990), where Allport traveled in the 1920s and first formulated many of
his ideas about personality.
Seventh, clinical psychology was once the traditional ally of per-
sonality psychology, but clinical psychology seems to have gone
through a paradigm shift; psychodynamics has been replaced by cogni-
tive psychology. Along with this shift, many clinicians seem to have lost
interest in personality and personality assessment as traditionally un-
derstood. The loss of this important market for personality research
has happily been offset by a burgeoning new market in industrial
and organizational (110) psychology. Over the past six or seven years,
1/0 psychology has discovered personality and personality measure-
ment (e.g., Barrick & Mount, 1991; Guion, 1987; Hough et al., 1990;
Staw & Ross, 1985; Weiss & Adler, 1984). This development is consis-
tent with Stagner's early interest in using personality to understand
labor-management problems and other issues in industrial psychology.
Eighth, research on the biological origins of personality is attract-
ing widespread interest in the scientific community. Perhaps the best-
known example of this is the research at the University of Minnesota
that examines the personality similarities of identical twins reared
apart (Tellegen et al., 1988). This research allows us to estimate the
heritability of the components of personality with greater precision
than before. Using models taken from the study of plant genetics,
molecular biologists are beginning to study the genetic origins of per-
sonality traits in dogs (e.g., shyness, aggressiveness), and it is only a
matter of time before we will be able to use these methods to study the
origins of individual differences in human personality.
Finally, well-constructed personality measures often predict occu-
pational performance about as well as cognitive tests do, but they tend
not to have adverse impact (i.e., minority groups' scores are usually
about the same as those of majority group members). Thoughtful ob-
servers of the U. S. Department of Justice note a trend in federal
legislation that will soon compel employers to focus more attention on
296 ROBERT HOGAN
their prospective employees' personality patterns and perhaps increase

the use of personality measures in preemployment testing. This trend
should further enhance popular interest in personality and lead to a
broader acceptance of personality assessment.
REFERENCES
Allport, G. W., & Odbert, H. S. (1936). 'lrait-names: A psycholexical study. Psychological
Monographs, 47, (211).
Angleitner, A., Ostendorf, F., & John, 0. P. (1990). Towards a taxonomy of personality
descriptors in German: A psycholexical study. European Journal of Personality,
4, 89-118.
Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job
performance. Personnel Psychology, 44, 1-25.
Baumeister, R. F., & Tice, D. (1988). Metatraits. Journal of Personality, 56, 571-598.
Block, J. (1965). The challenge of response sets. New York: Appleton-Century-Crofts.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University
Press.
Buss, D. (1991). Evolutionary personality psychology. Annual Review of Psychology, 89,
551-556.
Cheek, J. M. (1982). Aggregation, moderator variables, and the validity of personality
tests. Journal of Personality and Social Psychology, 48, 1254-1269.
of person perception. Journal of Personality and Social Psychology, 52, 303-315.
Digman, J. M. (1989). Five robust trait dimensions: Development, stability, and utility.
Emler, N. P. (1990). A social psychology of reputation. European Review of Social Psy-
chology, 1, 171-191.
Funder, D. C. (1991). Global traits: A neo-Allportian approach to personality. Psycholog-
ical Science, 2, 31-39.
Furnham, A. (1992). Personality at Work. London: Routledge.
Guion, R. (1987). Changing views for personnel selection research. Personnel Psychol-
ogy, 40, 199-213.
Hough, L. M., Eaton, N. K., Dunnette, M. D., Kamp, J. D., & McCloy, R. A. (1990).
Criterion-related validities of personality constructs and the effect of response dis-
tortion on those validities. Journal of Applied Psychology, 75, 581-595.
Ickes, W. (1982). A basic paradigm for studying personality, roles, and social behavior. In
W. Ickes & E. S. Knowles (Eds.), Personality, roles, and social behavior (pp. 305-
341). New York: Springer-Verlag.
John, 0. P. (1990). The "Big Five" factor taxonomy. In L. A. Pervin (Ed.), Handbook of
personality (pp. 66-100). New York: Springer-Verlag.
Jones, E. E. (1990). Interpersonal perception. New York: Freeman.
Kenrick, D. T. (1989). A biosocial perspective on mates and traits: Reuniting personality
and social psychology. In D. Buss & N. Cantor (Eds.), Personality psychology
(pp. 308-319). New York: Springer-Verlag.
AN OPTIMISTIC FORECAST 297
Kenrick, D. T., & Funder, D. C. (1988). Profiting from controversy: Lessons from the
person-situation debate. American Psychologist, 49, ~.
McAdams, D.P. (1985). Biography, narrative, and lives. Journal of Personality, 56, 1-18.
McCrae, R. R. (1990). Traits and trait names. European Journal of Personality, 4,
119-129.
Moskowitz, D. S. (1986). Comparison of self-reports, reports by knowledgeable observers,
and behavioral observation data. Journal of Personality, 54, 294--317.
Rorer, L. G. (1965). The great response style myth. Psychological Bulletin, 69, 129-156.
Ross, L., & Nisbet, R. E. (1991). The person and the situation. New York: McGraw-Hill.
Runyan, W. M. (1990). Individual lives and the structure of personality psychology. In A.
I. Rabin, R. A. Zucker, R. A. Emmons, & S. Frank (Eds.), Studying persons and
lives. New York: Springer-Verlag.
Sarbin, T. R. (1986). Narrative psychology. New York: Praeger.
Snyder, M. (1981). On the influence of individuals on situations. In N. Cantor & J. F.
Kihlstrom (Eds.), Personality, cognition, and social interaction (pp. 309-329).
Staw, B. M., & Ross, J. (1985). Stability in the midst of change: A dispositional approach
to job attitudes. Journal of Applied Psychology, 70, 469-480.
Tellegen, A., Lykken, D. T., Wilcox, K. J., Rich, S., & Segal, N. L. (1988). Personality
similarity in twins reared together and apart. Journal of Personality and Social
Psychology, 54, 1031-1039.
Weiss, H. M., & Adler, S. (1984). Personality and organizational behavior. In B. M. Staw
& L. L. Cummings (Eds.), Research in organizational behavior (Vol. 6, pp. 1-50).
Greenwich, CT: JAI Press.
Author Index
Abramson, L.Y. 180, 185 Amelang, M. 244, 246-248, Bandura, A. 112, 141, 143
Adelson, J. 187, 193 274, 285, 288 Barclay, A.M. 101
Adler, S. 295, 297 Anderson, J.W. 54 Bargh, J.A. 257, 264, 266
Adorno, T.W. 30, 37 Anderson, M.L. 270, 288 Barlow, D.H. 124, 128
Ahlawat, K.S 277, 287 Angel, E. 101 Barrick, M.R. 223, 233,
Ainlay, S.C. 140, 143 Angleitner, A. 215, 218, 295, 296
Albright, J.S. 186, 204 232, 233, 248, 276, Barzun,J. 110,115
Aldrich, D.B. 154 277, 278, 279, 288, Bates, E. 235
295, 296 Baumeister, R.F. 15, 17,
Alexander, I.E. 15, 17,
Anonymous 17, 19, 147, 180,181,186,295,296
126, 127
162 Baumgarten, F. 216, 233
Alker, H.A. 87, 100
Argyle, M. 184, 185, 201,
Allen, A. 238, 240-248 Beers, C.W. 168, 174
202, 204
Allen, B.P. 218, 232 Bellezza, A.G. 186
Antonovsky, A. 142, 143
Alloy, L.B. 180, 185 Bern, D.J. 87, 92, 100, 137,
Armstrong, D.M. 263, 265
Allport, F.H. 26, 49-51, 54, 143, 238, 240-248
Aronoff, J. 101, 115, 193,
188,189,193,271,288 Berkowitz, L. 186, 204, 214
194
Allport, G.W. 3--19, 23--37, Berman, H.J. 173, 174
Aronson, E. 214, 267
39-55, 57--65, 69--73, Bernieri, F. 249
Asch, S.E. 75, 83, 209,
75-82,83,86-87,92, 213,262,265 Bernstein, M. 268
100, 112-115, 119--121, Ashton, S.G. 271, 288 Bertocci, P.A. 59--60, 63, 64
124, 127, 131-143, Atkinson, J.W. 190, 194 Betts, G.H. 9, 19
147-162, 165-174, Atkinson, R.C. 214 Bhaskar, R. 253, 265
177-185, 187-190, 193, Atwood, G. 54, 125, 128 Birch, D. 190, 194
197-204,207,208,213, Blalock, H.M. 266
215-219,225,227-232, Baer, R. 180, 186 Blarney, K. 145
237-242,245,247,252, Bagb~ E.9, 19,25,37 Blank, A.L. 204
253, 263, 265, 271, Bailor, E.M. 32 Blankstein, K.R. 258, 266
286,287,288,293,296 Baldwin, A.L. 148, 151, Block, J. 80, 83, 124, 128,
Allport, R. 45, 54 162, 166, 174 213, 221, 223, 227,
Alpert-Gillis, L.J. 112-115 Baltes, P. 144 229, 233, 264, 265,
Alston, W.P. 193 Banaji, M.R. 186 294,297
299
300 AUTHOR INDEX
Block, J.H. 227, 233 Cantor, N. 15, 16, 80, 83, Crapanzano, V. 135, 144,
Booher, E. 30-31 186, 191, 192, 194, 173, 174
Borgatta, E.F. 81, 83 214,233-235,264,268, Cronbach, L.J. 210, 212,
Boring, E.G. 32-33 295-297 214
Borkenau, P. 16, 18, 244, Cantril, H. 60, 64 Cross, C. 144
246-248,274,285,288 Carlson, R. 82, 83, 124, Cummings, L.L. 297
Bowers, K.S. 87, 100, 237, 125, 128, 208, 214 Cutler, B.L. 290
248 Carpenter, B.N. 289
Bragman, L.J. 168, 174 Carver, M.S. 189, 194 Dailey, C.A. 124, 128
Cason, H. 28 D'Andrade, R.G. 262, 267
Brehm, J.W. 180, 186
Cather, W. 169, 174 Davidson, P.O. 124, 128
Briggs, S.R. 15, 18, 19,
Cattell, J. McK., 26, 37 Davis, J.H. 267
194, 223, 227, 230,
Cattell, R.B. 220, 224, 226, Davis, T. 204
231, 233, 235, 289,
233 Deaux, K. 167, 170, 174
275, 279, 287, 288,
Champagne, B. 77, 83 DeBoy, T. 249
289, 214
Chaplin, W.F. 218, 233, DeCharms, R.C. 90, 100
Brim, O.G., Jr. 144
244,245,247,248 Deci, E.L. 90, 100
Brookings, J.B. 258, 265
Chapman, J.P. 285,288 Deichmann, A.K. 268
Brown, C.E. 258,265 Chapman, L.J. 285, 288
Brown, J. 180, 186, 256, Dennett, D.C. 260, 266
Chassan, J.B. 124, 128 DePaulo, B.M. 16, 17,
268 Cheek, J.M. 205, 235, 244, 199-201,204,211,214,
Brown, P. 109, 115 246, 248, 275, 279, 295, 296
Brown, W. 9, 19 287,288,295,296 DeSoto, C.B. 262, 266
Bruner, J. 166, 174, 175, Chew, B.R. 186
Devine, P. 258, 266
294,297 Christal, R.C. 220, 235
DeWaele, J.P. 172, 174
Buehler, C. 166, 169, 175 Christie, R. 167
Dewey, J. 62, 168, 175
Bugental, D.E. 203, 204 Chiriboga, D. 137, 143
Diener, E. 187, 188, 194
Bull, V. 266 Churchland, P.M. 260, 261,
Dienstbier, R.A. 128
Bunch, M. 28 265,266
Digman, J.M. 216, 220,
Burisch, M. 282, 283, 286, Cohler, B.J. 15-17, 162
221, 224, 228, 233,
288 Cole, J.K. 205
294, 296
Burke, P.A. 246, 248 Collings, W.A. 233
Dobroth, K.M. 210, 214,
Burt, G. 244, 248 Colvin, C.R. 210, 213, 214
277, 285, 289
Burton, A. 128 Conant, J.B. 89
Dollard, J. 73, 83, 139,
Connell, J.P. 112-115
Buss, A.H. 83, 89, 276, 145, 166, 174
Conway, B.E. 268
286,288 Donley, R.W. 169, 174
Cook, M. 202, 204
Buss, D.M. 77, 83, 188, Drought, N. 34, 38
Cook, T.D. 289
191, 193, 194, 214, Dunnette, M.D. 296
Cooper, J. 186
225,230,233-235,264, Dworkin, R.H. 248
Costa, P.T., Jr. 47, 55, 76,
265, 295, 296
83, 216, 222-224, 227,
Buss, F.T. 28 Eaton, N.K. 296
231, 235, 252, 267,
284, 289 Eber, H. W. 233
Cabot, R. 4 Costello, C.G. 124, 128 Einhorn, H.J. 257-266
Cacioppo, J.T. 283, 288 Cox, J .R. 258, 266 Eissler, K.R. 54
Calkins, M. 60 Craik, K.H. 3, 19, 20, 36, Ekehammar, B. 87, 93, 100
Cameron, N. 28 37, 77, 81, 83, 131, Ekman, P. 202-205
Campbell, A. 184, 186 144, 191, 193, 194, Elde~ G.,J~ 140,144
Campbell, D.T. 271, 276, 208, 225, 230, 232, Ellenberger, H.F. 101
288 233,263,264,265 Ellsworth, P.C. 202, 205
AUTHOR INDEX 301
Elms, A.C. 15, 16, 44, 55, Funder, D.C. 15, 17, 204, Harlow, H. 28
112, 156, 158, 162 205,209-214,262,266, Harre, R. 172, 174
Emler, N.P. 295,296 267, 277,285, 289, Harris, R.E. 128
Emmons, R.A. 15, 17, 74, 294-297 Harrison, A.A. 167, 175
83, 140, 144, 187, 188, Furnham, A. 74, 84,260, Hartshorne, H. 18, 19,
190,192-194,229,233, 266,234,295,296 237,238,249
297 Hastie, 262, 266
Endler, N.S. 93, 100, 233 Gage, N.L. 210, 214 Healy, J.M. 173, 175
Epstein, S. 87, 100, 141, Garcia, T. 249 Heelas, P. 19
144, 238, 248, 274, Gergen, K. 142, 144
Heider, F. 74, 83
283,288 Gergen, M. 142, 144
Helson, R. 15, 16
Ergas, Y. 174 Gerow, J.R. 34,37 Henderson, M. 260, 266
Erikson, E.H. 51-52, 55, Gianetto, R.M. 204 Henmon, V.A.C. 29
112, 125, 128, 178, Gilbert, D.T. 258,266
Hensley, W.E. 283, 289
182, 183, 186 Giles, H. 205
Herdt, H. 84
Evans, R.I. 40-42, 55, 139, Ginsburg, G.P. 174
Hermann, M.G. 172, 175
144, 157, 162 Goffman, E. 200, 205
Hersen, M. 124, 128
Everett, J.E. 221, 233 Goldberg, A. 145
Higgins, E.T. 196, 257,
Exline, R.V. 202, 205 Goldberg, L.R. 15, 216,
264,266
Exner, J.E. 196 218,221,224,231-234,
Hilgard, E.R. 15, 103, 115
Eysenck, H.J. 224, 226, 271, 244, 245, 247,
Hinkle, S. 186
227, 233, 234 248, 288, 289
Hinz, V.B. 258, 266
Goleman, D. 73, 83
Faber, M.D. 42, 55 Hirschberg, N. 263, 267
Gordon, R.G. 9, 19
Farr, R.M. 12, 19 Hogan,J.224,226,234
Gorsuch, R.L. 230, 235,
Fazio, R.H. 249 285,289 Hogan, R.T. 15, 18, 19,
Fehr, B.J. 202, 205 Gough, H.G. 222, 234 137, 144, 162, 194,
Fekken, G.C. 277, 289 Gould, S.J. 286,289 195, 214, 224, 226,
Feldman, R.S. 204 234,273,284,289
Greene, D. 90, 101
Feldstein, S. 205 Greenwald, A.G. 15, 16, Hogarth, R.M. 257,266
Fenigstein, A. 276, 288 54, 140, 145, 180, 181, Holden, R.R. 276, 278, 289
Fenton, M. 186 186,256,266 Holt, R.R. 132, 134, 136,
Feyerabend, P. 254, 266 Greenwood, J.D. 253,266 144
Fincham, F.D. 264, 266 Gregory, R.L. 209, 214 Hooker, C.A. 253, 267
Fiske, D.W. 87, 100, 220, Griggs, F. 258, 266 Hoover, C.W. 204, 214, 296
234,271,276,288 Guilford, J.P. 220, 234 Hough, L.M. 295, 296
Fiske, S.T. 251, 256, 266 Guion, R. 295, 296 Houts, A.C. 284, 289
Fletcher, G.J.O. 16, 18, Howard, G.S. 253, 254, 267
254-267,277,288 Ha, Y.W. 257,267 Howe, H.E. 128
Flett, G.L. 258, 266 Habermas, J. 135, 144 Hull, C. 28
Frank, S. 297 Haig, B. 254, 257, 260, 266 Hunter, J.D. 143
Franz, C. 145 Hairfield, J.G. 204 Husband, R. 28
Freeman, M. 135, 144 Hall, C.S. 103, 105, 107, Hyland, M.E. 189, 195
Frenkel-Brnnswik, E. 37, 109, 111, 115, 148, 162
166, 175 Hamilton, M.M. 266 Ickes, W. 187, 195, 296
Frese, M. 189, 195 Hampson, S.E. 231, 232, Irwin, M.E. 289
Freud, S. 92, 95, 100, 201, 234, 282, 289 Items, 170, 175
205 Hanna, R. 167, 170, 174 Izard, C.E., 202, 204, 205
Friedman, H. 201, 205 Hansson, R.O. 289
Friesen, W.V. 202-205 Hanushek, E.A. 245, 249 Jaccard, J.J. 238, 249
302 AUTHOR INDEX
Jackson, D.N. 69, 83, 244, Kihlstrom, J.F. 80, 83, 178, Leary, M.R. 283, 289, 199,
246,247,249,272, 186, 191, 192, 194, 205,206
275,279,280-282,286, 295--297 Lennox, R.D. 290
287,289 Kim, M.P. 170, 175 Leonard, W.E. 168, 169,
Jackson, H.J. 245, 249 King, L.A. 190, 194 175
Jackson, J.M. 232 Kirkendol, S.E. 201, 204 Leplin,J.267
James, W. 24, 25, 37, 62, Klages, L. 216 Lepper, M.R. 90, 101, 257,
79, 83, 88, 141, 144 Klayrnan,J. 257,267 267
Jandorf, E.M. 64 Klein, G. 140, 141, 142, 144 Levinson, D.J. 37
Jemmott, J.B. 190, 195 Klein, S.B. 186 Levy, L.H. 109-U5
Jensen, F. 5, 19 Klinger, E. 191, 195 Lewin, K. 16, 19, 25, 60,
Jessor, R. 271, 276, 289 Kluckhohn, C.M. 225, 235 64, 73,83
Jessor, S.L. 276, 289 Knowles, E.S. 274, 289, 296 Le-Xuan-Hy, G.M. 268
John, O.P. 16, 17, 19, 215, Koestner, R. 195, 249 Lewis, M. 78, 83
216,218,220-224,226, Koffka, K. 60, 64 Lief, A. 60, 64
228, 229, 231, 233, Kohlberg, L. 112 Liker, J. 144
234, 236, 276, 288, Kohnstamm, E. 235 Linder, D.E. 180, 186
289,295,296 Kohut, H. 140, 141, 142, Lindzey, G. 15, 37, 103,
Johnson, R.C. 271, 277, 290 144, 145 105, 107, 109, 115,
Johnston, R.E. 203, 205 Kolligian, J. 191, 195 271, 288, 148, 162,
Jones, D.K. 84 Korchin, S. 15 214,267
Jones, E.E. 186, 209, 214, Krahe, B. 260, 267 Lippa, R. 200, 205
294,296 Kratochwil, T.R. 124, 128 Little, B. 137, 140, 145,
Jones, R.A. 124, 128, 166, Kraut, R.E. 203, 205, 248 191, 195, 229, 235
172, 175 Krout, M. 30 Lock, A. 19
Jones, W.H. 19, 175, 205, Kruglanski, A.W. 90, 100, Loehlin, J.C. 115
194, 195, 214 204,205 Lohr, F.J. 276, 288
Jung, C.G. 95, 101 Krull, D.S. 258, 266 London, H. 196
Kuhn, T.S. 254, 267 Lott, C.L. 268
Love, L.R. 204
Kagan, J. 140, 144
Lachman, S.J. 37 Ludwig, L.M. 202, 205
Kamp, J.D. 296
Lambert, W.W. 81, 83 Lundeen, E.J. 289
Kao, C.F. 288 Lamiell, J.T. 125, 128 Lykken, D.T. 297
Karniol, R. 101 Landis, C. 203, 205 Lynch, M. 144
Karwoski, T.F. 32 Lange, C. 266
Kasmer, J.A. 275, 290 Langer, E. 180, 186, 202, MacFarlane, S. 195
Kegan, R. 140, 144 205 Machotka, P. 15
Kelly, G.A. 76, 83, 96, 101, Langston, C. 15, 16, 191, Maddi, S.R. 15--17, 47, 55,
124,128,229,235 194 92-99, 101, 108, 109,
Kemp, J. 249 Lanier, K. 204 113, 115
Kendon, A. 201, 204 Larsen, R.J. 194 Magnusson, D. 93, 100,
Kenny, D.A. 204, 210, 214, Lasch, C. 140, 145 233,243,249
296 Lashley, K. 32--33 Maher, B.A. 233
Kenrick, D.T. 191, 195, Lassiter, G.D. 204 Maher, W.B. 233
246, 249, 262, 267, Lau, R.R. 257, 267 Malorzo, L. 289
294-296, 297 LaVoie, L. 210, 214 Manicas, P. T. 253, 267
Ketron, J.L. 268 Layton, L. 145 Manosevitz, M. 115
Kiefer, H.D. 193 Lazarsfeld, P. 168, 175 Marcia, J.E. 182, 186
Kiell, N. 168, 175 Lazarus, R.S. 83 Maresca, B.C. 249
AUTHOR INDEX 303
Markus, H. 179, 186, 251, Murphy, G. 5, 19 Peterson, D.R. 73, 84

267 Murray, H.A. 16, 19, 23, Pettigrew, T.F. 9, 20, 45, 54
Maser, J.D. 235 26, 31-37, 44-45, 54, Petty, R.E. 288
Maslow, A.H. 7, 13, 29, 55, 57--U5, 76, 84, 92, Phares, E.J. 148, 163
132, 145 93, 101, 121, 124, 128, Piaget,J. 142,145
Matlin, M. 277, 289 134, 153, 163, 166, Picioccio, M. 277
May, M.A. 11, 18, 19, 96, 175, 213, 214, 225, Pliner, P. 258, 266
101,237,238,249 235, 279, 289 Popper, K. 253, 267
McAdams, D.P. 126, 128, Mussen, P. 112 Porter, W. 100
144, 190, 191, 193, Potkay, C.R. 218, 232
195, 223, 227, 229, Nagao, D.H. 267 Price, R.H. 20
235,294,297 Newman, L.S. 263,267 Prince, M. 27
McClelland, D.C. 72, 76, Neufeld, R.W. 124, 128
84, 92, 93, 101, 190, Nickles, T. 267 Rabin, A.I. 101, 115
191, 193, 195 Niedenthal, P.M. 186 Rabin, I. 297
McCloy, R.A. 296 Nisbett, R.E. 209, 214, Rabinow, P. 135, 136, 145
McCrae, R.R. 76, 83, 222, 251,256,267,294,297 Rafferty, J.E. 124, 128
223, 227, 228, 231, Norem-Hebeisen, A. 144 Rank, A.D. 172, 175
235, 252, 267, 284, Norman, W.T.217,224,235 Read, S.J. 74, 76, 84
289,216,295,297 Nurius, P.S. 179, 186 Reeder, G.D. 258, 266
McCurdy, H.C. 124, 128 Reiss, S. 90, 101
O'Brien, E.J. 238, 248
McMullen, E. 253, 267 Revelle, W. 223, 235, 280,
Ochberg, R. 145
Meehl, P.E. 210, 212, 214 290
Odbert, H.S. 17, 19, 34,
Mendelsohn, G.A. 16, 17, Rhodes, G. 266
215-219,230,241,249,
19,232,268 Ricoeur, P. 135, 145
294, 296
Mershon, B. 230, 235 Rich, S. 297
Ogilvie, D.M. 45, 54, 55
Merton, R.K. 89, 98, 101 Oliver, P.V. 204, 214, 296 Rime, B. 204
Meyer, A. 60 Olson, J.M. 249 Rippere, V. 260, 267
Meyers, M. 283,285,289 Roback, A.A. 8, 20, 23, 24,
OSS Assessment Staff 35,
Michalson, L. 78, 83 37
37
Miller, J.T. 9, 19 Ostendorf, F. 218, 232, 196 Robertson, B.A. 267
Miller, L.C. 84 O'Sullivan, M. 203, 205 Robins, R.W. 16, 17, 19
Miller, N.E. 73, 83, 139, Oxenberg, J. 268 Rocklin, T.280,290
145 Rogers, C.R. 92, 96, 101,
Mischel, W. 25, 82, 84, 87, Page, M.M. 195 142, 145
92, 101, 109, 115, 192, Paige, J. 162, 163 Rorer, L.G. 187, 195, 294,
226, 237, 243, 245, Palys, T.S. 191, 195 297
246, 249, 261, 267, Parsons, T. 57 Rosanowski, J. 266
275, 284, 286 Paunonen, S. V. 69, 83, 244, Rosenberg, M. 141, 145
Modell, J. 174 246,247,249,282,289 Rosenberg, S. 124, 128,
Montello, D.R. 195 Peabody, D. 221, 235 166, 170, 172, 175,
Morey, L.C. 42, 55 Peake, P.K. 243, 245, 246, 261,267
Morgan, C.D. 128 249 Rosenthal, R. 205, 249
Moscovici, S. 12, 19 Pelham, B.W. 258, 266 Rosenwald, G. 137, 144
Moskowitz, D.S. 112, 115, Perri, M. 283, 290 Rosenzweig, M.R. 100
295,297 Pervin, L.A. 15-20, 70, 73, Rosenzweig, S. 63, 64
Mount, M.K. 223, 233, 295, 74, 76, 77, 78, 79, 81, Ross, J.M. 271, 288, 295,
296 84, 144, 189, 194, 195, 297
Muir, M.S. 90, 100 234,248,249,296 Ross, L. 209, 214, 251,
Munitz, M.K. 193 Peterson, C. 148, 163 256,257,267,294,297
304 AUTHOR INDEX
Ross, M. 89, 101, 254,257, Shweder, R.A. 82, 84, 262, Taylor, R.B. 266
267 268 Taylor, S.E. 180, 186, 251,
Roth, J. 137, 145 Siegman, A. W. 205 256,266,268
Rothbart, M.K. 227, 232, Sills, D.L. 213 Taylor, W.S. 46
235 Simonton, D.K. 112, 115, Tellegen, A. 221, 224, 227,
Rothstein, M. 101 124, 128 235, 243, 244, 249,
Rotter, J.B. 124, 129 Singer, J.L. 191, 195 295,297
Rubin, D. 140, 145 Sirgy, M.J. 140, 145 Tetlock, P.E. 170, 172, 175,
Runyan, W.M. 19, 125, Skinner, B.F. 29, 34-35, 37 258,268
126, 128, 137, 138, Smelser, N.J. 14, 20 Thayer, W.M. 9, 20
145,294,297 Smelser, W. T. 14, 20 Thomas, W.l. 169, 175
Rushton, J.P. 244, 249 Smith, K. 186 Thompson, J.B. 145
Ryckman, R.M. 148, 163 Smith, M.B. 12, 15, 16, 20, Thorpe, L.P. 9, 20
Ryle, G. 226, 235, 263, 267 37, 54, 57, 64, 65, 166, Tice, D.M. 180, 182, 186,
175 295,296
Sabini, J. 189, 195 Snyder, M. 187, 195, 297 Tindale, J.E. 267
Saeed, L. 167, 175 Sorrentino, R.M. 195 Tomaka, J.W. 283, 290
Sanford, R.N. 15, 37 Spence, K.W. 29 Tomkins, S.S. 76, 77, 84,
Sarbin, T.R. 59, 64, 144, Squires, P.C. 168, 175 124, 128, 129
294,297 Stagner, R. 3-20, 23~, Triplett, R.G. 64, 65
Satoris, J.M. 249 69-73, 75-81, 84, 106, Trope, Y. 256, 268
Saucier, G. 232 114, 115, 133, 145, Tuma, A.H. 235
Saunders, D.R. 243, 249 188, 193, 195, 293 Tupes,E.C.220, 235
Scharfstein, B. 125, 128 Staw, B.M. 295, 297
Scheibe, K.E. 79, 84, 144, Stepansky, P. 145 Uleman, J.S. 263, 267
270,290 Stern, W. 31, 35, 139, 145
Sternberg, R.J. 260, 268 Valentine, P.F. 9, 20
Scheier, M.F. 189, 194,
Stewart, A. 137, 145, 162, Vallacher, R. 140, 146, 189,
276, 288
190, 196 195
Scherer, K.R. 199, 206
Stigler, J.W. 84 Van Heck, G. 234
Schlenker, B.R. 79, 80, 84,
Stolorow, R.D. 125, 128 Vernon, P.E. 26, 37, 271, 288
181, 199, 200, 204,
205,207 Stone, J.l. 204
Stone, P.J. 54, 163 Wachtel, P.L. 237, 249
Schwartzman, A.E. 112, Waggenspach, B.M. 283,
Stones, M.J. 244, 249
115 289
Stringfield, D.O. 246, 249
Scott, W.A. 271, 277, 290 Waller, N.G. 221, 235
Strube, M.J. 256, 268, 283,
Scroggs, J.R. 148, 163 Walsh, W.B. 14, 20
290
Sears, R.R. 169, 175 Watson, D. 249
Suedfeld, P. 172, 175
Sechrest, L. 187, 195 Webb, W. 204,214,296
Sullivan, 135, 136, 145
Secord, P.F. 253, 267 Wegner, D. 140, 146, 189,
Sushinsky, L.W. 90, 101
Segal, N.L. 297 195
Swann, W.B. 184, 186
Semin, G.R. 260, 267 Swede, S.W. 170, 175 Weinberger, J. 191, 195
Shadish, W.R., Jr. 289 Szasz, T.S. 271, 272, 290 Weiner, B. 196
Shapiro, J.J. 182, 186 Weintraub, K.J. 137, 146
Shapiro, M.B. 124, 128 Taft, R. 210, 214 Weiss, D.S. 262, 268
Shaver, P. 205 Takemoto-Chock, N .K. Weiss, H.M. 295, 297
Sheldon, W.H. 28 220,221,224,233 Welch, L.K. 290
Sherif, M. 60, 64 Tatsuoka, M.M. 233 Wheeler, L. 205
Showers, C. 264, 268 Taylor, C. 135, 145 White, R.W. 124, 129, 134,
Shuttleworth, F.K. 18, 19 Taylor, E. 54 166, 175
AUTHOR INDEX 305
Whitney, K. 232 Wishman, S. 73, 84 Zajonc, R.B. 251, 267

Widiger, T.A. 187, 195 Wolfe, R.N. 16, 18, 19, Zanna, M.P. 246, 249
Wiggins, J.S. 84, 222, 225, 232,275,283,289,290 Zawadski, B. 168, 175
235,248,263,268 Wolfgang, A.P. 283, 290 Zimmerman, W.S. 220, 234
Wilcox, K.J. 297 Wright, J.C. 226, 235, 275, Zirkel, S. 191, 194
Wilson, J.P. 193,194 284,286,290 Znaniecki, F. 169, 175
Windleband, W. 59, 136, Wrightsman, L.S. 15, 17, Zucke~ R.A. 101,115,297
146 166, 171, 175 Zuckerman, M. 244, 246,
Wink, P. 15 247,249
Winnicott, D.W. 140, 146 York, K.L. 229, 236 Zukier, H. 75, 83, 262, 265
Winter, D.G. 16, 17, 169, Young, K. 28, 58, 65 Zuroff, D.C. 70, 84, 226,
174, 190, 196 Yuille, J. 204 227,236,263,264,268
Subject Index
Abnormal psychology, 103 Attitudes, 139, 246

Absolute consistency, 243-244 Attribution,74,75,90
Accuracy project, 212-213 Attributional Complexity Scale, 258--259
Act-frequency approach, 193, 275 Attributional schemata, 258-259
Action self, 180, 182, 183-184 Authoritarian personality, 30, 36
Activation, 95 Autobiographical memory, 140
Activities, 217, 218 Autobiographies, 171, 172
Actualization, 95, 96
Addiction, 78 Base rate fallacy, 209
Affect, 76--79 Bayes's theorem, 209, 257
Aggregation principle, 238 Behavioral consistency, see Consistency
Agreeableness, 215, 222, 228, 252, 284
Behavioral psychology, 5, 25-26, 29, 85,
Agreement, 210-211 87, 132, 139, 143
American Association of Applied
Behavior prediction, 17-18, 237-248
Psychology, 33
moderator variable strategy in, 243,
American Council of Learned Societies,
244, 246--247
173
trait applicability in, 243, 244, 245-246,
"American Icarus" (Murray), 124
246
American Psychoanalytic Association, 32
trait appropriateness in, 246
American Psychological Association, 5,
trait extremity in, 244-245
23, 32, 33, 293
American Psychosomatic Society, 32 trait importance in, 244, 246
Anal personality type, 92 trait relevance in, 244-245, 246, 247
Analytic psychology, 5, 104 trait variability in, 244-245, 246, 247
Applied Character Analysis in Human Benevolent eclecticism, 99
Conservation (Miller), 9 A Bibliography of Character and Person-
Approaches to Personality: SO'YM Contem- ality (Roback), 8
purary Conceptions Used in Psychol- Big Five, 18,221-228,252,284
ogy and Psychiatry (Jensen & Big Two, 222
Murphy), 5 Biological psychology, 98
The Assessment of Lives (Dailey), 124 Biophysical traits, 217
Assisted autobiographies, 172 Birth order, 88
307
308 SU&JECT INDEX
British Journal of Educatiorwl Construct validity, 212

Psychology, 59 Content validity, 281
Control, 179-180, 184
California Adult Q-Set, 223 Convergent validity, 275, 279
Canonical theory, 112, 113 Core statements, 92, 95-96
Cardinal dispositions, 134 Cross-situational consistency, 77, 238,
Cardinal traits, 139 238-240, 243--244, 245-246, 251, 254,
Career sequences, 137 262, see also Situational variability
Case studies, 23, 81, 166, 231 Culture, 284
Causality, 255 Allport on, 58-61
Causal view of traits, 225-228 Maslow on, 7, 13
Central acts, 275 Stagner on, 10-11, 12-14
Central tendency, 278-279 Current concerns, 191, 192
Central traits, 139 Current Controversies and Issues in Per-
Child psychology, 140 sorwlity (Pervin), 79
Clean personality theory, 42-46, 49, 156
Coefficient alpha, 279, 280 Death Comes for the Archbishop
Cognitive-behavioral therapy, 77 (Cather), 169
Cognitive dissonance, 95, 96, 180, 254 Defense mechanisms, 155
Cognitive evaluation hypothesis, 90 Delay of gratification hypothesis, 90
Cognitive psychology, 86, 103, 112, 261 Delinquency, 260
Cognitive-rational models, 256 Depression, 121, 260
Cognitive-structural models, 255 Depth psychology, 147, 171
Commonsense personality measurement, Developmental psychology, 86, 89
269-287 Developmental statements, 92, 93, 96
constraints of, 284-286 Diaries, 171
items in, 275-282, 286-287 Dirt phobia, Allport's 41--42, 43, 44--45,
scales in, see Test scales 49, 51,52
veridicality in, 272-274 Discriminant validity, 275, 279, 281
Commonsense psychology, 251-265, see Dispersion, 278-279
also Commonsense personality mea- Dispositions, 12, 140
surement cardinal, 134
realist theory and, 252-255, 264 in commonsense psychology, 261,
social cognition in, 252-252, 255-259 262-264,275,285
Common traits, 242 consistency and, 238
Comparative analysis, 94-97, 99, 105 motivational, 188, 190, 191
Competing response hypothesis, 90 personal documents and, 138
Conceptual seH, 180, 181-183 stylistic, 188
Conflict, 72, 73 traits distinguished from, 134
in Letters from Jenny, 152-153 Drive theory, 139
psychosocial model of, 95, 96 Dyrwmic Theory of Persorwlity (Lewin),
Conscientiousness, 215, 222, 223, 227, 241- 16
242,243,245,252,284
Consensus information, 254 Ego,60-61, 62-63,140,183
Consistency, 95, 96, 104, 188 Ego control, 227
absolute, 243-244 Ego development, 113
cross-situational, see Cross-situational "The Ego in Contemporary Psychology''
consistency (Allport), 177
of expressive behaviors, 198-199, 200 Ego psychology, 61-62, 177-178
relative, 243-244 Emotional stability, 284
SUBJECT INDEX 309
Emphasis, 198 Functional autonomy, 11, 44, 48, 60-61,
Empiricism, 97, 252-253, 254 64, 135, 143
Endogenous attribution, 90 in Letters from Jenny, 156-157
Enduring traits, 217 Fundamental attribution error, 75, 251,
Epigenetic approach, 134-135 258
Equifinality, 72
defined, 73 Galen, 23
Equipotentiality, 72 A Gallery of Women (Dreiser), 172
defined, 73 Geisteswissenschaften, 136
Existential psychology, 94, 96, 134, 147, Generative concept of causality, 255
171, 179 Geneticism, 59-60, 62, 88
Exogenous attribution, 90 Genital personality type, 92
Expansiveness, 198 Gestalt psychology, 5, 29, 120
Experience-distant concepts, 140 "Glenn," 147, 150-153, 154, 155
Experience-near concepts, 140 true identify of, 148-150
Experience sampling, 81 Goals, 72-76, 182, 188, 189, 191
An Experimental Study of the Traits of Great Depression, 10, 13
Personality (Allport), 39 Group mind, 51
Explanatory depth, 254, 255, 257 Habits, 139
Explorations in Personality (Murray), Hamburger Depressions-Skala, 283
16, 23, 54, 121 Handbook of Personality: Theory and Re-
Expressive behaviors, 197-204 search (Borgatta & Lambert), 81
consistency of, 198-199, 200 Handbook of Social Psychology, 131
deliberate regulation of, 199-202 Handedness, 285
Expressive traits, 71 Heuristic realism, 252-253
External coherence, 254, 257 Homeostasis, 32, 60, 133
Extraversion, 198, 199, 200-201, 215, 222, Homogeneity, in personality measure-
223,227,252,260,284 ment, 279
Honesty, 238, 242
Face, 202-204
Humors, 24
Facet scales, 231
Hypotheticodeductive model, 253-254, 257
Facial Coding System (Ekman's), 202
Factor analysis, 59, 124, 222, 239, 261, 280 Id, 614>2, 63
Factorial concept, 76 Identity, 177, 178, 181-183, 184-185
Fascist attitude studies, 30, 36 defined, 181
Fertility, 254-255 Identity conflict, 183
FFM, see Five-factor model Identity crisis, 178, 182-183
Fisherian analysis, 257 Identity deficit, 183
Five-Factor Model (FFM), 18, 215-232, Idiographic approach, 35, 59, 112, 134
294 in behavior prediction, 240, 241-244
Allport's criticisms of, 225-231 in FFM, 216, 228-230, 231, 232
Big Five in, see Big Five in Letters from Jenny, 148, 153-154
prototype model of, 218-219 in motivation, 188, 189, 191, 192, 193
psycholexical study in, 216-217, 218-219 nomothetic approach vs., 136-137, 153
reducing semantic problem in, 219-221 to personal documents, 138
Folk psychology, 259-264, 265 in single-case analysis, 123, 124
Free-response descriptions, 77 Imago, 190
Free self-descriptions, 229 Implicit motives, 191
Friendliness, 241-242, 243, 244, 245 Incomplete sentences methodology, 124
Fulfillment, 95 Individuality, 6, 10-12, 70
310 SUBJECT INDEX
Individual psychology, 5 Life tasks, 191, 192-193

Individual traits, 242 Likert format, 274
Industrial psychology, 5, 36, 295 Lives in Progress (White), 124
Inference, 59, 257 The Locomotive God (Allport), 39
Inner self, 79 Locomotive God (Leonard), 168, 169
Instinct theory, 87 Lonely hearts columns, 167-168, 170
Instinct transitoriness, 88
Institute for Personality Assessment Re- Machiavellianism scale, 167
search (U.C. Berkeley), 127 Meaning characteristics, 277, 278
Intentional stance, 260 Mediation, 152-153
lnteractionism, 237 Memoirs, 171
Interconnectedness, 140 Middle-level theorizing, 89-90, 98, 113
Inter-item correlations, 279 A Mind That Found Itself (Beers), 168
Internal coherence, 254, 257 Minnesota Multiphasic Personality Inven-
Internal states, 217 tory (MMPI), 134
Interpretive perspective, 135-137, 294 MMPI, see Minnesota Multiphasic Person-
Intrapsychic conflict model, 95 ality Inventory
Introversion, 260 Moderated regression analysis, 244
Intuition, 59 Moderator variable strategy, 243, 244,
Ipsatized variance, 241, 247 246-247
"Isabel," 147, 150-153, 154, 155, 158, 160
Morphogenic approach, 12, 137, 138, 140,
true identify of, 148, 149, 150
148, 153
Item-total correlations, 279
Motivational dispositions, 188, 190, 191
"Jenny," see Letters from Jenny Motives, 75-76, 81, 187-193, 256
Journal of Abnormal and Social Psychol- Allport on, 6, 11, 17, 59-61, 70-72, 75,
ogy, 53, 134 133, 134, 187, 188, 190
Journal of Personality, 111, 125, 188 implicit, 191
Journal of Personality and Social Psy- self-attributed, 191
chology, 111 social, 190
Judgment, see Personality judgment Stagner on, 6, 13, 70-72, 75, 188
Muller-Lyer effects, 209
Knower, the, 180-181 Multidetermination, 72-73
Kuder-Richardson reliability, 279 Multitrait-multimethod analysis, 271, 272,
281, 287
Learned helplessness, 184
Lebenspsychologie (Buehler), 169
Narcissism, 141
Letters from Jenny (Allport), 17, 39, 53,
National Academy of Sciences, 293
124, 138, 147-162, 166, 170-171, see
National Institute of Mental Health, 122
also "Glenn"; "Isabel"; "Ross"
Natural Self, 180
conflict and mediation in, 152-153
normal and abnormal explored in,
The Nature of Prejudice (Allport), 131
155-156 NEO Personality Inventory, 222
therapeutic skepticism in, 154-155 Neuroticism, 222, 227, 252
The Library of Living Philosophy New Introductory Lectures (Freud), 47, 53
(Dewey), 168 Nomothetic approach, 59, 216, 294
Life chart, 137 in behavior prediction, 240, 242-243
Life histories, 62, 125, 229 in FFM, 228-230, 231, 232
Life Histories and Psychobiography idiographic approach vs., 136-137, 153
(Runyan), 125 to personal documents, 138
Life space, 62 Normative model of inference, 257
SUBJECT INDEX 311
Objective personality assessments, 270, Phallic personality type, 92

271,273 Phenomenology, 29, 31, 62, 132, 136, 140
Openness, 215,252 Phenotypes, 71, 72
Oral personality type, 92 Philosophies of Human Nature Scale, 165
Originology, 42 Phrenological theory, 9
Outer self, 79 Physical states, 217
Overjustification hypothesis, 90 Pleasure principle, 61
Overlapping traits, 230 Plutarch, 23
Overspecialization, 90-91 Ponzo effects, 209
Possible selves, 179
Partisan zealotry, 98--99
Potentiality, 179
Pattern and Growth in Personality
Prediction of behavior, see Behavior pre-
(Allport), 3, 7, 133
diction
Pejorative labeling, 270
Predictive validity, 287
Perfection, 95, 96
Prejudice, 152
Peripheral acts, 275
The Principles of Psychology (James), 141
Peripheral statements, 92-93, 96
Private self-consciousness, 276
Personal construct theory, 104
Projective tests, 156, 270, 271, see also
Personal documents, 34, 35, 137-139,
specific tests
165-174
Propriate strivings, 72, 92, 141, 143
defined, 166--167
Proprium, 139, 140-142
methodology used for, 171-173
purpose of, 166--179 Prototypical states, 218, 219
theoretical perspectives on, 170-171 Prototypical traits, 218, 219
Personality: A Psychological Interpreta- Psychoanalytic psychology, 5, 33-34, 73,
tion (Allport), ix, 3, 4, 8, 46--48, 49, 82, 114, 132
50-51, 52, 53, 54, 58, 166 Allport on, 60, 120, 133, 143, 177-178
Personality and Assessment (Mischel), Letters from Jenny and, 171
237 Murray and, 27, 31, 33, 34
Personality judgment, 17, 207-213, 262 subjectivism in, 140
accuracy project on, 212-213 Psychobiographies, 112
criteria for accuracy, 211-212 Psychodynamic model, 134
expressive behaviors and, see Expres- Psychogenic needs theory, 62, 279
sive behaviors Psycholexical study, in FFM, 216--217,
interjudge agreement in, 210-211 218-219
Personality measurement, 36, see also Psychological Foundations of Personal-
Commonsense personality measure- ity: A Guide for Students and Teach-
ment ers (Thorpe), 9
Personality Research Form, 279-280 Psychological Review, 133
Personality scales, see Test scales Psychological Types (Jung), 27
Personality textbooks, 105-107 The Psychology of Personal Constructs
Personal projects, 137-138, 140, 191, 192, (Kelly), 124
229 Psychology of Personality (Bagby), 25
Personal strivings, 189-190, 191, 192, 193, Psychology of Personality (Stagner), ix,
229 3, 106
Personal variables, 85, 87-88, 294 Psychosocial conflict model, 95, 96
Personology, 92, 93, 94, 97 Public self, 181
Personology: Method and Content in Per-
sonality Assessment and Psychobiog- Q-sort items, 228
raphy (Alexander), 126 Questionnaires, 222-223, 224, 228
312 SUBJECT INDEX
The Quest of Nellie Wise Allpart Situational variability, 85, 87-88, 104, 237,
(Allport), 158--159 238, 241-242, 243-244, 294; see also
Cross-situational consistency
Reactance, 184 Sixteen Personality Factors Question-
Reactance theory, 180 naire, 220
Realist theory, 252-255, 264 Social cognition, 74-76
Regression analysis, 244-245 in commonsense psychology, 251-252,
Regularity theory of causality, 255 255-259
Relational learning, 29 Social Cognition (Fiske & Taylor), 252
Relative consistency, 243-244 Social judgment, 254
REP test, 124, 229 Social learning theory, 104, 112
Roles, 217 Social motives, 190
Rorschach test, 36, 121, 128 Social psychology, 81, 85, 86, 89, 103,
"Ross," 147, 148, 14~150, 152-153, 154, 294-295
155, 157-161, 171 commonsense psychology in, 254, 255,
as double, 15~161 256, 25~264
Russian experiment, 14 teaching of, 98
Social Science Citation Index, 242
Scapegoating, 88 Social Science Research Council (SSRC),
Script theory, 124 167, 173
Secondary traits, 139 Social stimulus value, 24
Self, 6, 76, 141-142, 177-185, 188 Society for Personology, 125
action, 180, 182, 183-184 Society for Projective Techniques, 32
conceptual, 180, 181-183 Society for the Psychological Study of So-
inner, 79 cial Issues, 33, 36
natural, 180 Sociocultural factors, see Culture
organization of, 79-80 SSRC, see Social Science Research
outer, 79 Council
personality confused with, 178 State-state analysis, 137
possible, 179 Stasis and flow of behavior, 76, 81
public, 181 States, 217, 218, 219
unity of, 178--179 Stephenson's Q technique, 124
Self-attributed motives, 191 Structural-dynamic psychology, 147, 171
Self-concept, 178, 179 "The Study of Personality by the Intu-
defined, 181 itive Method" (Allport), 136
Self-consistency, 80 A Study of Values (Allport & Vernon),
Self-esteem, 140, 141, 179, 181, 183 135,288
Self-perception theory, 254 Stylistic dispositions, 188
Self-presentation, 167, 181, 270 Subjectivism, 140-141
Self-rated variability, 241-242, 246, 247 Suicide, 89
Semantic distortion hypothesis, 262 Summary constructs, 284
Serial program, 191 Summary view of traits, 225-228
Shame, 78--79 Superego, 63
Simple and sovereign theorizing, 86-87, 88 Surface characteristics, 276-277, 278
Simplicity, 255 Surgency, see Extraversion
Single-case analysis, 17, 11~127 Sutton-Boveri hypothesis, 264
inhibiting forces in, 121-123 Systems theory, 132-133
sustaining forces in, 123-125
"The Single Case in Clinical Psychologi- Thct, Push, and Principle (Thayer), 9
cal Research" (Shapiro), 124 TAT, see Thematic Apperception Test
SUBJECT INDEX 313
Teaching of personality, 98--99 Traits (cont.)
''Teleonomic Description in the Study of defined, 139, 239
Personality" (Allport), 188 enduring, 217
Teleonomic trend, 189 expressive, 71
Telic significance, 71 in FFM, 215, 217, 218, 219-221, 224,
Temporal stability, 281 225--228, 230-231
Temporary states, 217 individual, 242
Test scales, 275-276, 279--280, 281-284, motives and, 187, 188, 193
285, 286-287 Murray on, 134
breadth vs. length in, 282-284 overlapping, 230
Thematic Apperception Test (TAT), 35, prototypical, 218, 219
36, 121, 190 secondary, 139
Theories of Learning (Hilgard), 103 Stagner on, 70-72, 75
Theories of Personality (Hall & Lindzey),
summary view of, 225--228
103 Trait variability, 244-245, 246, 247
Thurstone scale, 26, 34
Twin studies, 295
Time-series analysis, 124
Trait applicability, 243, 244, 245--246
Trait appropriateness, 246 Unifying power, 254, 257
Trait extremity, 244-245 Uniqueness, 50-51, 58-59, 70
Trait importance, 244, 246 Unity of personality, 11, 52-53
'Prait-NarY~.es: A Psycholexical Study
University of California, Berkeley, 127
(Odbert), 34 The Use of Personal DocurY~.ents in Psy-
Trait relevance, 244-245, 246, 247 chological Science (Allport), 39,
Traits, 75--76,77,82,104,213,254,295 165--166
Allport on, 6, 11-12, 70-72, 75, 95, 133,
134, 137, 139-140, 187, 188, 263 Values, 182
behavior prediction and, see Behavior Verstehende psychology, 120
prediction Verstehen psychology, 31, 35, 136
biophysical, 217 Veterans Administration, 122
cardinal, 139
causal view of, 225-228 Woodworth Personal Data Sheet, 29
central, 139 World War I, 159-161
common, 242 World War II, 121-122, 123
in commonsense psychology, 261, 263,
264,275,284,285--286 Zeitgeist, 142

(Perspectives On Individual Differences) Kenneth H. Craik (Auth.), Kenneth H. Craik, Robert Hogan, Raymond N. Wolfe (Eds.) - Fifty Years of Personality Psychology-Springer US (1993) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Perspectives On Individual Differences) Kenneth H. Craik (Auth.), Kenneth H. Craik, Robert Hogan, Raymond N. Wolfe (Eds.) - Fifty Years of Personality Psychology-Springer US (1993) PDF

Uploaded by

Copyright:

Available Formats

Fifty Years of

Current volumes in the series

FIFTY YEARS OF PERSONALITY PSYCHOLOGY

HANDBOOK OF MULTIVARIATE EXPERIMENTAL

HISTORICAL FOUNDATIONS OF EDUCATIONAL PSYCHOLOGY

INDIVIDUAL DIFFERENCES IN CARDIOVASCULAR

LEARNING STRATEGIES AND LEARNING STYLES

THE NEUROPSYCHOLOGY OF INDIVIDUAL DIFFERENCES

PERSONALITY, SOCIAL SKILLS, AND PSYCHOPATHOLOGY

THEORETICAL FOUNDATIONS OF BEHAVIOR THERAPY

SPRINGER SCIENCE+BUSINESS MEDIA, LLC

Flfty years of personallty psychology / edlted by Kenneth H. Cralk,

10 1993 Springer Science+Business Media New York

All rights reserved

N o part of this book may be reproduced, stored in a retrieval system,

Irving E. Alexander, Department of Psychology, Duke University,

Oliver P. John, Department of Psychology, University of California,

This volume celebrates the textbooks Personality: A Psychological In-

book is organized into four sections. First, the introductory chapter

Chapter 1 The 1937 Allport and Stagner Texts in

PART II. HISTORICAL AND PERSONAL

Chapter 2 Fifty Years of the Psychology of Personality:

Chapter 3 Allport's Personality and Allport's Personality . . 39

Chapter 4 Allport and Murray on Allport's Personality:

PART III. CURRENT STATE OF PERSONALITY

Chapter 5 Pattern and Organization: Current Trends

CUITent Status and Prospects for the Future . . . . . . . . . . . . . . . . . . . 80

Chapter 6 The Continuing Relevance of Personality

Chapter 7 It's Time to Put Theories of Personality in

PART IV. PRESENT-DAY PERSPECTIVES

The Current Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

Chapter 9 Describing Lives: Gordon Allport and the

Chapter 10 Gordon Allport and "Letters from Jenny" ..... 147

Chapter 11 Allport's Personal Documents:

Chapter 12 Conceptions of Self and Identity: A Modern

Chapter 13 Current Status of the Motive Concept . . . . . . . . 187

Chapter 14 The Ability to Judge Others from Their

Chapter 15 Judgments of Personality and Personality

Chapter 16 Gordon Allport: Father and Critic of the

Chapter 17 To Predict Some of the People More of

Chapter 18 The Scientific Credibility of Commonsense

Chapter 19 A Commonsense Approach to Personality

Chapter 20 An Optimistic Forecast . . . . . . . . . . . . . . . . . . . . . . . . 293

Author Index ............................................... 299

Subject Index .............................................. 307

The 1937 Allport and Stagner

In August 1937, Gordon W. Allport and Ross Stagner completed and

KENNETH H. CRAIK • Institute of Personality and Social Research, University of Cali-

organization of this volume, it is helpful to reintroduce briefly these

ALLPORT AND STAGNER:

In the 1920s and 1930s, an increasingly vigorous, empirically ori-

ALLPORT AND STAGNER:

Although generated independently, these two pioneering textbooks

definitions of personality and confidently introduces his own, now well-

Perhaps it is indicative of the then early stage of personality

ing the new field's existence. As a second-generation personality psy-

Thayer's Tact, Push, and Principle (1885), whose organization is fore-

A MAJOR CONTRAST: AN ANALYSIS OF THE INDIVIDUAL

A basic divergence in the formulations guiding the 1937 Allport

umes themselves. In contrast, Allport's textbook is written with a gen-

ALLPORT: THE INDIVIDUAL TRANSCENDENT

Allport, while not denying their role, narrowly circumscribed his

and continued to address it, notably through his notions of the