Vous êtes sur la page 1sur 408


and the
Volumes in this series are edited by successive presidents of the American Sociological
Association and are based upon sessions at the Annual Meeting of the organization.
Volumes in this series are listed below.


The Social Fabric: Dimensions and Issues (1986)
in association with
Social Structures and Human Lives:
Social Change and the Lire Course, Volume 1 (1988)
Sociological Lives: Social Change and the Lire Course, Volume 2 (1989)
Cross-National Research in Sociology (1989)
Sociology in America (1990)
Macro-Micro Linkages in Sociology (1991)
The Social Context or AIDS (1991)
Sociology and the Public Agenda (1993)
The above volumes are available from Sage Publications.

Approaches to the Study or Social Structure (1975, out of print)
The Uses or Controversy in Sociology (1976, out of print)
Major Social Issues: A Multidisciplinary View (1978, out of print)
Societal Growth: Processes and Implications (1979, out of print)
Sociological Theory and Research: A Critical Approach (1980, out of print)
Gender and the Lire Course (1985, Aldine Publishing Co.)
The Nature or Work: Sociological Perspectives (1990, Yale University Press)
and the


hericil SlCillllilll ASSlCillili Presidulill Series

SAGE Publications
International Educational and Professional Publisher
Newbury Park London New Delhi
Copyright © 1993 by Sage Publications, Inc.

All rights reserved. No part of this book may be reproduced or utilized in

any form or by any means, electronic or mechanical, including photocopy-
ing, recording, or by any information storage and retrieval system, without
permission in writing from the publisher.

For information address:

SAGE Publications, Inc.

2455 Teller Road
Newbury Park, California 91320
SAGE Publications Ltd.
6 Bonhill Street
London EC2A 4PU
United Kingdom
SAGE Publications India Pvt. Ltd.
M-32 Market
Greater Kailash I
New Delhi 110048 India

Printed in the United States of America

Library of Congress Cataloging-in-Publication Data

Main entry under title:

Sociology and the public agenda I William Julius Wilson, editor.

p. cm.-(American Sociological Association presidential
Includes bibliographical references and index.
ISBN 0-8039-5082-9.-ISBN 0-8039-5083-7 (pbk.)
1. Social policy-Congresses. 2. Sociology-Congresses.
I. Wilson, William J., 1935- . II. Series.
HN28.S63 1993
36 1.6' I-dc20 92-39719

93 94 95 96 97 10 9 8 7 6 5 4 3 2 1

Sage Production Editor: Astrid Virding


About the Authors vii

Preface xiii

Introduction: Sociology, Social Science,

and the Public Policy Agenda
1. Can Sociology Playa Greater Role in Shaping
the National Agenda?
William Julius Wilson 3
2. The Interaction of the Sociological Agenda
and Public Policy
Carol H. Weiss 23
3. How Do Issues Get on Public Policy Agendas?
John W. Kingdon 40
4. The Powers and the Intellectuals:
Benchmark Texts and Changing Conditions
Steven Brint 51

Part I: The Politics of Citizenship

5. Migrants Into Citizens? Traditions of Nationhood and
Politics of Citizenship in France and Germany
Rogers Brubaker 73
6. Citizenship and Welfare: Social Democratic
and Liberal Perspectives
J. Donald Moon 97
7. "Social Citizenship," Work, and Social Solidarity:
Historical Comparisons Between Britain and Sweden
Roger Lawson 119

Part II: Organizations, Social Movements, and Public Policy

8. Organizations as (Secondary) Citizens
Philippe C. Schmitter 143
9. Networks as Political Glue:
Explaining Public Policy-Making
David Knoke 164
10. Financial Reorganization of American
Corporations in the 1980s
Neil F1igstein and Linda Markowitz 185
11. The Conservative Revolution That Wasn't: New Right
Populism and the Preservation of New Deal Liberalism
Anne Wortham 207

Part III: The Public Agenda: Addressing High Priority

Social Problems
12. How Families Manage Risk and Opportunity
in Dangerous Neighborhoods
Frank F. Furstenberg, Jr. 231
13. The Community Context of Violent Crime
Robert J. Sampson 259
14. The Politics of Homelessness
Peter H. Rossi 287
15. Inner-City Education: A Theoretical
and Intervention Model
James P. Comer 300
16. Mothers, Children, and Low-Wage Work:
The Ability to Earn a Family Wage
Roberta M. Spalter-Roth, Heidi I. Hartmann,
and Linda M. Andrews 316

Part IV: Issues for the Public Agenda

17. The French Child Welfare System: An Excellent
System We Could Adapt and Afford
Barbara R. Bergmann 341
18. Employment as a Human Right
Philip Harvey 351
Index 375
About the Authors

LINDA M. ANDREWS is currently a research associate with the Institute

for Women's Policy Research. She received an M.S. in statistics and a B.S.
in mathematics and economics, both from the University of South Caro-
lina. Since joining the Institute for Women's Policy Research in 1988, she
has contributed to several research projects by performing computer pro-
gramming and statistical analysis. In this capacity, she has acquired
substantial experience working with census microdata files, especially the
Survey of Income and Program Participation (SIPP). She is currently
working on a study of the impact of pay equity on state civil service
workers and a study on the work and welfare patterns of AFDC recipients.
Before coming to IWPR, she worked for 3 years in the actuarial field
monitoring premium rates for group health insurance and Medicare sup-
plement policies.

BARBARA R. BERGMANN, currently Distinguished Professor of Eco-

nomics at American University, Washington, D.C., has a doctorate from
Harvard University. Her research interests include social policy, sex roles
in the economy, and computer simulation methods. She has previously
served on the staff of the Bureau of Labor Statistics, the Council of
Economic Advisors, and the U.S. Agency for International Development.
Her most recent book, The Economic Emergence of Women (1986), ex-
plored the reasons for changes in women's economic role and outlined
needed changes in the workplace, the marketplace, government policy, and
the family. She is also coauthor (with Robert L. Bennett) of A Microsimulated
Transactions Model of the United States Economy (1986).

STEVEN BRINT is Professor of Sociology at University of California,

Riverside. He is the author of The Diverted Dream: Community Colleges
and the Promise ofEducational Opportunity in America, 1900-1980 (with
Jerome Karabel), and of the forthcoming Retainers, Merchants, and Priests:
Politics and Culture in the Professional Middle Class in America (Princeton
University Press).

ROGERS BRUBAKER is Associate Professor of Sociology at the Uni-

versity of California, Los Angeles. He is author, most recently, of Citizen-
ship and Nationhood in France and Germany (1992). He is currently
working on the national question in post-Soviet Eurasia.


JAMES P. COMER, M.D., is the Maurice Falk Professor of Child Psychiatry

at the Yale Child Study Center, Associate Dean of the Yale School of
Medicine, and Director of the School Development Program. His preven-
tive psychiatry work in schools began in 1968 and his school improvement
model is now used in school districts in 14 states. He has published more
than 90 scientific articles, more than 20 book chapters, and four books, the
latest, Maggie's American Dream: The Life and Times ofa Black Family,
based on the life of his own family and his work in schools. He is a graduate
oflndiana University, Howard University College of Medicine, University
of Michigan School of Public Health, and did his training in psychiatry at
Yale School of Medicine.

NEIL FLIGSTEIN is Professor of Sociology at the University of Califor-

nia-Berkeley. He has recently published a book, The Transformation of
Corporate Control (1990), that provides a sociological view of the emer-
gence and transformation of the large American corporation in the past
century. He is currently working on a book manuscript with Doug McAdam
that attempts to theorize the problem of strategic action. He is also doing
a study of organizational and political change in the European Community
as a result of the 1992 Single Market Program.

FRANK F. FURSTENBERG, Jr., is the Zellerbach Family Professor of

Sociology and Research Associate in the Population Studies Center at the
University of Pennsylvania. His interest in the American family began at
Columbia University where he received his Ph.D. in 1967. His most recent
books include: Adolescent Mothers in Later Life. with J. Brooks-Gunn and
S. Philip Morgan and Divided Families: What Happens to Children When
Parents Part, with Andrew Cherlin.

HEIDI I. HARTMANN is currently Director of the Washington-based

Institute for Women's Policy Research, a scientific research organization
concerned with policy issues of importance to women. She is an economist
with a B.A. from Swarthmore College and M.Phil. and Ph.D. degrees from
Yale University, all in economics. In spring 1988, she was Director of the
Women's Studies Program at Rutgers University. During 1986-1987 she held
an American Statistical Association fellowship at the Census Bureau where
she conducted research on women's poverty. For 8 years previously she was
a staff member of the National Research CounciVNational Academy of
Sciences, where she contributed to many reports on women's employment
issues, and she served as Associate Executive Director of the Commission on
Behavioral and Social Sciences and Education. Her work on feminist theory
and the political economy of gender has been widely published.
Aboutthe Authors ix

PHILIP HARVEY is a practicing attorney in New York City. He received

his law degree from Yale and Ph.D. in economics from the New School
for Social Research. He is the author of Securing the Right to Employment
(1989) and coauthor with Theodore Marmor and Jerry Mashaw ofAmerica's
Misunderstood Welfare State (1990).

JOHN W. KINGDON is Professor in the Department of Political Science

at the University of Michigan, and has been Chair of that Department. He
has written widely on American governmental institutions, and has con-
ducted major studies of legislative decision making and public policy
formation at the national level in the United States. He is a Fellow of the
American Academy of Arts and Sciences, and has been a John Simon
Guggenheim Fellow and a Fellow at the Center for Advanced Study in the
Behavioral Sciences.

DAVID KNOKE is Professor and Chair of Sociology at the University of

Minnesota. He is Chair-Elect (1991-1992) of the Organizations and Occu-
pations Section of the American Sociological Association. His most recent
books are Organizing for Collective Action (1990); Political Networks
(1990); and Basic Social Statistics (with George Bohrnstedt, 1991). Cur-
rent projects are a survey of u.S. organizations' human resources policies
(with Arne Kalleberg, Peter Marsden, and Joe Spaeth) and a comparison
of U.S., German, and Japanese labor policy domain networks (with Franz
Urban Pappi, Jeffrey Broadbent, and Yutaka Tsujinaka).

ROGER LA WSON, Senior Lecturer in Social Policy at the University of

Southampton, England, has published widely on social policy, poverty,
and inequality in the United Kingdom and other Western European coun-
tries. His books include Responses to Poverty: Lessons From Europe (with
R. Walker and P. Townsend) and Poverty and Inequality in Common
Market Countries, which he edited with Vic George. He is currently
completing a book on the development of the welfare states in Britain,
Germany, and Sweden.

LINDA MARKOWITZ is a graduate student in the Department of Soci-

ology at the University of Arizona. Her current research interests are in
the areas of gender and stratification. She is writing a dissertation about
the intended and unintended actions of labor unions that have helped
create occupational sex segregation.

J. DONALD MOON received his Ph.D. from the University of Minnesota

and now teaches political science at Wesleyan University. He is the author

of the chapter on the "Logic of Political Inquiry" in the Handbook of

Political Science, and has published a number of articles on the philosophy
of social inquiry. He is coeditor of Dissent and Affirmation and edited
Responsibility, Rights, and Welfare (1988), and has written a number of
articles on the political theory of the welfare state. He has just completed
a book on the foundations of liberal theory, entitled Thin Selves, Rich
Lives, Tragic Conflicts.

PETER H. ROSSI is Stuart A. Rice Professor Emeritus of Sociology and

Director Emeritus, Social and Demographic Research Institute, at the
University of Massachusetts at Amherst. He is past-president of the Amer-
ican Sociological Association and was the 1985 recipient of the Common-
wealth Award for contributions to sociology. His recent books include
Down and Out in America (1989), Thinking About Evaluation (with R. A.
Berk, 1990), and Evaluation: A Systematic Approach (with the late How-
ard E. Freeman, 1993). He is a Fellow of the American Academy of Arts
and Sciences and of the American Association for the Advancement of

ROBERT J. SAMPSON is Professor of Sociology at the University of

Chicago. His major research interests include the social sources of violence,
multilevel theories of social organization, crime and delinquency over the life
course, victimization, escape from poverty, and the history of criminological
theory. Articles on these topics have appeared or are forthcoming in the
American Sociological Review, Criminology, American Journal ofSociology,
and Social Forces. He has also recently completed a paper on the community
context of violent crime for the National Academy of Sciences, Panel on the
Understanding and Control of Violent Behavior.

PHILIPPE C. SCHMITTER is Professor of Political Science at Stanford

University, formerly having taught at the European University Institute
and at the University of Chicago. His current research interests are in the
consolidation of democracy in Southern and Eastern Europe and Latin
America, and he is finishing a book on the emerging European polity.

ROBERTA M. SPALTER-ROTH, Deputy Director for Research of the

Institute for Women's Policy Research, is a sociologist with a Ph.D. in
sociology from American University and a B.A. degree from Indiana
University. She joined IWPR at its inception in 1987. As Director for
Research and Director for Research on Work and Family Policies, she has
overall responsibility for the coordination of IWPR's research program
and directs many specific projects. Prior to joining IWPR, she conducted
About the Authors xi

research for more than 10 years for the Anti-Poverty Program and the
Greater Washington Research Center. She is also an Assistant Professor
of Sociology at the American University.

CAROL H. WEISS is a professor at Harvard University in the Graduate

School of Education. She holds a doctorate in sociology from Columbia
University. She is the author or editor of eight books, including Evaluation
Research: Methods of Assessing Program Effectiveness, Social Science
Research and Decision Making, Reporting of Social Science in the Na-
tional Media (with E. Singer), and Organizations for Policy Analysis:
Helping Government Think. Much of her career has been devoted to
studying the linkage between social science and policy-making. On the
policy side, she has been an ASA Congressional Fellow, a guest scholar
at the Brookings Institution, a senior fellow at the U.S. Department of
Education, and a Visiting Scholar at the U.S. General Accounting Office,
as well as consultant to a score of government agencies.

WILLIAM JULIUS WILSON is Lucy Flower University Professor of

Sociology and Public Policy and Director of the Center for the Study of
Urban Inequality at the University of Chicago. He is the past President of
the American Sociological Association and is a MacArthur Prize Fellow.
He has been elected to the National Academy of Sciences, the American
Academy of Arts and Sciences, and the American Philosophical Society.
His most recent books include The Truly Disadvantaged: The Inner City,
the Underclass, and Public Policy and A Broader Vision: Race, Class, and
Poverty in Urban America (forthcoming).

ANNE WORTHAM, Associate Professor of Sociology at Illinois State

University and continuing Visiting Scholar at Stanford University's Hoo-
ver Institution, has taught at Wellesley College, Harvard University's
Kennedy School of Government, and Washington and Lee University. She
holds a B.S. degree from Tuskegee University and a Ph.D. from Boston
College, and is author of The Other Side ofRacism: A Philosophical Study
of Black Race Consciousness (1981). She is currently writing a book on
theories of social and cultural marginality, and conducting a study of
intellectuals on the right.

Five decades ago, in 1939, Robert Lynd wrote that

therewould be nosocialsciencesif therewerenotperplexitiesin livingin culture

that call for solution. And it is precisely the role of the social sciences to be
troublesome, to disconcert the habitual arrangements by which we manage to
live along, and to demonstrate the possibility of change in more adequate
directions. (p. 181)

Lynd strongly emphasized that social science is an organized segment

of culture that exists to help society in continually comprehending and
reconstructing its culture. This is achieved, argued Lynd, through the
unique role of asking long-range and, if necessary, sharply irreverent
questions of our democratic institutions; and of doggedly pursuing these
questions with systematic research. As Lynd put it, "if social science is to
be free to be science, it must have the courage to fight for its freedom from
the dragging undertow of a culture preoccupied with short-run statements
of long-run problems" (1939, p. 203).
The 1990 Program Committee of the American Sociological Associa-
tion developed the theme, "Sociology and the Public Agenda" to encour-
age discussion of the issues raised by Robert Lynd in the thematic sessions
of the 1990 Annual Meeting of the Association in Washington, D.C. Our
aim was to explore the problem of both protecting the tradition of free
intellectual inquiring and of promoting the political and social responsi-
bility of social science.
In other words, the Program Committee wanted sociologists and the
other social scientists to reflect upon the factors in the larger culture that
bear on sociology as a discipline (what topics get researched, how funds
are allocated, how and what data are collected, how these data are analyzed
and by whom, etc.); upon how sociologists as a collectivity deal with these
factors, including the protection of scientific autonomy; and upon the
political and civic responsibility of sociological research vis-a-vis the
larger society, including ways that sociology can contribute meaningfully
and responsibly to public policy.
Except for my introductory essay-"Can Sociology Playa Greater Role
in Shaping the National Agenda?"-the chapters included in this volume
were selected among those presented in the thematic sessions of the 1990
meeting. They have been organized around five topics including Sociology,
Social Science, and the Public Policy Agenda; The Politics of Citizenship;


Organizations, Social Movements, and Public Policy; The Public Agenda:

Addressing High Priority Social Problems; and Issues for the Public
Agenda. These topics represent the range of issues explored in the thematic
papers presented at the Annual Meeting in Washington. My introductory
essay incorporates arguments presented i r each of the chapters. And I
relate many of these arguments to provocative questions about the rela-
tionship between the future of sociology as a discipline and its role in
shaping the national agenda.
I would like to give a special thanks to Kathryn Neckerman, Judy
Birgen, and Edward Walker for their assistance in preparing this volume
for publication.

-William Julius Wilson


Sociology, Social Science,

and the Public Policy Agenda
Can Sociology Playa Greater Role
in Shaping the National Agenda?

William Julius Wilson

In the past several years a number of articles in the print media have
speculated about the demise of sociology as an academic discipline. These
reports have been fueled by the closing of the departments of sociology at the
University of Rochester and at Washington University in St. Louis; by the
recent decision to reduce the size of the Department of Sociology at Yale by
40%, following threats to abolish the Department altogether; and by the sharp
drop in the number of students in the United States receiving bachelor's
degrees in sociology-from 35,996 in 1973 to only 14,393 in 1989.
However, arguments concerning the fall of sociology are overstated.
Enrollment has slightly increased in the approximately 2,000 sociology
departments around the country in the past 2 years, and the quality of
applicants to graduate departments of sociology has noticeably improved
since 1980. Nonetheless, the recent negative actions taken against some
of the most visible departments of sociology in the country suggest that
the discipline remains vulnerable. Accordingly, I believe that unless soci-
ology can position itself to have a real influence on shaping the national
agenda in the ensuing years, questions about its continued viability as an
academic discipline will continue to be raised.
A discussion of the theme of sociology and the public agenda could not
be more timely. We have entered a period of immense national and
international change and turmoil. Social problems abound. Our nation is
saddled with a 4 trillion dollar debt ana 400 billion dollar annual deficits.
Our cities, stripped of needed revenue from depleted federal sources,
struggle to maintain an adequate quality of life. Problems associated with
the growing concentration of ghetto poverty are spilling over into the


larger urban environment, and the riot in Los Angeles following the verdict
on the Rodney King incident is a dramatic reminder. The poor and the
working classes struggle to make ends meet, and even the middle class has
experienced a decline in its living standard. And, as discussed in Peter H.
Rossi's perceptive Chapter 14 in this volume, perhaps no problem more
dramatically symbolizes the depths of America's predicament than the rise
of homeless individuals and families and their increased visibility in urban
areas in the 1980s.
The hard economic times have aggravated racial relations, as evidenced
in a recent spate of racially motivated violent incidents in a number of
cities, and in increases in racial tensions in schools and on college cam-
puses. Similar problems have surfaced in other countries. Stagnant econ-
omies and slack labor markets in Europe have placed strains on the welfare
state at the very time when the immigrant population has become more
dependent on public assistance for survival. In his comparison of the
problems of immigration and citizenship in France and Germany in Chap-
ter 5, Rogers Brubaker states:

During the 1950s and early 1960s, most foreign workers were either single or
separated from their families. Many lived in isolated workers' hostels. Outside
the workplace, they were largely invisible, participating little in the social,
cultural, or political life of the host society. In the last two decades, however,
the sojourners have become settlers. Single workers were joined by their fami-
lies, or formed new families. Immigrants became neighbors, schoolmates, and
joint users of public spaces. An increasingly vocal second generation emerged,
tenuously rooted in the culture of the parents' generation, yet economically and
socially marginalized in the country of residence. Groups marked by dress,
language, religion, and custom as "culturally distinct" comprised the fastest
growing segment of the immigrant community. Immigrants in both countries
have clustered in particular regions and, within cities, in particular neighbor-
hoods. All these developments made immigrants more visible.

As economic conditions have worsened, many in the majority white

population view the growth of minorities and immigrants as part of the
problem, thereby increasing racist and anti-immigrant feelings.
These social problems are discussed and debated in public forums.
Representing prominent issues on the public agenda, they are clearly
within the purview of sociology. Nonetheless, I maintain that, in the
United States at least, sociology is not yet in a position to become a major
player in the national public policy forum because of its present orientation
toward, approach to, and handling of research and scholarship relevant to
issues of public concern. Accordingly, I believe that unless certain funda-
mental changes are made, sociology's influence on the future national
William Julius Wilson 5

agenda will be minimal, and it will be more difficult to erase or overcome

the image of a fading academic discipline. Let me discuss the kinds of
changes I have in mind.



In a recent issue of Newsweek, James S. Coleman was quoted as saying

that it is "extremely important for sociology to demonstrate its utility to
society if it's going to be viable in the long run" (Kantrowitz 1992, p. 55).
I very strongly agree. However, there remains within the discipline a
strong sentiment against the practical application or utility of sociology.
On the one hand, there are the advocates of "pure" social science who
argue that it is not the business of sociologists to point out what ought to
be done to address a social problem. An effective rejoinder to this argu-
ment was provided more than 40 years ago by Robert Lynd (1939) in his
classic book, Knowledge For What? Lynd stated that:

Either the social sciences know more than do the "hardheaded" businessman, the
"practical" politician and administrator, and the other defacto leaders of culture
as to what the findings of research mean, as to the options the institutional system
presents, as to what human personalities want, why they want them, and how
desirable changes can be effected, or the vast current industry of social science
is an empty facade. (p. 18)

Indeed, to dismiss the idea that sociologists can address the issue of
"what ought to be done" is to blur the distinction between what the
philosopher of science Carl G. Hempel has described as instrumental
versus categorical value judgments. An instrumental value judgment is "a
statement which expresses a universal or probabilistic kind of means-ends
relationship, and which contains no terms of moral discourse ... at all"
(Hempel 1965, p. 85). For example, if research has firmly established that
the least prejudiced individuals are those who have grown up in a permis-
sive rather than a restrictive environment, then an instrumental value
judgment would state that if our society wants to embark on a long-term
program to reduce prejudice, then it is better to raise children in a permis-
sive than in a restrictive way. Such a statement clearly represents an
empirical assertion amenable to scientific test.
However, to assert that it is right to have the goal of reducing prejudice
in society represents a categorical, or absolute, judgment of value. To state
that it is right to reduce prejudice is to express a norm for behavior or a

standard for moral valuation. Since the statement does not express an
assertion that can be empirically tested or since it does not purport to
describe a fact that is directly observable, it falls outside the purview of
science. Accordingly, although it would be inappropriate for sociologists,
in their role as social scientists, to make categorical value judgments, they
can certainly make instrumental judgments of value, judgments that would
be consistent with the assumptions of the logic of scientific inquiry
(Hempel 1965).
However, this is not to say that sociologists or social scientists cannot or
should not address the issues of categorical value judgments in their studies
of basic values and belief systems in society. Often a social scientist can point
out the contradictions or inconsistencies in what are taken to be categorical
judgments of values. For example, in Chapter 18 Philip Harvey argues
persuasively that employment is rarely considered or discussed as a human
right in the United States because it is not defined as such by our Constitution.
It therefore is not included among the "rights that even democratically elected
majorities may not sacrifice for the sake of a net increase in aggregate utility."
Harvey states that "analysts who would never dream of advocating a policy
that violated human rights think nothing of supporting policies they know will
cause unemployment. They simply do not think of such policies as raising
human rights issues." Harvey's chapter presents a discussion of employment
as a human right both in terms of an instrumental value judgment (if the goal
is healthy individual and communities, then unemployment should be ad-
dressed) and in terms of the inconsistencies in America's basic categorical
judgments of value.
Some sociologists would accept the argument concerning instrumental
and categorical value judgments, but would still object to the practical
application of sociology on pragmatic grounds. They would argue that it
is good that sociological research draws very little attention from policy-
makers and the media because it both insulates the discipline from outside
pressures to pursue certain research topics, particularly those that are
topical, and protects the discipline from being sanctioned by the state if
the research does not support a particular political agenda or ideology.
There is some merit to this argument. However, if sociologists are con-
cerned with the present and future state of the discipline, it is shortsighted.
Why? Simply because the more sociology is ignored by policymakers and the
media, the less attention it receives as an academic discipline and therefore
the more removed it is from the decision-making arena, the fewer students it
attracts, and the more difficulty it has in trying to obtain funding support from
private foundations and government agencies. Moreover, as Carol H. Weiss
reminds us in Chapter 2, "if sociological work is to influence the course of
events, some of that influence will have to flow, sooner or later, through
William Julius Wilson 7

agencies of government. There are few mechanisms with the same poten-
tial for wide-scale impact on social issues."
However, overcoming resistance to the practical application of sociol-
ogy is just part of the problem. Sociology is also hampered by a narrow
interpretation of the use of sociological knowledge in the policy arena, a
point to which I now tum.



There is often intense pressure to address problems and dilemmas that

concern the nation or that threaten cherished values. Yet many sociologists
argue that we ought to wait until a sufficient amount of good data are
accumulated before making any policy recommendations or entering the
policy debate. If sociologists wait for more data before entering the policy
arena, or if they avoid issues of public controversy because of lack of data
even though their theoretical ideas or hypothesis would elevate the level
of the debate and broaden perspectives, decisions will be made and
policies will be formulated anyway-without their input (Lynd 1939).
Let me pursue this issue further by challenging the assumption that
sociology should not attempt to influence the national agenda until there
are "sufficient" or "adequate" data by arguing for: (1) a broader concep-
tion of the use and application of policy relevant sociological data, even
preliminary data; (2) an increase in the role of theoretical ideas, hypothe-
ses, and concepts in national policy debates; (3) a consideration of alter-
native ways or mechanisms to communicate insights from sociological
data, theories and concepts; and (4) the need to address the formalistic
fallacy in public policy research.

Broadening the Conception of the Use

of Policy Relevant Sociological Data
Just as one will rarely find in the social sciences a data set that would
unambiguously and incontrovertibly determine the validity of a major
theory or the correctness of a major factual question (Lieberson 1989), so
too is it uncommon to produce a data set that would unambiguously and
incontrovertibly resolve a public controversy. Although any social scien-
tist would like to have the greatest confidence in his or her data, sometimes
preliminary data can be used to reveal the narrowness of a public debate
or to challenge the general consensus on an issue, and thereby demonstrate
the need to take other factors into consideration.

For example, a few years ago the public policy debate over the causes
of the breakdown of the black family focused narrowly on the adverse
affects of welfare. In a paper first presented at a national welfare confer-
ence in Virginia in December of 1984, Kathryn Neckerman and I argued
for the need to consider the role of male joblessness in the growth of poor
single-parent black families (Wilson and Neckerman 1986). The aggregate
census-type data we presented in support of our position clearly suggested,
rather than firmly established, a positive relationship between male job-
lessness and solo-parent families. Nonetheless, the paper created a splash.
It not only altered the terms of the debate in academic circles and triggered
a round of new research among poverty researchers, but made policymak-
ers on Capitol Hill more aware that the issues surrounding the rapid growth
of single-parent poor black families were more complex than had been
previously assumed. Today male joblessness is routinely identified as one
of several important factors in the growth of solo-parent black families,
and the discussion of contributing factors no longer narrowly focuses on
the receipt of welfare. I
Also, in Chapter 17, Barbara R. Bergmann's discussion of the French
child welfare system raises important issues about the limitations of
overall child welfare, including child care, in the United States and
challenges us to broaden the terms of the child welfare debate in the United
States. The data that informed Bergmann's conclusions are limited to
France; however, she did not have to wait for pilot research on the
application of France's child welfare programs in order to provide provoc-
ative arguments for the public policy debate on child welfare in the United
Perhaps the best example of getting an important word out before all the
data are in is the initial reports on Dr. James P. Comer's creative interven-
tion model on inner-city education, described in Chapter 15. Comer pro-
duced a theoretical framework that relates the individual development,
including academic learning, of the child not only to his or her attachment
and bonding to adults who mediate experiences, teach and motivate in
primary social networks and in schools; but to opportunities and con-
straints in the broader community and society as well. Comer developed
an intervention program based on this framework and used it initially in
two elementary inner-city schools in the city of New Haven (CT). Of the
33 elementary schools in New Haven, these two schools were ranked last
and next to last in Language Arts and Mathematics achievement by the
fourth grade. "In 1984, 4 years after the program had been institutional-
ized, the two schools were tied for the third and fourth highest levels of
achievement in Language Arts and Mathematics by the fourth grade."
Although this model is now being tested elsewhere, because the findings
William Julius Wilson 9

in New Haven represent only one test case, the remarkable success of this
program in New Haven challenges the view that the problems in inner-city
schools are intractable and broadens our conception of ways to address the
educational retardation of ghetto children.
In addition to recasting the terms in which a public policy debate is
discussed, even with preliminary data, sociologists can play another im-
portant role in the policy arena by drawing attention to an important
problem that has been ignored by or has received insufficient attention
from policymakers. For example, Philippe C. Schmitter in Chapter 8,
"Organizations as (Secondary) Citizens," reveals that associations have
displaced individuals in the political process. This problem has received
only minimal discussion in the national policy forum. "Modern democracy
is increasingly becoming organized democracy," states Schmitter, "as
associations grow in number, extend their coverage, augment their re-
sources and consolidate their access to public authorities, the making of
binding collective decisions becomes more and more the outcome of
interorganizational bargaining, less and less a struggle between organiza-
tions and individuals, and even less a competition among individuals."
My book, The Truly Disadvantaged: The Inner-City. The Underclass,
and Public Policy (Wilson 1987), also emphasized a problem that had
received insufficient attention. Following the publication of the book I was
informed by Senator Bill Bradley of New Jersey that as far as he and a
number of other senators on Capitol Hill were concerned, the significance
of the The Truly Disadvantaged was not the specific public policy recom-
mendations advanced, but the fact that the book illuminated their under-
standing of the problems of ghetto poverty, raised their consciousness, and
increased their awareness of the need for effective public policy to address
these problems.
Finally, Roberta M. Spalter-Roth, Heidi I. Hartmann, and Linda M.
Andrews's discussion in Chapter 16 of the dismal plight of working
mothers increases our awareness of a serious problem. Their chapter also
calls our attention to the need for effective strategies to increase the
likelihood that working mothers can earn an adequate family wage.
Thus the issue here is not whether the data are adequate to advance
policy recommendations, or whether one can recast the terms of a policy
debate with certain kinds of data. Rather the issue is whether the descrip-
tion and sociological analysis of a problem that would be considered
important if fully recognized is sufficiently compelling and thought pro-
voking to enlighten or raise the consciousness and concern of policymak-
ers and the general public.
However, sociology's contribution to the policy arena need not be based
solely on empirical studies or research findings. As Carol Weiss points out

in Chapter 2, its theories, its ideas, its concepts "may also help to shape
what it is that the public thinks about and what it is that governments do."

Theoretical Ideas, Hypotheses, and Concepts

in National Policy Debates
Weiss argues that

although good data are useful and build credibility, equally important is the
sociological perspective on entities, processes, and events. Participants in the
policy process can profit from an understanding of the forces and currents that
shape events, and from the structures of meaning that sociologists derive from
their theories and research.

An important function of sociology is to use existing theories or theoretical

frameworks to advance our understanding of social processes and structures
(Lieberson 1992). In other words, sociologists can provide what Morris
Janowitz has called "enlightenment" (Janowitz 1970). Because of a narrow
vision of sociology's approach to public policy, we often overlook the fact,
as Carol Weiss points out in her chapter, that the public discourse on issues
such as persistent poverty, urban planning, pollution control, and criminal
justice has changed because of thought-provoking ideas from the social
sciences. Theories of class conflict and mobility have influenced government
policies in education, social services, and community development. Concepts
such as participatory decision making, labeling, concentration effects, or
maintenance of native language competence, have been incorporated in
policy discussion concerning criminal justice, mental health, poverty, and
education. "Sociological ideas, more than discrete pieces of data, have influ-
enced the way that policy actors think about issues and the types of measures
they have been willing to consider," states Weiss. The social sciences "bring
fresh perspectives into the policy arena, new understandings of cause and
effect; they challenge assumptions that have been taken for granted and give
credibility to options that were viewed as beyond the pale. They provide
enlightenment." Likewise. as John W. Kingdon points out in Chapter 3, "How
Do Issues Get on Public Policy Agendas?" although social scientists "can be
very good at documenting the existence, frequency, incidence, and intensity
of a condition," they are also frequently "able to show policymakers that the
world works in ways that might not have occurred to them." Kingdon also
argues that social scientists' knowledge of the way the world works enables
them to make better cause and effect connections than others.
The observations by Weiss and Kingdon are demonstrated in several of
the chapters in this volume. J. Donald Moon's stimulating discussion in
William Julius Wilson 11

Chapter 6 on citizenship and welfare contrast the liberal and social dem-
ocratic perspectives on social policies. Carefully explicating these two
different conceptions of the welfare state, Moon enhances our understand-
ing of the conflicts over the design of social policy, including those that
involve targeting and those that define and highlight the conditions in-
volved in the social obligation of citizenship.
Anne Wortham's insightful comments in Chapter 11 provide a unique
perspective to help us understand why the influence of New Right activ-
ists, who "contributed substantially to the election of Ronald Reagan to
two terms as President," failed to result in the expected retreat from the
New Deal welfare state. Wortham reveals how the forces that fed this
failure can be found in the self-contradictions of American populism, the
paradoxical relationship of democracy and Protestantism, and the de-
mands of modernity.
Moreover, Roger Lawson's informative comparison in Chapter 7 of
social policy in Great Britain and Sweden reveals why it is so important
to understand the important relationships between social rights and labor
rights (and more generally class politics), or welfare policy and labor
market policy in understanding why societies differ in their approaches to
social inequality.
Finally, in Chapter 13, Robert J. Sampson's community-level perspective
on both offending and victimization increases our theoretical understand-
ing of violence. His theoretical integration of community level factors
(including concentrated ghetto poverty, residential mobility and popula-
tion turnover, housing/population density, family disruption, local social
organization, and opportunity structures) in a framework to explain violent
crimes has profound implications for public policies regarding housing
and municipal services.
The chapters by Moon, Wortham, Lawson, and Sampson provide the kind
of fresh perspectives and theoretical insights that would be useful in discus-
sion and debates in the national policy forum. It would be shortsighted
therefore to discourage or overlook the use of such knowledge to inform
public policy debates as we wait for the compilation of "adequate" data.

Alternative Ways or Mechanisms to Communicate Insights

From Sociological Data, Theories, and Concepts
Some of the best sociological insights never reach policymakers, how-
ever, because sociologists seldom take advantage of useful mechanisms to
get their ideas out. Academic journals are infrequently read by officials of
government. One ofthe best ways to communicate sociological knowledge
is through the media. As Weiss points out, just as sociological ideas can

influence a reporter's perception and coverage of the news, so too do the

articles of a reporter often influence the public policy agenda. Although
occasionally an enterprising reporter will prepare a story based on an
article in an academic journal, sociologists who conduct public agenda
research should be encouraged to work with their university's department
of public information in preparing press releases and reports for the media.
The important work of Douglas Massey on hyper-segregation reached a
very wide audience through this mechanism (Massey and Denton 1987;
Massey and Eggers 1990).
Another important mechanism to bring sociological analysis into the
policy arena is through op-ed articles. Let me once again draw from my
own experience. As a sociologist I feel that my understanding of the
dynamics of race relations can provide insights into ways to address the
apparent growth of racial antagonisms in the United States and around the
globe. In my formal writings I have emphasized the situational aspects of
racial tension. I have argued, in other words, that racial antagonisms or the
manifestation of racial tensions are products of situations-economic
situations, political situations, and social situations. In a 1992 op-ed article
in The New York Times I used this argument in pointing out why it is
important for political leaders to channel the frustrations of citizens in
positive or constructive directions when economic times are hard.
I pointed to the political campaign of President Bill Clinton, who not
only explicitly acknowledged the growing racial tension in America and
the need for political leadership to unite, not divide, the races, but who
had actually developed a public rhetoric that reflected these concerns. This
campaign rhetoric emphasized that demagogy forces Americans to turn on
each other-race against race-instead of associating their declining real
incomes, increasing job insecurity, and growing pessimism with the real
source of the problems. I pointed out that the use of this public rhetoric
during a period of intense racial tension enabled Clinton to bring together
antagonistic racial groups in an effective political coalition, even in Lou-
isiana where a majority of white voters supported David Duke in last
year's gubernatorial election (Wilson 1992).
However, the media preoccupied with the candidate's personal past
failed to record the significance of this event. The op-ed piece, which drew
upon my insights as a sociologist, triggered a series of reports in national
electronic and print media on Clinton's approach in developing this mul-
tiracial coalition, and thereby demonstrated how the use of sociological
ideas can increase our awareness and understanding of important events
that relate to the national agenda.
From what I can judge, social scientists who have presented their ideas
in op-ed articles tend to represent what Steven Brint, in Chapter 4, "The
WilliamJulius Wilson 13

Powers and the Intellectuals: Benchmark Texts and Changing Condi-

tions," calls the "academic intellectual-experts." Their "political signifi-
cance derives principally from their ability to influence the climate of
educated opinion," states Brint. "They are, for the most part, interested in
particular issues, or sets of related issues, rather than in broad questions
about the proper order of society and government generally." Academic
intellectual-experts write, for example, about the conditions confronting
women in the work force, affirmative action, the state of American edu-
cation, the basis of economic decline among nations. "In these writings,
academic intellectual-experts mix normative judgments with scholarly
considerations of all of the diverse phenomena bearing on their subjects-
from legal opinions, to social change, to research findings, to past intel-
lectual arguments." Because they focus on a particular issue or a set of
relations issues, their main ideas could be easily presented or summarized
in op-ed pieces that average, say, a thousand words.

Public Policy Research and Overcoming the Formalistic Fallacy

Any discussion of the need to expand the domain of policy relevant
scholarship has to address the problem of what Stanley Lieberson has
called the formalistic fallacy. In public policy research, the formalistic
fallacy is the view that data for generating policy recommendations ought
to be obtained from the use of certain formal procedures or techniques
(Lieberson 1992). Nonquantitative research, for example, ethnographic
research, is therefore considered inappropriate for generating policy rec-
Although all scholarly work should be subjected to critical review,
concern should focus on the logic of inquiry-the structure of explanation,
the significance of concepts, and the nature of evidence-not on the
procedures or techniques used (cf. Lieberson 1992, p. 4). Let me elaborate
on this point as it relates to public policy relevant research.
Quantitative social science established its hegemony in the 1970s, aided
by the development of quantitative techniques in history. Ethnographic
research in fields such as urban poverty, which had been revived in the
1960s, was basically dormant in the 1970s. In the 1980s, however, we were
beginning to see a shift in focus away from quantitative versus qualitative
research to an approach that emphasized integrating the two strategies in
empirical studies that focused on problems such as urban poverty.
There are several intellectual and practical issues involved in the inte-
gration of quantitative and qualitative techniques. These issues relate to
the important distinction between the context of discovery and the context
of validation. Whereas the context of discovery is a matter of empirical

inquiry concerned with the way in which fruitful concepts, hypotheses,

and theories are discovered, the context of validation is concerned with
the evaluation of the products of science and therefore with making the
evaluative criteria as explicit as possible.
I emphasize this distinction because in recent discussions a number of
people have maintained that the best way to integrate quantitative and
ethnographic research is to use the former in the context of discovery and
the latter in the context of validation. In other words, it is argued that
ethnography ought to be used to generate the hypotheses that could then
be tested with quantitative research.
More specifically, the major objection to using ethnographic research
in the context of validation is the inherent difficulty in generating a sample
representative of a larger population. However, there is another type of
sampling crucial to theory testing that addresses the issue of whether the
conditions specified by the theoretical assumptions that guide the research
are represented. This is known as theoretical sampling, defined as select-
ing a number of natural cases that fit the conditions appropriate to the
assumptions of the theory.
For example, I have been working on a theory of the social transforma-
tion of the inner city and a number of the key hypotheses incorporate the
notion of "concentration effects"-the effects of living in highly concen-
trated poverty areas. One of these hypotheses states that individuals living
in extreme poverty areas are much less likely to be tied into the job network
system than those living in marginal poverty areas. I contend that this
hypothesis could be tested by a participant observer who selects a neigh-
borhood that represents an extreme poverty area and a neighborhood that
represents a marginal poverty area, and who observes patterns of work-re-
lated behavior in each neighborhood over an extended period of time.
Some people may want to question the degree of rigor involved in testing
such a hypothesis with participant observation techniques, but this ap-
proach is clearly consistent with the logic of validation.
The study by Frank F. Furstenberg, Jr., in Chapter 12, "How Families
Manage Risk and Opportunity in Dangerous Neighborhoods." is an excel-
lent example of the use of ethnographic data in the context of validation.
Based on the field research of five of his research assistants, that included
"extended open-ended interviews with parents and their teenage children
in five distinct inner-city neighborhoods," Furstenberg concluded that
family management is profoundly influenced by the communities in which
the families reside. "Ordinary parents are likely to have more success when
they reside in communities where the burden of raising children is seen as
a collective responsibility and where strong institutions sustain the efforts
of parents." .
William Julius Wilson 15

As an example of public agenda research, Furstenberg's findings have

clear implications for public policy designed to "strengthen the family or
improve the situation of children." Furstenberg points out that many of our
social programs focus solely on improving the material, informational, and
psychological resources of parents so that they might manage the task of
child rearing. Since the full burden of caretaking is attributed to parents,
they receive the full measure cfblame when their children do not succeed.
However, Furstenberg argues persuasively that if we are committed to
strengthening the family, more attention must be given to rebuilding
neighborhood centers, recreational services, schools, churches, and other
local institutions that support families. "Rebuilding local community
institutions may be a potent way of supporting beleaguered poor parents
and ensuring a better future for their children."
Ideally, and as Furstenberg emphasizes, one would want to test this hypoth-
eses with more quantitative data sources that would include a large number
of individuals from a variety of urban poverty and nonpoverty areas. But the
ethnographic research, including leisure conversations with people about
their experiences over extended periods of time, could uncover many subtle
patterns of behavior and experience that are difficult, if not impossible, to
ascertain with the more conventional research techniques.
I contend that combining research techniques within the context of
validation provides the richest data set not only to test theoretical hypoth-
eses but also to uncover new, important issues not raised or suggested in
the original theoretical formulations. There is an increasing awareness that
despite all of the highly sophisticated quantitative studies of the 1970s, we
really learned very little about the important interaction between social
structure and cultural patterns of behavior among the poor, including the
inner-city poor. Increasingly, scholars have been calling for more ethno-
graphic studies to provide some of the answers.
Research from the University of Chicago's Urban Poverty and Family
Life Study (UPFLS) of Chicago combines survey and ethnographic re-
search. And it is clearly apparent that the results from these two modes of
research provide deeper insights on the changing structure and experiences
of urban poverty than if we had relied on a single approach. Indeed, I have
drawn a number of specific policy recommendations from the UPFLS that
were based on findings from both the quantitative survey data and the
ethnographic data (Wilson 1991). For all these reasons, I recommend
greater reliance on ethnographic research as sociology addresses policy
relevant issues. To eschew such research in the policy domain on formal-
istic grounds is to limit sociology's potential to shape the national agenda.



In my foregoing remarks, I have focused on the need to expand the

domain of policy relevant scholarship in sociology to enhance our ability
to shape the national agenda. There are, however, other problems that have
to be overcome that involve our selection of research topics, the issues we
tend to stress or highlight, and/or the issues we are willing to pursue in the
public policy arena. These problems often reflect the impact of ideology
both within and outside the discipline.
For many people in the policy arena, sociology has the image of being
a "champion of the underdog." Sociology has a rich and proud tradition of
research on the disadvantaged. However, there is a general view in some
circles that the policy relevant studies of sociology tend to be skewed to
reflect the interests and concerns of the disadvantaged, especially the more
oppressed groups in society. Conversely, there is the perception that
sociological studies on or about the more advantaged groups or powerful
institutions of the society reflect a negative bias . .This image, a bias that
favors the disadvantaged, is not associated with the discipline of econom-
ics, which is taken more seriously or viewed as having more credibility
among those who help shape the national agenda.
I believe that sociology has been adversely affected by this stereotype
because it affords the opportunity for skeptics to dismiss those of us concerned
about policy relevant research as ideologues. It would therefore be beneficial
if greater public visibility could be given to some of the sociological studies
that undermine this stereotype, if for no other reason than to demonstrate the
great diversity of policy relevant research in the discipline. I have in mind
two studies relevant to social policy presented in this volume: David Knoke's
"Networks as Political Glue" (Chapter 9) and Neil Fligstein and Linda
Markowitz's "Financial Reorganization of American Corporations in the
1980s" (Chapter 10); and Stanley Lieberson's (1971) insightful earlier study
of the relationship between industry and the military.
David Knoke argues that an adequate understanding of public policy-
making requires thorough attention to the structure of power relations that
connect actors, events, and decisions. Knoke presents a comprehensive
theoretical framework that shows how "social actors mobilize political
power to achieve their preferred policy objectives." He illustrates the
process by drawing examples from research on individual voter decisions,
on decision making in collective action organizations, and on social
networks in elite circles in national policy domains.
Fligstein and Markowitz's interesting Chapter 10 reveals that the fi-
nance conception of the large firm, that is, "a cultural frame that views the
William Julius Wilson 17

essence of the corporation in primarily financial terms," influenced the

1979-1987 financial reorganization of the largest firms. In organizations
already dominated by the finance conception, managers were more likely
to pursue mergers, divestiture, stock repurchases, and investment in the
stocks of other firms. Corporate firms that were not dominated by this
perspective often received pressure from new board members who repre-
sented the finance conception to realign in accordance with this perspec-
tive. "If top executives or important board members fail to force realign-
ment with that perspective, then firms are liable to become the targets of
executives who do." The findings by Fligstein and Markowitz show that
market relations are tied up with social relations. One has to understand
the context of the social relations to understand actions involved in market
relations. The authors hypothesize that one of the factors contributing to
the "intensification of the finance orientation toward firms" was the
"Reagan Administration's suspension of the antitrust laws and its tax
policies in the early I980s," policies that reinforce the orientations of
finance-oriented managers and institutional investors.
The studies of both Knoke and Fligstein and Markowitz focus on the
actions of powerful institutions and actors in the United States and provide
perspectives that inform our understanding of the origins and direction of
social policy. They therefore undermine the stereotype that policy relevant
studies in sociology reflect the interest and concerns of the disadvantaged.
One of the most important sociological studies that challenges this
stereotype was written by Stanley Lieberson (1971). His research sug-
gested that the notion of the military-industrial complex needs to be
reexamined. The data that Lieberson collected and analyzed showed that
increased military spending benefited only a minority of American indus-
tries, that a majority of industries would gain more if the same sum for
military spending were marked for nonmilitary spending, and that in-
creased military spending could occur even if most industries in America
did not stand to gain. These important findings, buried in an academic
journal, were apparently not discussed by the media and were probably
ignored by policymakers. Lieberson's main arguments would have been
an ideal focus for an op-ed article in national newspapers such as the Wall
Street Journal, The New York Times, or The Washington Post.
A related problem associated with public policy relevant scholarship in
the discipline is the pursuit of the "correct perspective.t'/ An important
role of the sociologist as a social scientist is the ability to capture and
analyze the perspectives of different groups on similar or different social
issues. As straightforward as this seems, a problem arises when the
investigator avoids either describing or describing and analyzing objec-
tively a perspective of which the discipline, or in some cases the larger

society, generally disapproves. Although we are often perfectly content to

try to understand and communicate to others the viewpoints of groups
whose positions we support or with whom we are in sympathy, we are
frequently reluctant to do the same for groups whose positions we disap-
prove or whom we disdain (Lieberson 1988). Nowhere is this tendency
more pronounced than among those who study intergroup relations.
Whereas the perspective of, say, poor minorities is often described care-
fully and with considerable sympathy, the perspective of working- or
lower middle-class whites on issues of race and culture is often ignored,
misunderstood, or treated with contempt.
Lower middle-class culture reflects an emphasis on the family, the church,
and the neighborhood. A community's continuity is valued more highly than
individual advancement, social solidarity is favored over social mobility, and
the maintenance of existing ways takes precedence over conventional ideals
of success. Parents want their children to succeed in life, but they also want
them to be considerate of their elders, willingly bear their responsibilities, and
to show courage under adversity. "More concerned with honor than with
worldly ambition, they have less interest in the future than do upper-middle-
class parents, who try to equip their children with the qualities required for
competitive advancement" (Lasch 1991, p. 490).
There is no doubt that lower middle-class culture is provincial and
narrow and "it has produced racism, nativism, anti-intellectualism, and all
the other evils so often cited by liberal critics" (Lasch, p. 17). But, in their
zeal to condemn the objectionable traits, sociologists-with a few notable
exceptions (Gans 1988 and Rieder 1985)-have failed to provide a socio-
logical understanding of lower middle-class perspective on the abortion
debate, family values, the politics of race (particularly the politics of
busing and affirmative action), and other issues that dominate liberal
discussions ofthe "white backlash."
Indeed, lower income whites, like inner-city minorities, have felt the
full impact of the urban fiscal crisis in the United States. Unlike middle-
class whites, they have been forced by financial exigencies to remain in
the poorer parts of the cities and suffer the strains of crime, poorer
services, and higher taxes. Unlike the more affluent whites who choose to
remain in the wealthier sections of the cities, they cannot easily escape the
problems of deteriorating public schools by sending their children to
private schools, a problem made worse by the sharp decrease in the number
of urban parochial schools.
Many of these people originally bought relatively inexpensive homes
near their industrial jobs. Because of the deconcentration of industry, the
racially changing neighborhoods bordering their communities, the prob-
lems of neighborhood crime, and the surplus of inner-city housing created
William Julius Wilson 19

by the population shift to the suburbs, housing values in their neighborhoods

have failed to keep pace with those in the suburbs. As the industries in which
they are employed become suburbanized, a growing number of lower income
whites in our central cities find that not only are they trapped in their
neighborhoods because of the high costs of suburban housing, but they are
also physically removed from job opportunities as well. This situation in-
creases the potential for racial tension as they compete with blacks and the
rapidly growing Hispanic population for access to and control of the remain-
ing decent schools, housing, and neighborhoods in the central city. And
explanations that their negative response to minority encroachment is due to
racial prejudice hardly captures the complex issues that have shaped their
unique perspective. To repeat, one does not have to agree with this perspective
in order to analyze it thoroughly. Indeed, an attempt to provide a more
comprehensive understanding of perspectives that do not draw sympathy in
our field would not only enhance our credibility as social scientists, but would
also provide important information in the policy arena when steps are consid-
ered to reduce intergroup conflict.
Related to the problem of selective attention to group perspectives in
social policy relevant studies is the pressure to provide politically correct
descriptions, including the use of appropriate concepts and conclusions.
Weare fully cognizant as a discipline of the need to be on guard against
efforts by powerful segments of the larger society to sanction studies
whose conclusions challenge strongly held political assumptions, beliefs,
or sacred programs, but we tend to overlook the danger when we ourselves
insist on politically appropriate studies.
For example, sociologists working in the field of urban poverty have
felt pressure from their colleagues to consider the political and/or social
implications of their work, even to the point of suppressing results or of
avoiding certain research topics. The vitriolic attacks and acrimonious
debate that characterized the controversy over the Daniel Patrick Moyni-
han report on the Negro family in the late 1960s and early 1970s, a
controversy that emerged because his ideas were seriously misrepresented
in the media and in the work of some social scientists, is a case in point.
The controversy following the report intimidated many sociologists and
other social scientists conducting research on poverty and family struc-
ture. Indeed, in the aftermath of this controversy and in an effort to protect
their work from the charge of racism or of "blaming the victim," sociolo-
gists, like other liberal social scientists, tended to avoid describing any
behavior that could be construed as unflattering or stigmatizing to racial
minorities. Accordingly, for a period of several years, and well after this
controversy had subsided, the problems of social dislocation in the inner-
city ghetto did not attract serious research attention. From the early 1970s

to the mid-1980s, there was minimal contribution from sociology in

particular and the social sciences in general to the public policy agenda on
combating inner-city poverty.
Finally, it is necessary for sociologists to give much more attention to what
Stanley Lieberson (1988) calls the side effects of certain social policies
recommended by social scientists on the basis of their research or study. In
other words, in policy research we have to be concerned not only about the
effect of intervention A on outcome B, but also the possible effect of
intervention A on other factors, including those that further affect or change
outcome B. For example, in the early 1970s a number of notable sociologists
strongly urged busing as a way to enhance the educational experiences of
black students. The strong recommendations for busing continued even after
it was apparent that urban public schools were being abandoned by whites. It
is reasonable to assume that the inflexible position was in part a function of
a strong ideological commitment to school desegregation.
For many whites, however, the decline in public schools was associated not
with the broader changes in society but with forced integration that sent their
children to schools in inner-city neighborhoods and brought black children
into their white enclaves (Califano 1988). From a public policy perspective,
it would have been useful if those sociologists who recommended busing as
a way to enhance the education of black children also considered the idea of
alerting policymakers to the possibility that a vigorous program of busing
could accelerate white flight from urban public schools and therefore further
increase segregation in the public schools.


Scholars involved in public agenda studies confront a unique problem

that has to be acknowledged and anticipated-the problem that their work
is much more likely to be the subject of ideological attack. Indeed, it would
be naive to think that such studies can avoid critical scrutiny that is
motivated more by ideological than by substantive, theoretical, or meth-
odological considerations. Many sociologists are fully aware of this prob-
lem and avoid public policy research, or are not interested in seeing to it
that their ideas or contributions actually enter the public policy domain.
However, as I have tried to argue above, the future of the discipline may
very well rest on the degree of influence that sociology wields in shaping
the national agenda.
Our discipline is plagued with feelings of insecurity, and is concerned about
and tries very hard to avoid criticism. Many sociologists are therefore reluc-
tant to become participants in the public policy debate before we have
WilliamJulius Wilson 21

"adequate" information, even though their views on the issues are far more
sophisticated and comprehensive than those of even some of the most influ-
ential opinion leaders. George Bernard Shaw once said that it is better to be
criticized or misunderstood than to be ignored. This not only applies to the
work of individual scholars, it also applies to an entire academic discipline.
I have advanced some suggestions that I believe would increase sociology's
visibility in the national forum. I believe that we have to be concerned about
the future of our discipline and we need to recognize that our continued
viability is inextricably connected with success in becoming a more important
player in the public policy arena. Accordingly, resistance to the practical
application of sociology must be overcome. A more aggressive and positive
orientation toward public agenda research and scholarship in sociology has
to emerge. This includes expanding the domain of policy relevant scholarship
so that we become more flexible in the kinds of data that we use and the ways
in which we use them; recognizing the important role of sociological theories,
concepts, and ideas in the formulation and discussion of public policy issues;
expanding the outlets for sociological insights; and recognizing the influence
of ideology not only in the way that we select, interpret, and analyze issues
for study that relate to the public agenda, but in the public policy recommen-
dations that we advance based on our research.
It will become clear to the reader that the chapters in this volume raise
issues and advance arguments consistent with many of the suggestions that
I have presented to enhance sociology's image and increase the discipline's
visibility in the national forum.


1. Another problem in the way some sociologists use policy relevant data is the failure to
discuss the possible wider implications of their findings. In other words, sociologists may
have intensively studied a problem and are fairly confident about the empirical results but
they avoid extrapolating their data, however tentatively stated, into what Lynd has called the
"realm of wide meaning." This simply invites others, representing organizations that range
from the far right to the far left and who are presumably more biased than the typical
sociologist, "to thrust upon the culture their interpretation of the meaning of the situation"
(Lynd 1939, p. 186).
2. In the remaining parts of this section I am indebted to Lieberson 1988.


Califano, Joseph A., Jr. 1988. "Tough Talk for Democrats." The New York Times Magazine
(January 8):29.

Gans, Herbert J. 1988. Middle American Individualism. New York: Free Press.
Kantrowitz, Barbara. 1992. "Sociology's Lonely Crowd." Newsweek (February 3):55.
Lynd, Robert S. 1939. Knowledge for What? The Place of Social Science in American
Culture. Princeton, NJ: Princeton University Press.
Hempel, Carl G. 1965. Aspects of Scientific Explanation. New York: Free Press.
Janowitz, Morris. 1970. "Sociological Models and Social Policy." pp. 243-59 in Political
Conflict: Essays in Political Sociology, edited by Morris Janowitz. Chicago: Quadrangle.
Lasch, Christopher. 1991. The True and Only Heaven: Progress and Its Critics. New York:
Lieberson, Stanley. 1971. "An Empirical Study of Military-Industrial Linkages." American
Journal of Sociology 76:562-84.
- - - . 1988. "Asking Too Much, Expecting Too Little." Sociological Perspectives 31(Oc-
---.1989. "When Right Results Are Wrong." Society 26(1uly/August):6G-66.
---.1992. "Einstein, Renoir and Greeley: Evidence in Sociology." American Sociologi-
cal Review 57(February):I-l5.
Lynd, Robert S. 1939. Knowledge for What? The Place of Social Science in American
Culture. Princeton: Princeton University Press.
Massey, Douglas S. and Nancy A. Denton. 1987. "Trends in the Residential Segregation of
Blacks, Hispanics and Asians: 1970-1980." American Sociological Review 52:802-25.
Massey, Douglas S. and Mitchell L. Eggers. 1990. ''The Ecology of Inequality: Minorities
and the Concentration of Poverty, 1970-1980." American Journal of Sociology
Rieder, Jonathan. 1985. Canarsie: The Jews and Italians of Brooklyn Against Liberalism.
Cambridge, MA: Harvard University Press.
Wilson, William Julius. 1987. The Truly Disadvantaged: The Inner City, The Underclass,
and Public Policy. Chicago: University of Chicago Press.
---.1991. "Poverty, Family Structure and Joblessness in the Inner City: A Comparative
Perspective." Paper presented at the Chicago Urban Poverty and Family Life Confer-
ence, Chicago, October.
---.1992. "The Right Message." The New York Times (March 17): p. A3.
Wilson, William Julius and Kathryn Neckerman. 1986. "Poverty and Family Structure: The
Widening Gap Between Evidence and Public Policy Issues." Pp. 232-59 in Fighting
Poverty: What Works and What Doesn't, edited by Sheldon Danziger and Daniel
Weinberg. Cambridge, MA: Harvard University Press.
The Interaction of the Sociological Agenda
and Public Policy

Carol H. Weiss

The theme of the 85th Annual Meeting of the American Sociological

Association, "Sociology and the Public Agenda," couples sociology and
the public agenda with an innocuous "and." My question is: Which influ-
ences which? To what extent does the public agenda influence sociology?
To what extent does sociology affect the public agenda?
To make the questions manageable, I will limit discussion to the subset
of the public agenda that involves governmental policy and governmental
action. I want to consider the ways in which government policy and the
sociological agenda influence one another. I will limit the discussion
largely to the subset of sociology that is practiced in universities and
freestanding research organizations.
There are four main hypotheses to be considered (see Figure 2.1). The first
is that public policy influences the sociological agenda for research and
theorizing. This is very likely to be the case, for example, to the extent that
government funding affects which sociological research gets done. Whether
or not public policy intends to discourage research on certain topics, it can do
so by pouring resources into some substantive fields and ignoring others.
When public policy supports evaluation research or studies of AIDS, sociol-
ogists may be likely to jump on the bandwagon---or the gravy train-and do
research that fits current policy priorities. A second hypothesis is that sociol-
ogy influences the public agenda. Through its theories and data, its concepts,
ideas, and the findings of its research studies, sociology may help to shape
what it is that the public thinks about and what it is that governments do.

AUTHOR'S NOTE: I wish to thank Robert Smith and Bruce Fuller for their comments.


Hypothesis I. Government policy influences the sociological agenda.

Hypothesis 2. Sociology influences the public policy agenda.


Hypothesis 3. Sociology and public policy are separate and independent.

Hypothesis 4. Sociology and public policy each respond independently to the zeitgeist.

Figure 2.1. Hypotheses About Mutual Influences of Government Policy and the
Sociological Agenda

A third hypothesis is that sociology and government policy proceed

relatively independently with little influence flowing in either direction.
In many ways, this is the most plausible hypothesis, because relatively few
sociologists receive government research funds and are thus directly
Carol H. Weiss 25

susceptible to government influence, and the influence of sociology on

government is not obvious from casual inspection.
Finally, it is possible that sociology and government policy independently
respond to outside events and currents of thought. Even if we find what appear
to be convergences between them, the convergences may be due not so much
to direct impingement of one on the other as to their separate and independent
reactions to larger social currents. Sociologists, in their statuses as citizens,
parents, workers, voters, patients, and so on, engage themselves in the issues
of the day, and that engagement can influence their sociological work. The
concerns of the larger society, which stimulate public policy responses, may
also stimulate sociological research and theory.
The four hypotheses are, of course, not mutually exclusive. It is possible
that some combination of them is operative, or even that all of them receive
support at certain times and places. If several of them have some explanatory
power, then we will want to discuss the conditions under which each expla-
nation of the interaction of sociology and public policy is likely to hold.



The means that any government has to influence an academic discipline

are limited. Its main "policy instruments" are taxation and regulation,
resource allocation, direct provision of services, and persuasion. Because
there have been few governmental efforts to regulate, persuade, or provide
alternative sociological services, the case rests largely on the importance
of resource allocation. And the ways in which resources affect the socio-
logical agenda have been primarily through support of research.
There is little doubt that governmental funding of research in particular
areas immediately increases the amount of research done in that area.
Robert Haveman (1987) has documented the effect of government atten-
tion to poverty on the rise of poverty research (not only in sociology; even
more obviously in economics). Research grant programs of the National
Institute of Mental Health (NIMH) from the 1950s to the 1970s led to the
growth of research on mental health topics (here psychology was the
primary recipient although sociology benefited as well). Currently the
increase in spending for AIDS research is affecting the research agenda of
some sociologists, including several large survey research operations.
Current funding at the National Science Foundation (NSF) for research on
math and science education is attracting sociological attention to those
fields. In fact, sociologists have lamented the responsiveness of the dis-
cipline to topics legitimated and promoted by government functionaries.

(There has been parallel concern when a government agency appears to

promote one type of research methodology, as when NSF exhibits a consistent
preference for large-scale quantitative studies.) When sociologists go along
with government priorities, it seems to be a sellout. It allows bureaucrats
with different interests and a very different agenda to distort the priorities
of the discipline and perhaps undermine the independence and intellectual
autonomy that sociologists should maintain.
But several factors mitigate the influence of government through resource
allocation. First, whereas a sizable number of sociologists may be tempted
to apply for grants, and many do apply, only a few actually receive grants
from government. So although government research funding may influ-
ence what some sociologists consider doing, its effect on the research that
sociologists actually do is limited. For sociologists in research firms or
nonprofit organizations dependent on grants and contracts, the effect is no
doubt greater than for those sociologists in academic departments.
Another reason why government research funding has probably not had
more impact on the discipline is that social scientists have been canny in
reshaping government mandates to fit their own research priorities. When
money was available for behavioral research on mental health, sociologists
convinced funding bodies that they could advance mental health by studying
neighborhood socialization patterns, assimilation of immigrants, dyadic rela-
tionships between marriage partners, and elite opinion leadership. Similarly,
sociologists are getting AIDS research funds to study social scripts for
intergroup relations, values formation in adolescent groups, and other sub-
jects on their own research agendas. The situation is much as it has always
been. As Rossi (1980) reminds us, the classic North-Hatt study of occupa-
tional prestige (Reiss 1961) was funded by a federal agency interested in the
declining prestige of federal employment. North and Hatt used the study as a
way to advance their interests in social stratification. Sociologists have long
bent government funding to serve sociological ends.
Perhaps a third reason that government money hasn't dominated soci-
ology is that it has never been the exclusive source of research funding.
Even when the amounts available from government overwhelmed all other
sources combined, there were still foundations, other organizations, and
university resources to fund studies on diverse topics.
The most dramatic resource move that the U.S. government has made in
the past decade has been the cutback in funds available for sociological
research. The cutback, although not overtly aimed at any particular element
on the sociological agenda, was directed against research that seemed to
promote government intervention, particularly for social programs. Clearly
the cutback has influenced the kinds of research that have and have not been
done. The effects of the cuts have been amplified by a strategy adopted by a
Carol H. Weiss 27

number of funding agencies, including NSF, the Department of Labor, and

the Department of Education, to maintain government funding of large-scale
longitudinal surveys and therefore, of necessity, to cut back on smaller
investigator-initiated studies. Public-use data from longitudinal studies are
available for reanalysis, and sociologists with good computer and statistical
skills are able to use them for their own purposes-if they are able to locate
variables that can serve as measures or surrogates of features they are
interested in. But sociologists are increasingly driven to "make do" with
variables that don't necessarily measure what they want to study, and some
recent research has a shaky linkage between the theories purportedly inves-
tigated and the variables actually used. And sociologists who lack statistical
expertise or who cannot locate relevant variables in existing data bases have
had a lean time of it. Fewer investigator-initiated studies are being done,
especially when these would involve new large-scale data collection.
Does government influence the development of sociology through any of
its other instruments-regulation, services, or persuasion? Government reg-
ulations can tighten the reporting requirements on federal grants, require
matching funds, and so forth. These kinds of rules can make the researcher's
work life more burdensome, but they are unlikely to affect the topics that are
studied. State governments can influence the fate of sociology in state
universities: they can expand or contract departments, raise or cut salaries,
and so on. These kinds of actions affect the health and status of the discipline;
they affect numbers of jobs, numbers of students, opportunities for advance-
ment, even perhaps the standing of the discipline among other disciplines.
But again they are not likely to affect the topics that are studied or the theories
that are propounded. For all the severity of the impacts on sociology, they do
not seem to influence the intellectual development of the discipline.



This is a subject that I have studied for more than a dozen years in terms
of both executive (Weiss 1977, 1978, 1986; Weiss and Bucuvalas 1980a,
1980b) and legislative (Weiss 1989) policy, at federal, state, and local levels.
Other sociologists, too, have looked at the effects of sociology on government
policy (e.g., Barber 1987; Lazarsfeld and Reitz 1975; Orlans 1973; Scott and
Shore 1979; van de Vall 1986), and we have been joined by political scientists
(e.g., Lindblom and Cohen 1979; Rich 1981; J. Wilson 1978), psychologists
(e.g., Caplan 1982; Melton 1987), and economists (e.g., Aaron 1978, 1989;
Nathan 1988; Haveman 1987). A considerable body of knowledge is devel-
oping on the subject of social science and public policy.

To be brief, this research shows that the kinds of effects that many social
scientists expected to see on government policy are only occasionally
evident. The social sciences are rarely applied to select Policy A over
Policy B or to change the features of government programs. Sociology is
not "used" for the specific answers, let alone the recommendations, that it
provides; it does not work to tell government what to do. On the other
hand, it does seem to work to tell government what to consider. Sociology
is sometimes-underline sometimes-used to illuminate the nature of
social problems and to recast the terms in which they are discussed. The
social sciences have changed the terms of discourse in many fields from
criminal justice to pollution control, from chronic poverty to urban plan-
ning. They bring fresh perspectives into the policy arena, new understand-
ings of cause and effect; they challenge assumptions that were taken for
granted and give credibility to options that were viewed as beyond the
pale. They provide enlightenment.
We have found, too, that policy actors find it difficult to pinpoint
individual studies that have altered their ways of seeing the world. They
have a sense that the social sciences have influenced the way they think,
but they cannot disentangle sociological information from the many other
sources of knowledge to which they are exposed. Sociological knowledge
enters their stock of knowledge and becomes part of the organizing
perspective they use to make sense of their work. This amorphous perco-
lation of sociological ideas into the policy arena has been called "knowl-
edge creep" (Weiss 1980).
Sociology has other effects, too. Policy actors sometimes use research
to support positions that they already want to take; they use the research
to provide legitimacy and justification for their cause. On occasion, they
use sociology to help persuade others that the cause is right. It provides
ammunition to mobilize supporters and develop coalitions. But it is the
"enlightenment" function of sociology (Janowitz 1970; Weiss 1977) that
is most pervasive. Broad theories, such as those on class conflict or
individual mobility, have influenced government policies in education,
community development, and social services. Frames or concepts, such as
those of labeling, participatory decision making, or maintenance of native
language competence, have had an impact in criminal justice, mental
health, poverty policy, and education. Sociological ideas, more than dis-
crete pieces of data, have influenced the way that policy actors think about
issues and the types of measures they have been willing to consider.
The influences are often indirect, sometimes amorphous, sometimes
slow, and perceptible more to policy actors than to the outside observer.
A famous quotation from John Maynard Keynes (1936) captures one of
the paths to influence:
Carol H. Weiss 29

The ideas of economists and political philosophers, both when they are right and
when they are wrong, are more powerful than is commonly understood. Indeed,
the world is ruled by little else. Practical men, who believe themselves to be quite
exempt from any intellectual influences, are usually the slaves of some defunct
academic economist. Madmen in authority, who hear voices in the air, are
distilling their frenzy from some academic scribbler of a few years back. I am
sure that the power of vested interests is vastly exaggerated compared with the
gradual encroachment of ideas.

Since Keynes's time, the permeation ofthe policy sphere with the data
and findings of social science has proceeded apace. Today it is a rare
public debate that doesn't rely in some considerable measure on studies,
statistics, and evaluations. In legislation, in administration, and increas-
ingly in judicial proceedings, evidence from the social sciences is drawn
into discussion. Participants in public policymaking almost have to be able
to speak the language of social science to make their points and to protect
their positions against attack. Even though some of this change in the
language of public discourse is little more thana change in fashion,
nevertheless it opens the door to serious consideration of social science
information. Once a participant has introduced sociological evidence into
debate, others-particularly those opposed to the position-will scrutinize
it in order to discredit the ideas. Through this kind of adversarial attention,
study findings move into circulation and into political currency.
Evidence from social science research can reduce disagreements over
matters of fact (e.g., whether fewer pregnant women are receiving prenatal
care, whether vocational education improves employability and job per-
formance). In doing so, it helps to raise the level of debate, freeing policy
actors to talk about matters of value-which are their proper province.
Analogously, the concepts and theories of sociology make a difference.
They are helping to make public decision makers more sophisticated about
social structure and group processes (less content with individual-level
explanations for social phenomena), and they are gradually infusing po-
litical thinking with more complex and subtle notions of conflict, social
disorganization, community norms, social movements, and other sociolog-
ical constructs. Obviously the processes of enlightenment are not evident
in all venues and at all times, but a comparison of the political debates of
today with those of a generation ago will reveal, I believe, a gain in
sophistication about the social (as well as the economic) world.
That there is still a long way to go almost goes without saying. The creep
of sociological ideas into policy arenas remains an erratic phenomenon.
Some ideas are distorted or misunderstood; other are ignored. Those
findings and theories that support existing positions are better heard than

those that do not. Still there are significant areas in which social science
has influenced the way that public problems are conceptualized and the
solutions that are considered.



This hypothesis suggests that public policy is little moved by sociology and
that sociology basically goes about its business without much regard for
government policy. If this is so, we have to see what alternative mechanisms
account for the current state of sociology. In the absence of direct or indirect
government influence, how is "the sociological agenda" set?
The standard answer is that the agenda for both research and theorizing
arises from the core problems of the discipline. By and large, I think that
is a lot of piffle. Some number of sociologists concern themselves with
what have become consensually known as the core problems of the dis-
cipline, but the direction of the arrows is the other way around. What smart
and creative sociologists do become the core problems of the discipline.
Sociology is such a large and sprawling field that there are scores of issues
that are potential candidates for "core problems of the discipline." Aside from
a few hardy mainstays, like social conflict and social order, the list of "core"
issues is capable of major turnover. (Even issues of conflict and order may be
highlighted only in particular historical periods.) When creative sociologists
come up with heuristic formulations or reformulations, their ideas are likely
to stimulate a host of further studies, and out of the ferment a new core issue
emerges. New concepts appear in mass communications when creative re-
search and thinking are done by people like Gans (1979) and Schudson
(1978). Core issues change in the fields of poverty and ethnicity when new
insights are developed by people like W. Wilson (1978, 1987). Because
sociological concerns range over almost the entire spectrum of human life in
social groups, the discipline can pick and choose its core.
Core issues change over the decades. The reason is certainly not because
we have solved the old problems but because new ideas and new issues
move to center stage. Kuhn's (1970) idea of scientific revolutions ap-
pealed more to social scientists than to the natural scientists on whom it
was based, because we know how apt the picture is for our fields. Actually
in sociology what we experience is not so much revolutions, where strong
theoretical models are overturned or replaced, as a fairly constant churn-
ing. Theories of delinquent behavior move from notions of differential
association to opportunity theories, which are replaced by labeling theo-
ries, which are overtaken by a more catholic acceptance of a variety of
Carol H. Weiss 31

Table 2.1 Sources of the Sociological Agenda

Individual Level Societal Level

Extra-Sociological Own experiences and personal Availability of research funding

interests and other opportunities
Contemporary social issues that
engage the researcher

Sociological "School" of graduate training Topics, methods, and theories

Topics promoted by professors, valued in the discipline, i.e.,
adviser preferred by journals, recognized
Methods and theories that through awards, prestige, etc.
researcher adopts Intrinsic interest of reigning
New questions raised by own theoretical/empirical questions
past research

explanations for the behavior of different youth at different phases of their

delinquent careers.
It is educational to read the agendas for sociology published in earlier
decades. I recently reread parts of Merton, Broom, and Cottrell's 1959
volume on problems and prospects for sociology. Although some of the
landscape is familiar, the emphases are quite different. It doesn't appear
as though we have made great headway in answering the questions that
were on the sociological agenda 30 years ago. Our attention has simply
shifted to new problems and prospects-not because of government poli-
cies but because events, ideas, and issues have affected sociologists just
as they have affected everyone else living in the United States in the latter
years of the 20th century.
How do sociologists choose the problems on which they work? Loubser
makes the important point that in sociology, problem selection is "random
and atomistic" (Loubser 1976:80). Little, if any, systematic research has
been done on the question, although it doesn't seem a difficult topic to
study. 1 After some unsystematic data collection of my own, I have classi-
fied individuals' choices of research topic into four categories: two at the
individual level and two at the social/societal level. 2 Of these, two are
external to sociology and two are internal. What we have, in Table 2.1, is
a classic two-by-two table.
At the individual level, researchers often choose their topics because of
their own experience. For sociologists who have been through divorce,
loss of employment, or pressures toward conformity, these concerns can
play an important part in setting their research agenda.
Sociological elements can also affect choices at the individual level.
The department in which the sociologist studied, its leading lights, theo-

retical orientation, methods of choice, or the influence of the student's

dissertation adviser may set the budding sociologist off on a particular
subject matter. Neil Smelser notes that some sociologists remain with the
same research subject through their entire careers. Put that way, the
person's research agenda sounds waterlogged. But it is often also the case
that the first research study opens up a series of new questions, and the
sociologist pursues the questions that emerge from each successive study
down through the years.
At the social level, factors internal to the discipline of sociology can play
a part in choice of problem (see the lower right-hand cell in Table 2.1). For
example, certain topics may be preferred by prestigious journals, by commit-
tees that make awards, or through other recognition by the discipline. Even if
leaders of the discipline do not consciously seek to reward specific topics,
reigning preferences for particular theories and methods may benefit some
topics over others. Thus, for example, the prestige attendant on sophisticated
quantitative analysis has tended to elevate the standing of studies of occupa-
tional mobility, inter alia. Sociologists looking for recognition will often
respond to the perceived rewards of particular research topics. Furthermore,
topics in sociological currency may attract attention because of their intellec-
tual appeal to some members of the field.
The research agenda can be influenced, too, by social/societal factors
outside of sociology, such as availability of funding. When a sociologist
works for and receives funding from an organization, the research agenda
is set largely by the interests of the organization. Even for those who work
in academic departments or research institutes, research priorities of
funders can have an effect-particularly for those whose salaries depend
on "soft money."
Finally, sociologists-and my informal surveys suggest that this is a
large number-choose research topics on the basis of current social issues.
When gang crime or inadequate health services for the poor or the political
influence of large contributions to election campaigns engage societal
attention, sociologists undertake research on the subjects. They do so
because their interests and sympathies are engaged. Which brings us to the
fourth hypothesis.



This, it seems to me, may be the strongest explanation of the relation-

ship. Sociologists take up issues, like gender roles and homelessness,
because these are matters that concern us as human beings. From the time
Carol H. Weiss 33

of Auguste Comte and Herbert Spencer, sociology has been a science

directed not only at making sense of chaotic conditions but at improving
them as well. Many of us went into sociology in the first place because of
a reforntist impulse. Sociology held the promise of allowing us to improve
the human condition even as we studied it. We could follow our intellec-
tual bent and still believe that we were contributing to social reform. As
Rossi (1980) said in his ASA Presidential Address: "Our ranks always
have been full of ministers and ex-ministers, radicals and ex-radicals, even
a few conservatives and ex-conservatives, all of whom were attracted to
sociology because our discipline appeared to have some relevance to
social reform" (p. 889). Even those who want no truck with government
under any circumstances, and those who are out of sympathy with current
directions of public policy, generally respond to their perceptions of the
needs of the social world.
In recent years few sociologists have sought or been sought out in
corridors of power, but topics that cry out for change remain popular in
sociology. Sociologists have long been interested in issues of race and
persistent poverty, because the United States as a society faces severe and
continuing problems with these issues. Sociologists have become engaged
with issues of gender as the Women's Movement has made us more
sensitive to the necessity for change in the status of women. The economy
has moved to the forefront in sociology as issues of economic organization
have exercised the nation.
All of us respond to the world we live in. Even though many sociologists
seek to maintain a distance from the political realm, at the same time they
want their work to have some influence. And if sociological work is to
influence the course of events, some of that influence will have to flow, sooner
or later, through agencies of government. There are few other mechanisms
with the same potential for wide-scale impact on social issues.


All four explanations about the interaction of sociology and policy have
at least a modicum of support in the evidence. Nor do these possibilities
exhaust the possible links between sociology and public policy. Another
hypothesis, for which there is occasional support, is a two-step flow of
influence: Sociological ideas influence economists who then influence the
development of the policy agenda. Or sociological ideas and methods of
study influence how reporters report the news, and reporters' views influ-
ence the public policy agenda. There is empirical justification for this
hypothesis as well.

Which explanation has the greatest explanatory power? The answer seems
to be: It depends. It depends on the subspecialty within sociology. Criminol-
ogists, demographers, sociologists of education, and organizational sociolo-
gists have different experiences. It depends on the historical period. The
influence that sociologists had on public policy during the heady period of
the late 1960s and early 1970s was only a nostalgic memory during the 1980s.
It depends on the place. The effects of government policy on sociology in
some nations, such as the Soviet Union and China, have sometimes been
devastating. The influence of sociology on criminal justice and education in
California is not replicated in most other states. Different groups of sociolog-
ical workers, different periods, different places all make a difference.
It would be useful to specify the conditions under which one or another
explanation has greater support, that is, when public policy influences
sociological work, when sociology influences policy. Not enough empir-
ical research has been done to reach conclusions of this sort. Neverthe-
less, we can do some sensible speculation. As a preliminary attempt, these
are some of the conditions that I think may affect the interactions of
sociology and public policy.


Public policy affects the sociological agenda primarily when:

(a) large-scale funding is made available for research and such addi-
tional perquisites as training grants, fellowships, postdoctoral stipends,
development and maintenance of data bases, and so forth, in particular
substantive fields. The availability of money will exert a magnetic pull on
some segment of the fraternity/sorority, and when funds are concentrated
in one area, that area will rise to sociological prominence. In the late 1960s
and early 1970s, the study of poverty profited from an inflow of this sort.
(b) government grants are almost a monopoly of all research funds
available. The absence of alternatives will concentrate the influence of
funding for one substantive specialization.
(c) a relatively large proportion of sociologists is dependent on soft
money. A scarcity of other resources increases the demand for available
research support. Government money concentrated in one substantive
domain, monopolizing research options for a large group of hungry soci-
ologists, is almost bound to influence the sociological agenda.

Sociology affects public policy primarily when:

(a) creative sociologists work on issues relevant to present or potential
government action. A precondition for influence surely has to be that able
sociologists have something relevant to say.
Carol H. Weiss 35

(b) sociologists produce cogent findings and/or interesting ideas on important

topics about which there is uncertainty or conflict. The idea here is that not only
are the sociologists creative and the work relevant, but that policy actors are
willing to listen because of the significance of the issues and the extent of
uncertainty or confusion about what to do. In times of crisis, policy actors may
pay particular attention to outside ideas that offer possible solutions.
(c) there are intrinsic or extrinsic rewards for sociologists seeking to
influence policy, or at least few penalties. There should be something in
it for sociologists to encourage them to make their work known outside
the walls of academe. Otherwise, the best and the brightest are likely to
shun the second-class status of "applied work" and attend to well-re-
warded concerns of the discipline.
(d) mechanisms exist for diffusion of sociology to policy actors, and
sociologists know how and are willing to use them. Even able, motivated
sociologists with something to say to government will have a hard time
influencing policy unless there are structures and arrangements available
to transmit their messages. Academic journals do not reach policy actors.
They need to use such mechanisms as popular articles, op-ed pieces, news
reports, special conferences or individual contacts, or intermediary insti-
tutions that translate sociology into political parlance, such as government
offices of policy analysis.
(e) government staffs include people with sufficient training and knowl-
edge of the social sciences to be able to understand and translate sociolog-
ical data and concepts. Just as sociologists will not have much influence
unless they send their messages through appropriate channels, people at
the receiving end are not likely to be influenced unless they can decipher
the messages and put them to use.
When all, or most, of these conditions obtain, I hypothesize that sociol-
ogy is likely to affect the direction of public policy.

Sociology and public policy are separate and uninvolved when the
reverse conditions exist. Divergence is especially likely when:
(a) sociology concerns itself predominantly with internal discipline-ori-
ented issues, such as grand theory or methodology. These are yawns to
policy actors, even those with graduate degrees in the discipline.
(b) public policy is dominated by ideologues. Ideologues tend to believe
that they have all the answers and do not need advice or ideas-even, or
maybe especially, empirical evidence. The drastic cuts in social science
funding in the early years of the Reagan Administration apparently evolved
from this perspective. Ideologues also have little interest in trying to
influence what sociologists do.
(c) sociology is dominated by ideologues. Sociologists who are inter-
ested in being ideologically pure and true to the faith have little desire to

influence policy. Particularly if the prevailing ideology in sociology tends

toward the left and that of government veers to the right, few opportunities
will arise for either domain to influence the other.

Sociology and public policy each respond to the imperative oftime, tide,
and events:
(a) almost constantly.
(b) the trend is probably heightened when a sizable proportion of
sociologists think of themselves as part of the intelligentsia. Whether or
not they engage in public action, their image as intellectuals sensitizes
them to the political and cultural currents of the day, and affects their
research agenda.


Those are provisional speculations about conditions that affect the

intersection of sociology and public policy. My preference is for sociology
to be more actively engaged in research on contemporary topics of policy
relevance. I would like to see sociologists use their research as a basis for
taking a more active part in the discussions that go on not only in
Washington but in state capitals, where many important policy decisions
are being made these days. Duncan McCrae did a count a few years ago
of the number of ASA members who also belonged to the Association for
Public Policy Analysis and Management (APPAM), the professional as-
sociation for academics and practitioners who try to bridge the space
between social science research and public policy. Only 22 ASA members
were also members of APPAM, and most of those sociologists were on the
faculties of professional schools. That is not a very sensitive indicator of
the connectedness of sociologists to policy concerns, but it is a straw in
the wind. I think we should be more involved.
On the other hand, I am aware of the limitations of the discipline as provider
of political prescriptions. Our research gives us reasonable findings and
interesting ideas, but it does not often give us the grounds to make authorita-
tive prescriptions for public policy. Nevertheless, we have concepts and
insights to offer to political debates, sometimes better and wiser than the ideas
in currency, and I think it is sensible that we make them heard-provided we
do not claim more knowledge and certainty than we have.
Still, there are sound reasons for caution. As many social scientists have
written recently, sociologists who have entered into collaboration with
policymakers appear to have made a "Faustian bargain," mortgaging their
true task and heritage in search of political influence (Wittrock et al. 1991;
Carol H. Weiss 37

see also Heckman 1990). Not only have their prescriptions for government
often turned out to be wrong, but their involvement in policy issues
diverted them from the proper pursuit of knowledge.
Furthermore, as they found out, it is not easy to make the transition from
sociological research findings to policy advice. Few findings within
sociology receive consensual support; disagreements abound. The one-time
"epistemological optimism" that methodological advances would bring about
cognitive mastery of the social world (Wittrock et at. 1991) has foundered.
(For a provocative critique, see Lindblom 1987.) Some sociologists who
worked with policymakers, unable to find substantive consensus within
sociological communities, resorted to something akin to political compromise
to reconcile differences among sociologists-and between sociologists and
political actors. Other sociologists hewed to the line of their political convic-
tions and let research evidence take second place. Others caved in to prevail-
ing orthodoxies in the political arena and became little more than method-
ological technicians for existing government policies. It is not unreasonable
to wish to steer clear of these kinds of involvement.
But there is the possibility for engagement with policy action that is not so
close as to require compromise or collusion. There can be a vigorous,
independent, original, daring, critical sociology that provides insight and
criticism to government, a sociology that is captive to the assumptions of
neither the right nor the left. It is not a rarity. It is the type of sociology that
many sociologists practice. But what is often lacking are two critical ele-
ments: (1) sustained effort to reach policy audiences with sociological mes-
sages; that is, serious attention to dissemination, contact, and continuing
conversation with actors in the policy process; or the use of intermediary
institutions to undertake the task, and (2) recognition that sociology has more
to offer to policy audiences than validated data from well-designed experi-
ments and studies. Although good data are useful and build credibility,
equally important is the sociological perspective on entities, processes, and
events. Participants in the policy process can profit from an understanding of
the forces and currents that shape events, and from the structures of meaning
that sociologists derive from their theories and research.


1. One of our doctoral students, Suman Bhattacharjea, did a study of educational research
conducted in India to see to what extent it conformed with World-Systems analysis assumptions
that Third World research is dependent on the ideas, models, and values of the First World
(Bhattacharjea 1990).

2. I thank Dean Gerstein for his helpful comments on categories. He credits Loubser, but
I was unable to find a specific discussion in Loubser's excellent paper. What Loubser does
say is that the community of social scientists does not attend to problem identification. He
proposes that sociology should establish a set of human values as the ethical basis of the
discipline and its research agenda, and he offers four enduring values as the basis.


Aaron, Henry J. 1978. Politics and the Professors: The Great Society in Perspective.
Washington, DC: Brookings Institution.
- - - . 1989. "Politics and the Professors Revisited." American Economic Review 79(2): 1-
Barber, Bernard. 1987. Effective Social Science. New York: Russell Sage.
Bhattacharjea, Suman. 1990. "The Landscape of Educational Research in India: An Explor-
atory Study." Unpublished Qualifying Paper, Harvard University Graduate School of
Education, Cambridge, MA, May.
Caplan, N. 1982. "Social Research and Public Policy at the National Level." In Social Science
Research and Public Policy-Making: A Reappraisal. edited by D.B.P. Kallen, G. B.
Kosse, H. C. Wagenaar, J.J.J. Kloprogge, and M. Verbeck. Windsor, UK: NFER-Nel-
son Publishing.
Gans, Herbert J. 1979. Deciding What's News. New York: Pantheon.
Haveman, Robert. 1987. Poverty Policy and Poverty Research. Madison: University of
Wisconsin Press.
Heckman, James J. 1990. "Social Science Research and Policy: Review Essay." Journal of
Human Resources 25(2):297-304.
Janowitz, Morris. 1970. "Sociological Models and Social Policy." Pp. 243-59 in Political
Conflict: Essays in Political Sociology, edited by Morris Janowitz. Chicago: Quadrangle.
Keynes, John Maynard. 1936. The General Theory ofEmployment. Interest. and Money. New
York: Harcourt, Brace.
Kuhn, Thomas S. 1970. The Structure of Scientific Revolutions. 2d. ed. Chicago: University
of Chicago Press.
Lazarsfeld, Paul and Jeffrey Reitz. 1975. An Introduction to Applied Sociology. Amsterdam
and New York: Elsevier.
Lindblom, Charles E. 1987. "Alternatives to Validity." Knowledge: Creation. Diffusion.
Utilization 8(3):509-20.
Lindblom, Charles E. and David K. Cohen. 1979. Usable Knowledge: Social Science and
Social Problem Solving. New Haven, CT: Yale University Press.
Loubser, J. J. 1976. "The Values Problem in Social Science in Developmental Perspective."
Pp. 75-89 in Explorations in General Theory in Social Science. Vol I, edited by J. J.
Loubser, R. C. Baum, A. Effrat, and V. M. Lidz. New York: Free Press.
Melton, Gary B., ed. 1987. Reforming the Law: Impact of Child Development Research. New
York: Guilford.
Merton, Robert K., Leonard Broom, and Leonard S. Cottrell, Jr. 1959. Sociology Today:
Problems and Prospects. New York: Basic Books.
Nathan, Richard P. 1988. Social Science in Government: Uses and Misuses. New York: Basic
Orlans, Harold. 1973. Contracting for Knowledge. San Francisco: Jessey-Bass.
Reiss, A. E. 1961. Occupations and Social Status. Glencoe, IL: Free Press.
Carol H. Weiss 39

Rich, Robert F. 1981. Social Science Information and Public Policy Making. San Francisco:
Rossi, Peter H. 1980. "The Presidential Address: The Challenge and Opportunities of Applied
Social Research." American Sociological Review 45(December):889-904.
Schudson, Michael. 1978. Discovering the News. New York: Basic Books.
Scott, Robert A. and Arnold R. Shore. 1979. Why Sociology Does Not Apply: A Study of the
Use ofSociology in Public Policy. New York: Elsevier.
van de Vall, Mark. 1986. "Policy Research: An Analysis of Function and Structure." In The
Use and Abuse of Social Science, edited by Frank Heller. London: Sage.
Weiss, Carol H. 1977. "Research for Policy's Sake: The Enlightenment Function of Social
Science Research." Policy Analysis 3(4):531-45.
- - - . 1978. "Improving the Linkage Between Social Research and Public Policy." pp.
23-81 in Knowledge and Policy: The Uncertain Connection, edited by Laurence E.
Lynn, Jr. Washington, DC: National Academy of Sciences.
- - - . 1980. "Knowledge Creep and Decision Accretion.' Knowledge: Creation, Diffusion,
Utilization 1(3):381-404.
- - - . 1986. "The Circuitry of Enlightenment." Knowledge: Creation, Diffusion, Utiliza-
tion 8(December):2.
- - - . 1989. "Congressional Committees as Users of Analysis." Journal of Policy Analysis
and Management 8(3):411-31.
Weiss, Carol H. with M. J. Bucuvalas. 1980a. Social Science Research and Decision-Making.
New York: Columbia University Press.
Weiss, Carol H. and M. J. Bucuvalas. 1980b. "Truth Tests and Utility Tests: Decision
Makers' Frames of Reference for Social Science Research." American Sociological
Review 45(2):302-13.
Wilson, James Q. 1978. "Social Science and Public Policy: A Personal Note." In Knowledge
and Policy: The Uncertain Connection, edited by Laurence E. Lynn, Jr. Washington,
DC: National Academy of Sciences.
Wilson, William Julius. 1978. The Declining Significance of Race. Chicago: University of
Chicago Press.
- - - . 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy.
Chicago: University of Chicago Press.
Wittrock, Bjorn, Peter Wagner, and Helmut Wollmann. 1991. "Social Science and the
Modern State: Policy Knowledge and Political Institutions in Western Europe and the
United States." In Social Sciences and Modern States: National Experiences and
Theoretical Crossroads, edited by Peter Wagner, Carol Hirschon Weiss, Bjorn Witt-
rock, and Helmut Wollmann. Cambridge: Cambridge University Press.
How Do Issues Get on Public Policy Agendas?

John W. Kingdon

The purpose of this chapter is to provide some context to other studies of

agenda-setting. It discusses both a context of governmental institutions
and a context of theory within which one might better understand the
setting of public policy agendas. I come to this task not as a sociologist,
but as a political scientist who has spent the bulk of his career studying
governmental institutions and trying to understand legislative processes
and the formation of public policies by governmental officials.
Let's first understand how I will use words. I take a governmental agenda
to be a list of subjects or problems to which governmental officials and those
close to them are paying serious attention. So an agenda-setting process
narrows a list of conceivable subjects within any domain (e.g., within health,
transportation, or social welfare) to those that actually are the focus of
attention. Within agenda subjects, a second and quite different process nar-
rows a very large set of possible alternatives from which choices could be
made to the set of alternatives that are seriously considered. Then an author-
itative choice (e.g., a legislative enactment or presidential decision) is made
among the alternatives in the choice set. So agenda-setting and alternative
specification are not the same as final choices. We're talking, here, not about
how issues get decided, nor about how decisions are implemented and what
impacts they have, but rather how issues come to be issues in the first place.
This chapter has three parts. First, I discuss some of the major conclu-
sions that emerged from my larger study of agenda-setting in the federal
government (Kingdon 1984). Second, I consider the place of ideas (as
opposed to forces like interest group pressure or career election incen-
tives) in public policy formation. Third, I reflect a bit on the place of social
science in public policy.

John W. Kingdon 41


Incrementalism is probably the most common model of policy change,

and it does describe some processes well. But agenda change often does
not proceed incrementally, in small steps. Instead, issues "hit" suddenly.
There's a tremendous flurry of activity, and government policy changes
in major ways all at once. The New Deal, the Great Society, the Reagan
Revolution in the first 10 months of 1981, all illustrate this pattern of
spasms of reform interspersed with periods of rest.
A satisfying model of agenda-setting needs to comprehend sudden
change, to tolerate enormous complexity, and to model messy processes
in orderly ways. A good place to start is with the Cohen-March-Olsen
(1972) model of "organized anarchies," large, multipurpose, fragmented
entities like universities or the federal government. According to this
model, separate streams of problems, solutions, participants, and choice
opportunities run through such organizations, each stream with a life of
its own and largely unrelated to the others. People generate solutions
(proposals) whether or not they are solving problems, for instance, and
then look for problems to which to hook their solutions. Urban mass
transit, for example, has at various times been portrayed by its advocates
as a solution to the problems of traffic congestion, air pollution, and energy
shortages as each of these problems has heated up. Outcomes then depend
heavily on how these fairly separate streams are coupled-which solutions
get linked to which problems, which participants are present when deci-
sions are made, which problems and which solutions stay under consider-
ation when a choice opportunity presents itself.
In my revised version of the Cohen-March-Olsen model, there are three
streams: problems, proposals, and politics. People in and around government
recognize and come to concentrate on certain problems rather than others,
they propose and refine policy proposals (alternatives), and political events
like shifts in national mood, changes of administration, or interest group
campaigns move along on their own. These streams develop largely indepen-
dently of one another. Proposals are generated whether or not they are solving
a problem, problems are recognized whether or not there is a solution, and
political events move along according to their own dynamics.
At certain critical times the three streams come together, and the great-
est agenda change occurs. A problem is recognized, a solution is available,
and the political conditions are right. Advocates of proposals seize on
those times of opportunity (open policy windows)-such as changes of
administration, renewals of enabling legislation, or shifts in partisan or
ideological balances in Congress-to hook their solutions to problems that
seem pressing or to take advantage of propitious political happenings.

Something is done when the window is open, or the opportunity is lost and
advocates must wait for the next window to open.
I now discuss each of the three streams in turn, and then discuss their

Problem Recognition
Why do important people in and around government pay attention to
some problems and not to others? Fairly often, government simply moni-
tors the performance of systems through standard indicators like highway
deaths, cost of Medicare, commuter ridership, or disease rates. Sudden
shifts in these indicators or performance contrary to expectations produces
attention. At other times, it takes some sort of focusing event, a disaster
or crisis. Plane crashes focus attention on airline safety, for instance, in a
way that one-by-one automobile deaths don't do the same job for highway
safety, even though highway deaths are far more numerous.
There is a difference between a problem and a condition. As one of my
respondents said, "If you have only four fingers on one hand, that's not a
problem; that's a situation." So problem recognition is not simply a matter
of observing objective conditions; it involves interpretation as well. Con-
ditions become problems when we feel we should do something to change
them. How do we come to that sort of conclusion?
Conditions sometimes become problems when they conflict with prevail-
ing values. Uneven access to health care is or is not a problem, for instance,
depending on whether one thinks of health care as a right. Another fascinating
way for conditions to become problems is to place them in one category or
another. If transportation of the handicapped is a transportation problem, for
example, then low-cost and effective solutions like dial-a-ride or subsidized
taxi service are appropriate. But if transportation of the handicapped is
classified into the category of civil rights, then retrofitting subways for
elevators and providing bus lifts is warranted. The struggle for appropriate
categories makes all the difference in framing the issue.
Getting people to see a condition as a problem is a central political

The Policy Stream of Proposals

Picture a community of specialists in health, transportation, or any other
area: researchers, congressional staffers, bureaucrats in planning and eval-
uation offices, academics, and analysts working for interest groups. Ideas
float around in such communities. Specialists have their notions of future
directions and their specific proposals. They tryout and revise their ideas
John W. Kingdon 43

by going to lunch, attending conferences, circulating papers, holding hear-

ings, presenting testimony, publishing articles, and drafting legislative pro-
posals. Many, many ideas are considered at some point along the way.
The process by which proposals are selected from this very large set
resembles biological natural selection. Much as molecules floated around
in the "primeval soup" before life came into being, ideas float around in
a "policy primeval soup" in these communities of specialists. Both mole-
cules and ideas bump into one another and combine in various ways. In
each type of evolution, new elements are sometimes introduced (muta-
tion), and others are formed from previously existing elements (recombi-
nation). Thinking of policy change as recombination solves an intriguing
puzzle: How one finds sudden and substantial policy change at the same
time that there is "no new thing under the sun." There is change, but it's
the recombination of familiar elements.
The policy primeval soup takes a long time to bubble around. Develop-
ment of proposals must be done long before the opportunity for actual
adoption presents itself, because at that point, it's too late to hone a
proposal. Gradually, a policy community arrives at a "short list of ideas,"
a consensus at least on what the few live options are. In the process of
policy evolution, some ideas fall away, others survive and prosper, and
some are selected to become serious contenders for adoption.
By what criteria are some ideas discarded? First, the proposal must be
technically feasible, and specialists ask such questions as the following:
Will it work? Can it be drafted? Are the details ironed out? Second, there
must be an acceptable budgetary cost, and items that seem too expensive
are shunted aside. Third, proposals must be acceptable to the mass public
or to attentive publics. Some ideas are dropped because specialists think
they would run into a buzz saw, and it's not worth their trouble. The values
held in important publics constrain the choices that can be made in
specialized communities.

The Political Stream

Our third stream, the political stream, is composed of changes of admin-
istration, shifts in partisan or ideological balances in Congress, shifts in
national mood, partisan realignments, or interest group pressure. This
stream is an important endogenous part of policymaking, but it proceeds
according to its own dynamics, such as elections, nominations, partisan
politics, and interest group campaigns.
The following examples suggest the tremendous importance of this
stream. A widespread perception that the national mood favors govern-
mental activism at one point in time or favors smaller government at

another time, for instance, profoundly affects which items are possible and
which ones are not. Or the values of elected officials-the current presi-
dent and administration, or the distributions of values in Congress-are
absolutely critical to what can be pushed and what must wait. Or social
movements like the civil rights, consumer, environmental, or taxpayer
revolt movements sometimes sweep across our land bowling over every-
thing in their paths.

Joining the Streams

Advocates (policy entrepreneurs) continually push their pet proposals or
push attention to particular problems, and try to keep issues alive in good
times and bad. But an open policy window presents them with a special
opportunity. These entrepreneurs keep their proposals at the ready, and
become particularly active when a problem floats by to which their proposals
can be the solution, or when a development in the political stream can be used
to their advantage. At those critical points in time, they playa major part in
joining the previously separate streams, by hooking their solutions to prob-
lems, or by seeing that proposals from the policy stream are considered when
the political conditions are right. Much like a window for a space shot, policy
windows stay open only so long. If you don't seize your opportunity when
you have it, you must wait for the next one to come along. Thus Lyndon
Johnson told his major advisers that the election of 1964 opened a window
for Great Society proposals that would last only 2 years, and a tremendous
amount of major legislation was passed before it closed.
Generally, alternatives or policy proposals are generated in the policy
stream, and agenda change comes from changes in the problems or politics
streams. So windows sometimes open because problems become pressing.
The stock market crash of 1987 created a golden opportunity to work on
the federal budget deficit, for example, which as it happened was not
seized. Or windows open because of a change in the political stream. The
election of the Reagan administration, for instance, created many oppor-
tunities for the Heritage Foundation at the same time that it shut down any
chance for comprehensive national health insurance. Sometimes windows
open predictably, as with the scheduled renewal of enabling legislation.
At other times, they open unpredictably, as with an airline crash.
Events in the problems or political streams may set the governmental
agenda (the list of subjects to which governmental officials are paying
serious attention). But to make it beyond that to an agenda of active
decisions, all three streams must be joined. People recognize a problem, a
proposal is ready that can be related to that problem, and the political
conditions are right. If one of the three is missing, then the item's place
John W. Kingdon 45

on a decision agenda is probably quite fleeting. There is no single-factor

explanation for agenda change; several things come together at once.
One of my respondents, a lobbyist, captured the whole process in an
absolutely beautiful image. He showed how some of the process is gov-
erned by large events not under anybody's control, how people wait for
their opportunities, how they must be prepared ahead of time, and how the
joining of the streams is crucial. He said:

Whenyou lobby for something, whatyou haveto do is put together yourcoalition,

you have to gear up, you have to get your political forces in line, and then you sit
thereand wait for the fortuitous event. For example, peoplewho weretryingto do
something aboutregulation of railroads triedto ridetheenvironment for a while, but
that wavedidn't washthem in to shore. So they grabbed their surfboards and they
tried to ride something else, but that didn't do the job. The Penn Centralcollapse
wasthebig wavethatbrought themin.As I seeit, peoplewhoare tryingto advocate
change are like surfers waiting for the big wave. You get out there, you have to be
ready to go, you have to be readyto paddle. If you're not readyto paddlewhenthe
big wavecomesalong, you're not goingto rideit in.


The foregoing discussion of agenda-setting makes clear that policy

formation is not simply driven by such conventional political processes as
reelection incentives, interest group pressure, and marshaling votes and
power. Instead, argumentation, persuasion, and marshaling evidence and
information are also important. In other words, participants traffic in the
world of ideas (see Kingdon 1988, for a more complete statement). I take
"ideas" to mean either (1) goals or motivations other than self-interested
pursuits, or (2) theories that people hold about how the world works (e.g.,
about cause and effect relationships) that can be used to further self-inter-
est as well as other goals.
One clue to the importance of ideas is the investment in time, money,
and energy that people in and around government make in them. They hire
analysts; they marshal arguments and collect evidence in support of those
arguments; they reason and persuade. Even the most hard-bitten, self-in-
terested lobbies have policy analysts, and don't rely simply on appeal to
their self-interest or on their political muscle. If ideas weren't important,
savvy people like these, who know what is a productive use of their
resources like their time and money, wouldn't invest so much in them.
Much of current writing on ideas and politics takes the posture that ideas are
quite different from self-interest, and that the two also work at counterpurposes.

So both research and theory attempt to distinguish the effects of ideas from
the effects of self-interest. A major problem with this attempt, however,
may be that the two cannot be disentangled, either empirically or theoret-
ically. Empirically, the two work together. A reelection minded congress-
man, for instance, also believes in certain principles of good public policy,
and most of the time he takes both the expedient and principled course at
once. Theoretically, people attach meaning to their behavior, even their
self-interested behavior. Those meanings are not simply after-the-fact
rationalizations for actions, but affect those actions. In persuading others
and in explaining his votes, for example, a politician also persuades
himself and his explanations affect his decisions. So people often find it
impossible to pursue only their "raw" self-interest. They can't arrive at
positions without both self-interest and principle.
If my account of agenda-setting above is right, ideas don't drive changes
by themselves, because they must be coupled with more conventional
political forces, but ideas do have a considerable independent impact on
the outcomes. One account of the movement in the 1970s and 1980s
toward deregulation in many public policy domains, for example, empha-
sizes the sheer power of the idea (Derthick and Quirk 1985). Economists'
work on natural monopoly, capture, and the inefficiencies of regulation,
together with conservative ideology, powerfully affected public policy,
and extended into transportation, communication, banking, retailing, and
many other areas. My account of agenda-setting, although certainly em-
phasizing the importance of such ideas, also stresses the importance of
coupling those ideas in the policy stream with such events in the political
stream as the taxpayer revolt, the rebellion against "big government" as
represented by some aspects of the Great Society programs, the election
first of Jimmy Carter and then of Ronald Reagan, and the political impulse
to "get government off our backs." The ideas are themselves important;
but their combination with other forces is critical.
How can an idea be introduced into public policy discussions? My
description of agenda-setting above makes clear the importance of timing.
The window, when officials are receptive to a new idea, is open for a short
time and that opportunity must be seized then. This observation has two
implications. First, social scientists and others who wish to have their
ideas considered cannot wait to develop their ideas or proposals until the
opportunity arises, because they are too late at that point. A long period
of research, honing, and settling on proposals is needed well in advance
of the policy-making opportunity in order to be ready to take advantage
of it when it (sometimes unpredictably) presents itself. Both the soundness
of the basic research and the policy implications must be made clear well
in advance. Second, advocates including social scientists must be willing
John W. Kingdon 47

to invest time in their ideas. Just getting the attention of governmental

officials is a major accomplishment, even under the best of circumstances.
Using open windows takes skill, knowledge, and luck, to be sure. But of
all the attributes of successful policy entrepreneurs that I could name,
sheer persistence is probably the most important.


Frustrated social scientists sometimes ask why governmental officials

don't pay greater attention to social science than they do. We can't seem
to get them interested in certain subjects that seem obviously important to
us, and their priorities seem elsewhere. Or when they do get interested,
they take our proposals and use, change, and bend them to their own
purposes sometimes beyond recognition. Why can't we get subjects on
their agendas? Why do we lose control of the process when we are
successful in getting them to pay attention?
One perfectly reasonable answer is another question: Why should poli-
ticians and other policymakers pay attention to social scientists? After all,
we don't have the claim on them that their constituents ought to have in a
democratic system, and we don't enjoy the legitimacy that is conferred by
virtue of being elected or being directly responsible to elected officials.
As for controlling the course of our proposals once a policy-making
process takes them over, this inability to control the process is true of
everybody, even the president. Proposals will always be bent around to
suit the purposes of various decision makers,and bargains and compro-
mises will always be the order of the day. Furthermore, social scientists
often bring their own biases into their recommendations just like every-
body else does, and a political process must recognize that fact and apply
the appropriate discounts. In general, how are social scientists different
from anybody else? What special claim might they be able to make?
One such claim would be an appeal to expertise. If public policy needs to
deal with drug use, teen pregnancy, racism, unemployment, or a host of other
problems, then perhaps social scientists who have studied these things would
be a useful resource. But that claim turns critically on the existence of a solid
foundation of basic research in which the methodology is sound, the findings
are firmly established, and the interpretations of data are not subject to
dispute. It turns furthermore on the relevance of that research for practical
public policy problems. Social scientists need to ask themselves realistically
how often their expertise really is a reliable guide to public policy, how sure
they really are of their findings and interpretations, and how much the
purposes of public policy really can be served by the research. Without solid

claims to expertise, social scientists are just another interest group in the
political process, no more and no less important.
So let's think about the comparative advantages of social scientists.
What are we good at, or what can we become good at? Hints at possible
answers are found in the nature of policy-making discussed above. First,
it's important to the process of making public policy that problems are
recognized as problems. Social scientists can be very good at documenting
the existence, frequency, incidence, and intensity of a condition. We know
how to gather data about all manner of conditions-frequencies of various
crimes, demographic trends, disease and mortality rates, transportation
ridership, unemployment, income dynamics, social worker caseloads, and
government spending authorizations and outlays, to name just a few. We
can even make reasonable projections (often with a little less assurance)
into the future. So the monitoring of conditions is both important to
agenda-setting and comfortably within a social scientist's kit bag.
Transforming conditions into problems (see above) is more problem-
atic. In the course of monitoring conditions, social scientists often make
that transformation implicitly, either by juxtaposing the condition to a
widely held value or by framing the data in a category that heightens its
importance. But our professional ground for doing these things is more
shaky than for documenting the conditions themselves. We can document
a certain poverty or unemployment level in a population, for instance, and
we may believe that the level is inappropriate or intolerable. But it's really
up to a political process to establish how inappropriate or intolerable it is.
Social scientists can playa part in that process as well as anyone else, but
we have less special claim to expertise in the realm of judging the values
involved than we do in documenting the condition itself.
Second, policymakers are often very much in need of solid information
about how the world works. They particularly need to know, if they undertake
a certain kind of intervention, what will follow from it. Readers will recall
that this concern for how the world works or for cause and effect is the second
type of "idea" as defined above. Social science sometimes has good answers
to such questions. We might especially be able to show policymakers that the
world works in ways that might not have occurred to them. We have studied
certain processes in basic research thoroughly enough to claim reasonably
that we know what is going on. We may not be able to predict all of the effects
of an intervention with great assurance, or we may be able to anticipate some
consequences better than other consequences. But because of our base of
knowledge about how the world works, we may be able to make these cause
and effect connections better than other people can.
Of course, there's an obvious caution buried in this observation. Social
scientists' claim on the policy-making process is weaker when our know ledge
John W. Kingdon 49

is less solid than we would like it to be. It takes both an impressive fund
of basic empirical research and reliable theory to generate a high degree
of confidence in claims of cause and effect. And there is ample reason for
caution. If social scientists make claims to policymakers that cannot be
reasonably sustained, particularly in the name of social science, then the
credibility of the enterprise is threatened. We need good, solid research
and interpretations of data on which we can agree, in order to convince
policymakers that we should be heard on the grounds of our expertise.
Unfortunately, that's easier said than done. Social phenomena are dev-
ilishly complicated. It is inherently difficult to sort out cause and effect.
We can document the incidence of poverty, for instance, but are much less
able to agree on the importance of all of its various hypothesized causes.
This inherent complexity of social phenomena weakens our confidence in
various possible interventions, or at least it should. All of this reason for
caution argues for a tentativeness and an experimental approach to inter-
vention. Again unfortunately, tentativeness is exactly what the political
system does not tolerate very well.
Third, social scientists can often be involved in the process of generating
proposals that takes place in what I have labeled the policy stream. There
are entire communities of specialists on a whole series of social policy
problems, and social scientists are and should be prominent in those
communities. If a policymaker would want to design a new approach to
long-term health care, for instance, social scientists who specialize in the
economics, psychology, and sociology of aging would naturally be among
those to consult. Social scientists who are actively involved in this policy
stream, however, quickly discover that their knowledge must be adapted
to the practical problems of policymakers to be useful. In other words,
policy proposals need to meet the criteria of technical feasibility, budget-
ary acceptability, and value acceptability (see above) with which practical
people wrestle continually.
One development within the policy stream to which social scientists
might contribute as much as anybody else is a gradual reorientation of
world views or categories within which people think. Within the commu-
nity of transportation specialists, for instance, it makes a tremendous
difference whether people think of transportation as naturally market-
driven or as something that needs central planning. Or within the health
insurance policy community, there's a major difference between using
insurance simply as a device for paying the bills and using insurance as an
instrument for redirecting incentives in health care delivery. Processes of
highlighting some problems rather than others or of demonstrating the
consequences of various alternatives, among the processes in which social
scientists can participate, sometimes contribute to reorienting whole

approaches within policy communities. When the problem of Medicare

costs became unmistakably severe, for instance, policymakers changed
their orientation toward insurance from simply paying the bills to creating
cost-saving incentives.


We often wonder how a given group can get something on a governmen-

tal agenda. For that matter, how might social scientists do SO? An impor-
tant part of an answer must be that there are very few actors who can "get"
something on an agenda. Persistence is important, of course, as is expertise
and cultivating the right connections. But being heard in the policy process
is often more a matter of being positioned to take advantage of opportuni-
ties when they arise, than it is doing a set of things fully under one's
control. And realistically, even the most knowledgeable, savvy, and influ-
ential activists lose some of the time. Both researchers on and activists in
the setting of policy agendas come to appreciate the complexity of the
processes, the inability fully to predict courses they will take, and the
frustrations of attempting to control them completely.


Cohen, Michael, James March, and Johan Olsen. 1972. "A Garbage Can Model of Organiza-
tional Choice." Administrative Science Quarterly 17:1-25.
Derthick, Martha, and Paul Quirk. 1985. The Politics of Deregulation. Washington, DC:
Brookings Institution.
Kingdon, John W. 1984. Agendas, Alternatives. and Public Policies. Boston: Little, Brown.
---.1988. "Ideas, Politics, and Public Policies." Paper delivered at the Annual Meeting
of the American Political Science Association.
The Powers and the Intellectuals:
Benchmark Texts and Changing Conditions

Steven Brint



Thirty years ago, social scientists and historians were in the midst of a
particularly active-and, in many respects, particularly productive-phase of
thinking about the relations between people of ideas and people of power.
The work that has come since has often been derivative, when it has not
been a step backwards, and, consequently, the period I have in mind
represents a still-relevant benchmark. I want to discuss the assumptions
and conclusions of that period at the beginning of this chapter, because
my work is intertwined with those assumptions and conclusions, even
where it disagrees and even where it urges a change of perspective.
The period is bounded by Edward Shils's "The Intellectuals and the
Powers" published in 1958 and by Lewis Coser's Men of Ideas, published
in 1965. In its span, it includes-to note only works by Americans-Henry
Kissinger's "The Policy-Maker and the Intellectual" (1959), Seymour
Martin Lipset's "American Intellectuals: Their Politics and Status" (1960,
chap. 10 of Political Man), and Richard Hofstadter's (1963) Anti-Intellec-
tualism in American Life. If we stretch the periodization a bit, we can also
include in its span the influential broadsides by C. Wright Mills (1951) on
"Brains, Inc" and by Lionel Trilling (1965), from the opposite direction,
on the "adversary culture" of intellectuals, as well as several other signif-
icant works of scholarship and commentary.'
These writings are particularly relevant to the situation of intellectuals
in the United States, and, in so far as they are generalizable, to the situation

of intellectuals in the liberal democracies. These types of societies will

also provide the focus for my remarks.

Concepts and Findings

There are, I think, several still-relevant conceptions and findings in the
writings of the midcentury Americans. How relevant they are, and in what
ways they are relevant, will be a matter of dispute--one that I will address
in this chapter. But anyone who is willing to keep an open mind about the
writers and their works will, I think, recognize several foundational ele-
ments on which others have built and will continue to build.
First, as a preliminary matter, there is the critical distinction between
intelligence and intellect. It is a distinction sensed perhaps by everyone,
but it was recognized as essential, and delineated with characteristic
penetration, for the first time by Hofstadter. Intelligence, he writes, is "an
excellence of mind ... employed within a fairly narrow, immediate and
predictable range; it is a manipulative, adjustive, unfailingly practical
quality." Intellect, on the other hand, is "the critical, creative, contempla-
tive side of mind" and has many of the opposite characteristics. "Whereas
intelligence seeks to grasp, manipulate, re-order, adjust, intellect exam-
ines, ponders, wonders, theorizes, criticizes, imagines. Intelligence will
seize the immediate meaning in a situation and evaluate it. Intellect
evaluates evaluations, and looks for the meanings of situations as a whole"
(Hofstadter 1963, p. 25).2
This definition, which focuses so completely on qualities of mind, did
not command full agreement. Some argued that intellectuals ought to be
defined in the conventional way: not simply by their qualities of mind, but
also by the scope of their interests and, in particular, by their "pronounced
concern" with moral and political theories and ideals. 3 This emphasis is
apparent in Coser's work, for example: "Intellectuals," he writes, "exhibit
in their activities a pronounced concern with the core values of society.
They are the men who seek to provide moral standards and meaningful
general symbols" (Coser 1965, p. viii)." These two definitions continue to
mark the sweep of sociological discussions of intellectuals.
There is a second still-relevant conception, I think, in the efforts of these
midcentury writers to order the types of relations found between people
of ideas and people of power. Coser's (1965) typology from part two of
Men of Ideas is particularly noteworthy. Many intellectuals, Coser ac-
knowledges, are disconnected from political life, except in so far as they
accept their minimal civic obligations. Of those who are in some fashion
politically active, however, there are four typical situations. (1) Intellec-
tuals sometimes, though rarely, hold power themselves, following periods
Steven Brint 53

of revolutionary change, as was true of the American Founding Fathers, the

Jacobins, and the Bolsheviks. (2) They more often work with greater or lesser
success as influential policy-shaping powers behind the throne, as has been
true of such recognizable groups as the Fabian socialists, the Swedish Social
Democrats of the Myrdal generation, the Brain Trusters during the New Deal,
the so-called wise men of American Cold War foreign policy, as well as
countless individuals throughout the modem era. (3) They may work to
legitimate (or undermine) power with their visions of the proper order of
society and their evaluations of the fidelity of the current ruling elites to that
proper order. (4) And, finally, in what is surely one of the more common
situations, they may protest power on the basis of their "pronounced concern"
with moral standards and the core values of society.
There is a third still-relevant contribution in the fair-minded and empir-
ically based consideration of the political orientations of intellectuals.
There have been many in the past and there are many today who see
intellectuals as a chronically discontented lot. The midcentury Americans,
by contrast, took the view that, within the intellectual community, many
varieties of political orientation could be found. A characteristic statement
is that of Shils: "Moderates and partisans in civil politics, quiet apolitical
concentration on specialized intellectual tasks, cynical anti-political pas-
sivity, and faithful acceptance (of) and service (to) the existing order are
all to be found in substantial proportion among modern intellectuals, as
among intellectuals in antiquity" (Shils 1958, p. 13).
If intellectuals were not inevitably alienated, they were, however, at
least very often at odds with their societies and with the elites of their
societies. Lipset's work, in particular, documents, with characteristic
subtlety and ingenuity, the association of 20th century American intellec-
tuals and left-of-center political views (see Lipset 1960, chap. 10; and,
later, Lipset and Dobson 1972). For the first time, empirical evidence
replaces anecdote, and we find that American academics, writers, artists,
and journalists incline toward the political left, and, furthermore, that the
more distinguished people in these professions are often the most critical.
There is a great deal of speculation about why this should be so. Intellectual
judgments and intellectual commitments were very often stressed. Coser (1965)
argues, very much in the line of Schumpeter (1942), but with a more positive
tone, that the very qualities of mind that make people intellectuals also dispose
them to be dissatisfied with uncritical acceptance of "mere" custom, utility, and
factuality. Hofstadter (1963) emphasized the intellectuals' reaction to the nar-
row-mindedness and rigidity ofthe dominant culture. Shils (1958) discussed the
"oppositional" traditions of intellectual life-romanticism, populism, scien-
tism, and millenarianism-which can be interpreted as a sort of functional
equivalent to the "union consciousness" of discontented workers.f

There is, finally, a fourth still-relevant contribution: the beginning-it

is very much just a beginning-of comparative and historical study. There
is specifically the thesis, endorsed by Shils (1955) and Lipset (1960, pp.
403-17), among others, of a worldwide trend among intellectuals away
from revolutionary goals and toward pragmatic reform, the so-called
end-of-ideology thesis, which perhaps ought to have been called the
"end-of-radical-ideology" thesis. And there is also the sense-never very
fully worked out-that the remaining national differences in the politics
of intellectuals ought to be explained by the specific social situation of
intellectuals in particular countries, or by dominant national cultural
traditions. Following Annan (1955) and others, Coser (1965), for example,
attributes the centrism of British intellectuals to the small size of the
British intellectual class and to its close association with future members
of the British elite at Oxford and Cambridge. Characteristic outlooks of
French and American intellectuals were attributed, in like fashion, to the
abiding influence of Jacobin versus pragmatic/utilitarian cultural tradi-
tions, as these developed out of social structural differences in the two
countries at the time of their political revolutions (Lipset 1963).

Mood and Assumptions

One of the difficulties of social analysis is that the best of it as much as
the worst can be colored by historical conditions that even the most
self-conscious writers may not at the time fully appreciate. Consequently,
the task of succeeding generations is not just to appreciate contributions
and to correct errors-as one version of social science cumulation would
have it-but also to appreciate, in the manner of an intellectual historian,
the conditions that influence the direction ideas take and the kinds of
assumptions that underlie them.
In this vein, it is the confident tone and secure sense of the intellectual
community that most strikes me in reviewing these works. The writings
are marked by a tone of stoic equipoise. They show a good deal of
awareness of the problems intellectuals face in steering a course between
too much closeness to power and too much distance from power. In the
words of Kissinger, "[The intellectual] must steer between the Scylla of
letting the bureaucracy prescribe what is relevant or useful and the Charybdis
of defining these criteria too abstractly. If he inclines too much toward the
former, he will turn into a promoter of technical remedies; if he chooses
the latter, he will run the risk of confusing dogmatism with morality"
(Kissinger 1959, p. 35).
And yet there is little doubt among these writers that many intellectuals,
like latter-day Odysseuses, will have the skill and ingenuity to navigate in
Steven Brint 55

these troublesome waters. The stoic tone does not efface-and in fact rather
enhances-the confidence these writings convey in the possibilities for intel-
lectual independence, for intellectual influence, and also in the potential for
coherence and self-consciousness in the intellectual community.
The potential independence of the intellectual community is assumed
virtually without question. For Kissinger, who had none of Weber's faith
in political heads of bureaucratic agencies, intellectuals represented the
best hope of infusing "purpose and vision" into the technical machinery
of bureaucracy. Even in the (slightly later) work of Talcott Parsons, a
theorist normally sensitive to constraints and interdependencies, the intel-
lectual "classes" are depicted as bidding in an entirely independent way
"to have an impact on the affairs of society commensurate with or perhaps
running somewhat ahead of their actual position of strategic importance
in it" (Parsons 1969, p. 22).
As for influence, Coser's typology makes evident the range and variety
of types of influence that intellectuals can have, and it is clear both in his
work and in other writings from the period that this influence has often
been decisive. One conclusion of Shils's, for example, is that modern
politics-the politics that commences with the constitutionalism of the
18th century-is very much a politics in which intellectuals have played
important roles. "Modern liberal and constitutional politics were largely
the creation of intellectuals with bourgeois affinities and sympathies,"
(Shils 1958, p. 12) and revolutionary politics were similarly very much a
product of dissenting intellectuals, usually, as Shils writes elsewhere,
those "not yet assimilated into intellectual-practical occupations, bohem-
ian free-lance intellectuals ... and occasionally already well-established
persons with unusually sensitive moral consciences" (Shils 1968, p. 413).
It is the sense of coherence and community that is so particularly
striking. Parsons's discussion of ideology as the primary instrument of the
"modern secular intellectual classes" is not at all uncharacteristic. These
midcentury Americans, although they often did not say it directly, clearly
thought of intellectuals as a distinct class in modern societies. 6 The
writers, moreover, found it not at all incongruous to address the entire
community of thinkers, as when Hofstadter sharply questions what he calls
the radical intellectuals' "fetishism of alienation":

to identify the maximum degree of alienation with the optimum exercise of

intellect and to prescribe it as a kind of moral obligation for intellectuals is, in
effect, to prescribe a single intellectual style for the entire community of think-
ers.... It would be tragic if all intellectuals [determined] to serve power, but it
would be equally tragic if all intellectuals who became associated with power
were driven to believe they no longer had any connection with the intellectual

community: their conclusion would almost inevitably be that their responsibili-

ties are to power alone. (Hofstadter 1963, p. 395)


Since the last of these works was published, great events have occurred
that have transformed our sense of the intellectual community, and, in-
deed, have raised questions about the very existence of such a community.
Ideologically, these events include the brief but intense challenge of "the
New Left," the waning of liberalism and social democracy, the rise of new
conservative ideologies, and perhaps most importantly the crumbling of
Marxist-Leninism and the state-based forms of socialism in the Soviet
Union and Eastern Europe. Great structural events have also occurred that
bear on our understanding of the relations of intellectuals and the state.
These include the ever-increasing capabilities of organized power centers,
such as political parties, state bureaucracies, mass media organizations,
and universities; and the now very large number of highly educated
professionals who pour out of our universities each year and into govern-
ment, or government-related activities.
In this context, a revaluation of these midcentury writers necessarily
leads in somewhat contradictory directions. On one side, the force of some
of their conceptions seems to me to remain essentially undiminished. On
the other, the ideological and structural changes mentioned above-and
our own changed mood and assumptions-do suggest the need for some
significant revisions. These revisions are based, for the most part, on a
greater sense of modesty about the powers of intellectuals. It is more
difficult now to make a case for the notion that the influence of intellec-
tuals is based primarily on the intrinsic qualities of the intellectual com-
munity-be they qualities of mind, or distinctive resources and interests.
Rather, the qualities of the situations in which intellectuals find them-
selves and the quality of their ties to other groups and organizations now
seem more often decisive. 7

The Influence of Intellectuals

It requires some digging to see this, because surface impressions do not
suggest that intellectuals have become less important politically. At least
they do not suggest this about the leading stratum of intellectuals. If
anything, the opposite would appear to be true. In 1960, intellectuals were
the decisive political class nowhere in the developed world. With the
collapse of Communism in Eastern Europe, intellectuals have come to
Steven Brint 57

power, at least temporarily, in Czechoslovakia, Hungary, and Poland.

There is, similarly, no escaping the sense that political ideology has
become a formidable power, not only in the post-Communist world, where
such a thing is to be expected, but also in the West where new conservative
ideologies have provided an important resource and support for new
conservative administrations. Indeed, when looking over the role that men
of such different political temperaments as Vaclav Havel and Irving
Kristol have played in recent years, it is certainly possible to argue that
ours has been an era of great triumph, on a grand scale, for ideologists-a
term I do not use pejoratively, but rather simply to identify those whose
primary concern is to project a philosophically grounded vision of the
proper order of society and the relation of ruling elites to that order.f
Nor do the memoirs of those who served in the recent administrations in
this (see, e.g., Stockman 1986; Anderson 1988; and Noonan 1990) and other
Western countries (see, e.g., Aberbach et al. 1981; Suleiman 1984; Heclo and
Madsen 1987; Wright 1989) lead us to conclude that policy intellectuals have
experienced any obvious decline in prominence or influence. To choose an
example that is close at hand: the memoirs of Reagan Administration staff
encourage very much the opposite impression.'
And there is, finally, another group of political intellectuals-left almost
entirely out of Coser' s scheme 10-that shows no signs of diminishing influ-
ence. These are the academic intellectual-experts, whose political signifi-
cance derives principally from their ability to influence the climate of edu-
cated opinion. Typically, the scope of these people is narrower than that of
the more ideologically oriented intellectuals, and less directly involved with
the nuts and bolts of governmental planning than that of the policy intellec-
tuals. They are, for the most part, interested in particular issues, or sets of
related issues, rather than in broad questions about the proper order of society
and government generally. They write, for example, about the current situa-
tion of woinen in the work force and at home, or of affirmative action, or of
the state of the educational system, or of the sources of economic decline
among nations. In these writings, academic intellectual-experts mix norma-
tive judgments with scholarly considerations of all of the diverse phenomena
bearing on their subject-from legal opinions, to social changes, to research
findings, to past intellectual arguments.
The more popular of these academic intellectual-experts are seen and
read by large numbers of people in their capacity as commentators on
current events. Often these commentaries have at best a rather superficial
and ephemeral impact. But, at the more scholarly end of the spectrum, the
impact of academic intellectual-experts may be both more durable and
more profound. If one thinks, to choose a few recent American examples,
about the extent to which Hochschild's (1989) "second shift" argument

has entered into recent discussions of women's issues, or how Glazer's

(1975) critique of "reverse discrimination" helped mobilize reactions to
affirmative action policies, or how Hirsch's (1987) advocacy of "cultural
literacy" helped to recast discussions of education in many areas of the
country, or the extent to which Kennedy's (1987) ideas have at once
stimulated and framed new thinking about the global situation of the
United States, it seems clear that the leading figures among the academic
intellectuals also constitute a rather important group, albeit one whose
precise influence would be difficult to measure.
And yet, in spite of this evidence, many people-and I include myself
in this group-would be inclined to turn a skeptical eye on those who speak
of the increasing influence of intellectuals and would want to emphasize
instead the dependence of intellectuals on forces greater and more power-
ful than the intellectual community itself. Indeed, the very notion of a
coherent intellectual community seems increasingly dubious at a time
when so many who are at least nominally tied to that community experi-
ence a profound sense of disagreement within and disengagement from the
regulation of that community.

Sources of Dependence
Let us look at the matter of dependence first. We can do this by
examining the fate of several of the categories of politically involved
intellectuals identified by Coser, beginning with the ideological intellec-
tuals-those who provide visions of the proper order of society that help
to legitimate or to delegitimate ruling elites. If we evaluate the situation
unsentimentally, we can see how much of the current influence of these
ideological intellectuals depends on forces that have created repeated
disruptions in the fiscal and political health of states.
A crisis of the state is also a crisis of legitimation. This is perhaps obvious
in the case of Eastern Europe, where the faltering fiscal and political health
of the state created a legitimation crisis of obvious profundity. But it is also
true in the West. Indeed, throughout the West, the dominant left-of-center,
egalitarian ideology of intellectuals suffered a tremendous shock in the early
1970s when the logic of Keynsianism came unstuck and increases in govern-
ment services started to be perceived-fairly or not-as greatly interfering
with, rather than enhancing, economic development.
Indeed, I think we can go further and say that in the liberal democracies,
at any rate, a stable ideology requires a reliably supportive political
coalition, and fiscal and distributive rules that appear to work in the
interests of both social and economic advancement. In the United States,
the "New Deal" coalition; the fiscal rules of Keynsianism; and the distrib-
Steven Brint 59

utive rules of an interest-group-based pluralism provided the necessary

foundation for the dominant liberal ideology among intellectuals. Though
the elements were never exactly the same, similarly stable rules and
conditions existed in many parts of Europe during the same period and
provided the same sort of foundations for the prevailing social democratic
views among intellectuals. It is possible to see now that the association
between intellectuals and the liberal left in the postwar era was supported
less by the so-called adversary culture of intellectuals than by effective
fiscal and distributive practices of the state that held together coalitions
supporting the growth of the public economy (see also, e.g., Cameron
1982; and Keohane 1984).
These conditions no longer exist. Nor have they been replaced, in many
countries, by stable configurations of another sort. This enhances the
importance of ideologists, and it also reduces the level of consensus that
ought to be expected among intellectuals in their ideological beliefs. At
the more elite levels, conservative ideologies have clearly flourished in
recent years. But, unless fiscal and political conditions stabilize, there is
no certainty that this will continue.
Other forces are also involved in the growing importance of ideology at
a somewhat less elevated level. One is the mass media. Many educated
people seem to feel most comfortable when they have a morally satisfying
vision of the proper order of society on which they can rely. The mass
media are, consequently, at least partly in the business of servicing these
needs for satisfying general pictures of proper social organization and
satisfying pictures of threats to social organization. They are, therefore,
inclined to pay serious attention to those who are willing to comment on
the various visions of proper social organization that are competing at any
given time. For very similar reasons, the political parties also have devel-
oped a strong functional need for ideologists, who can serve up appealing
images of threat and proper order.
To some degree, in a competitive situation, there is a self-reproducing
aspect in the success of ideology. It is often said that the bureaucratization
of one group in a conflict forces the bureaucratization of the other group.
In similar fashion, the success of ideology on one side of a political debate
encourages the development of a counterideology on the other side. This
sense of the needs of combat has certainly been an important force in the
development of the new conservative ideologies that arose in the 1970s.
As one writer close to the neoconservative movement in the United States
put it not too long ago: "Intellectual attacks on business require intellec-
tual apologetics" (Bruce-Briggs 1979, p. 207).
In the United States, confidence in the independence of another cate-
gory, the policy intellectuals, has, I think, faltered since the last of the

wntmgs of the midcentury writers, because of the highly publicized

involvement of policy intellectuals in the fabrications surrounding the war
in Southeast Asia and the economic forecasting of the Reagan administra-
tion. It is difficult in the wake of these events to feel much confidence in
the ability of policy intellectuals to offer the large vision and purpose that
Kissinger so hopefully expected in the face of the bureaucratic pressures
encouraging administrative loyalty.
Instead, policy intellectuals now seem more completely than before crea-
tures of the state-and perhaps also of rather self-contained policy commu-
nities closely tied to the state. The ties of policy intellectuals to the broader
intellectual community are, frequently enough, very tenuous indeed. Under
these circumstances it is usually too much to expect that their primary
affiliations will be other than to the political leaders who invite them into the
councils of government. Kissinger was astute in pointing out the potential
problems faced by policy intellectuals, but he underestimated their force. II
Of course, there is nothing wrong with helping to implement state
policies when those policies are wise and beneficial. The difficulty, which
is hard to identify outside of the obvious and notorious cases, is that
intellectuals in state employment often find themselves supporting estab-
lished state bureaucratic interests over more viable alternative conceptions
of the public interest, and that they often lose their capacity to criticize
and reconsider the policy decisions of their superiors. Thus, except under
extremely rare circumstances, policy intellectuals are better characterized
as state experts or state managers.
In many respects, it is the academic intellectual-experts who are now
best able to occupy a responsible, relatively independent position. But
even they are influenced, in ways that we can only begin to understand,
by their primary ties-to the changing moods of the educated classes and
of the cultural intermediaries who provide intellectual fare for those
classes. A good argument can be made that changing attitudes call forth
intellectual responses every bit as much as intellectual conceptions influ-
ence opinions. The history of thinking about "affirmative action" policies
to combat the long history of racial exclusion in this country provides one
compelling recent example. The notion of "reverse discrimination," pop-
ularized by Professor Glazer, among others, almost certainly had an
independent effect on the rhetoric and thinking surrounding affirmative
action at least in some quarters of the educated public. Yet isn't it at least
equally likely that the force of the argument was due to a degree of
preexisting change in attitudes?12
However one judges the relative importance of audiences and the ideas
intellectuals address to those audiences, it is obviously wrong to think that
academic intellectual-experts play an entirely independent, creative role.
Steven Brint 61

Instead, they stand in relation to a mix of changing facts-whether these be

increasing homicides, larger trade deficits, declining test scores, longer work
weeks-the selective spotlighting of these facts, and the development of
feelings of threat and desire in relation to them. There is, I should add, nothing
necessarily wrong with the dependence of intellectuals on a market for their
goods, nor do I mean to suggest that market success is uppermost in the minds
of many intellectuals. I wish only to point out that a kind of dependence exists,
even in this relatively independent quarter of the intellectual stratum.
What we have, then, is not only a set of constraining and enabling
situations, but also a set of fateful institutional ties. Ideological intellec-
tuals are closely connected to-and also, in many respects, clearly depen-
dent on-the mass media, the political parties, and the intellectual journals
of opinion. Similarly, policy intellectuals are connected to, and dependent
on, state bureaucracies and the political heads ofthose bureaucracies. And
so too academic intellectual-experts are connected to, and dependent on,
the prevailing climate of opinion in the educated classes, which is the basis
of their market.

Sources of Incoherence
One source of the current incoherence. in the intellectual community
stems from the sometimes contradictory directions in which the different
segments ofthe politically involved intellectual stratum are pulled by their
primary institutional ties. Because the interests of the political parties, the
state bureaucracies, and the educated public are often rather different, if
not explicitly opposed, there is always the possibility of a strong sense of
disconnection between ideological intellectuals, policy intellectuals, and
academic intellectual-experts. As employers of ideologists, the parties and
the mass media require, above all, vivid images and bold justifications of
general principles. (The leading intellectual journals of opinion require,
in addition, the seriousness of political and social philosophy.) The bu-
reaucracies most often require skillful plans to meet strategic needs of the
future and specific issues of the moment. And the educated public wishes
to ponder, with the guidance of a fresh, identifiable argument, contempo-
rary issues that fall between the extremes of generality and specificity.
To the extent that coherence exists between the major segments of the
politically active intellectual stratum in the West, it is unquestionably
more on the right nowadays than on the left. But even on the right
coherence is fragile and unstable. This is less because intellectuals are
given to fractiousness-although, of course, they are-than because the
institutional situation in which each segment works often encourages
contradictory outcomes.

Most notoriously, the actual strategies pursued by policy intellectuals and

the images used by sympathetic ideologists to legitimate policy can lead in
different directions. An image of toughness on terrorists, for example, may
run into conflict with policy efforts to encourage a more stable future in an
unstable region. Even where ideology and policy are at least generally in
harmony, however, tensions frequently exist. One classic tension is between
the idea-driven boldness of ideologists and the political realism and caution
of many policy intellectuals. 13 Still more seriously, the actual effects of policy
may at times undermine the promises of ideology. Thus the image of general
prosperity through tax cuts and deregulation cultivated by conservative ide-
ologists in the years preceding and during the Reagan Administration now
runs into conflict with some of the actual results of these policies, including
a still-growing deficit, a significantly larger income and wealth gap between
rich and poor, and a scandal of unprecedented proportions in at least one of
the deregulated industries.
Apart from the divisive potential of institutional differentiation in the
absence of strong countermanding ties, there are some further reasons for
the current incoherence of the intellectual community. The most obvious
is that our era has seen a tremendous intensification of ideological conflict
within the intellectual community-first arising from the challenge of the
antiwar movement and counterculture of the 1960s and then from the
faltering fiscal and political health of the Western states in the mid-1970s,
combined with the mobilization of conservative elites to meet those crises.
Apart from demonstrating in an incontestable way that no inevitable
connection exists between modern secular intellectuals, and antibourgeois,
left-of-center politics.i" the most recent episodes of intense ideological
conflict have also, for obvious reasons, damaged the unanimity of the
intellectual community. In one sense, this is a price we might gladly pay
for the potential revitalization that ideological conflict encourages. But,
practically speaking, years of discord have led to the loss of a secure center
and have deeply affected the ability of the intellectual community as a
whole to stand as a recognizable center of gravity on matters of import.
These years of discord may also have injured public regard for intellec-
tuals. It is ideological intellectuals who are most willing to embrace the
label "intellectual," and, under the circumstances, this is rather unfortun-
ate. Although the conflicts of social philosophy can certainly be of the
greatest possible cultural significance, they are also prone-particularly
during periods of intense conflict-to degenerate into ill-tempered polem-
ics, overly general blanket condemnations, and a rigid unwillingness to
rethink positions. Thus the characteristic fighting stance of self-proclaimed
intellectuals, as it appears to the public mind, often lacks precisely the
farsighted, contemplative qualities that ought ideally to be associated with
Steven Brint 63

intellect itself. It is difficult to hold intellectuals who seem intent on

subverting intellect in high esteem.
In fact, many-even many in the university-do not hold intellectuals
in high esteem. This brings us to the last major source of the incoherence
we find in the intellectual community-the conflict between "intellectu-
als" and another of the powers, the academic "professionals." This conflict
is evident in sociology, but it is also evident in many other fields in the
social sciences and humanities. One recent account (Jacoby 1987) has
gone so far as to argue that what the author perceives as a general decline
of intellectual life in America can be traced to the incorporation of
intellectuals into the universities, combined with the growing insistence
by universities on a strict and narrow form of professional training.
There is another possibility, however; that the conflict is based largely
on a common misunderstanding of what intellectual activity involves. It
seems to me likely that most professionals involved in the conflict share
the culture's largely negative conception of intellectuals as self-righteous
partisans of potentially one-sided moral and political theories. It is not
surprising that many professionals define intellectuals in this way, since,
as I indicated, many people who think of themselves as intellectuals also
define themselves in this way. The widespread acceptance of a definition
of the intellectual as political and moral idealist critic has had, among its
consequences, that of encouraging potential and would-be intellectuals to
overemphasize their sensitivity to general moral standards and the grand-
est sort of theories and to underappreciate other qualities of intellect and
other forms of intellectual activity.
Under the circumstances, a reconsideration of the connection between
intellectuals and professionals might do both sides good. If professionals
associated intellectuals with intellect, rather than with either intelligence
or ideological criticism, they would perhaps be less prone to dismiss the
importance of intellectuals. If would-be intellectuals understood that ideo-
logical forms of thought are only one part-and indeed sometimes a rather
small part-of the preoccupations of intellectuals, they would perhaps be
less inclined to enclose themselves in the chrysalis of political and moral
theory. They might also be more inclined to embrace professional skills
and knowledge as part of the tool kit with which they go about doing their
work in the world.
Unfortunately, clearing up the conflict is not simply a matter of clarify-
ing the meaning of the intellectual "role." In one important sense, the
analysis of Jacoby (1987) is correct. In the universities, the "specialist type"
of academic professional is in fact frequently ascendant over the "generalist
type." In this sense, the grounds for status conflict are always abundantly
present. If the status tensions between professionals and intellectuals have

been aggravated in recent years, it is perhaps not so much because of

declining alternative employments for intellectuals (were they ever great?
have they really diminished?), as to the sheer growth of the professoriate
and the consequent increasing relative deprivation of all involved. By
increasing the size of graduate programs over a period of many years in
the absence of an equal increase in demand, universities inadvertently
augmented the possibilities for an intensified conflict between largely
insecure "intellectuals" and only slightly more secure "professionals."


My revaluation of the midcentury Americans is now complete. As I

suggested it would, the revaluation leads in contradictory directions. From
our perspective a quarter century later, the balance sheet quite naturally
includes minuses as well as pluses. Let me summarize briefly in relation to
the principal conceptions and empirical findings of the midcentury writers.

The Midcentury Americans Reconsidered

(1) In one respect, the midcentury writers were more farseeing than we are
today. Against the widely accepted contemporary view of intellectuals as
intensely engaged idealist critics, stands the more accurate view of intellec-
tuals found in Hofstadter and Shils. The work of intellectuals does sometimes
involve the defense of culturally significant values-one should never under-
estimate the importance of this-but it is also to provide searching criticism
based on deep study and contemplation, and better, more fully comprehending
conceptions of natural and human life. If emendations are necessary to the
definitions of Hofstadter and Shils, these emendations have to do with the
extent to which intellectual activity is now a rather well-understood cultural
role and increasingly requires validation within the community of socially
recognized intellectuals (cf. Nettl 1969; Kadushin 1974).
(2) Coser's typology of the forms of intellectual relations to power
remains useful. It contains two weaknesses, however. One is the tendency,
despite many reasonable disclaimers on Coser's part, to bestow the term
policy intellectual too liberally. The great majority of so-called policy
intellectuals are better thought of as state-oriented policy experts. The
other weakness of the typology is that it neglects an important category of
politically involved intellectuals: the academic intellectual-experts. Against
the implications of Coser' s analysis, it is this group, rather than any of the
others, that nowadays most often represents a responsible, relatively inde-
pendent center of intellectual political activity.
Steven Brint 65

(3) The long-standing association of intellectuals with antibourgeois,

and (following World War II) also left-of-center politics no longer exists
to the same degree that it once did. One should not go too far here; some
connection does continue to exist. The available research suggests that
participants in feminist, environmental, antinuclear, and other "new social
movements" of the left do tend to be found disproportionately among
younger intellectuals (Pinard and Hamilton 1989), and it is also these
people who are most likely to indicate that they are sympathetic to protest
(Kreisi 1989). Most of the few holdouts against the free-market tides of
recent years have been found in the academic and literary intellectual
communities. However, significant splits have emerged throughout the
intellectual communities of the West (see, e.g., the articles in Gagnon
1987). In the broad scheme of electoral politics in the West, the intellectual
stratum as a whole is at most only moderately left of center, and it is not
at all left of the national political centers in those countries where a
relatively egalitarian, "social-welfare-oriented" political culture has been
achieved (Brint, forthcoming, chap. 11).15 The notion of an "adversary
culture" of intellectuals becomes more and more a rallying myth of the
political right, with little or no empirical validity.
(4) What remains useful in the comparative work of the rnidcentury writers
is their emphasis on developmental trends over cross-national variation. This
emphasis seems increasingly sound. In the years since 1965, both ideological
and structural changes have shown a largely common trajectory in the West.
Moreover, their inclination to emphasize the importance of national cultural
traditions as a source of remaining variation (rather than traditions and
structures specific to the intellectual community) also appears to be well
placed. What is less useful is the specific thesis of an "end-of-ideology."
Ideology, properly understood, has not ended, and it is difficult to see how it
will ever end so long as people look for philosophical justification for their
sense of proper order in society. While trends toward moderation and reform-
ism are certainly evident, indeed more today than at midcentury, the possi-
bility of some future, significant episodes of radicalism ought not to be
discounted entirely. Given the oppositional traditions of intellectual life, so
well discussed by Edward Shils, it is always unwise to predict that the future
can be known in this regard.
(5) Above all, what is missing from the work of the midcentury Ameri-
cans is a proper appreciation of the dependence of intellectuals on the
power of situations and external forces. My effort in this chapter has been
to elaborate on the sources of dependence that C. Wright Mills alone
among the midcentury writers caught sufficient glimpse of, but to do so
in a less expressionistically exaggerated way than Mills himself was able
to provide (see Note 7). It has also been to show how these dependencies,

among other factors, have influenced the coherence of the intellectual

community in the recent past.

Dependence and Independence

This emphasis on the power of situations and institutional ties encour-
ages a new appreciation, among other things, of the characteristic mood
and assumptions of the midcentury writers themselves. Indeed, there may
be as many as three distinct kinds of situational influences on the sense of
independence and potency they conveyed. One is the high and buoyant
level of prestige and confidence of virtually all elite groups during this
high point of the "pax Americana." There is also the specific mood of
imperium-the sense of the possibility of doing great things in the world,
and the conviction that prudence is required to do these great things in a
lasting way. And, finally, there are the many concrete points of consensus
in what Hodgson (1976) called the dominant "liberal-conservative" ideol-
ogy that emerged during this hyperion age: the sense that reform is
desirable and natural, but that all reform must be of a meliorist sort; the
faith in value-oriented elites; and the sense that all elite segments have an
important role to play in a powerful and progressive republic.
Yet I do not think we should be content to substitute a hard-boiled
realism about dependent ties for what we take to be the consensual
illusions of yesterday. Independence is not just a fact or a fallacy; for
intellectual communities, it is always, in some significant measure, an
achievement as well. We would be foolish to accuse the midcentury writers
of a lack of self-consciousness on this point. The best of the midcentury
writers realized the fragility of intellectual independence, intellectual
influence, and intellectual community. Coser, for example, tells us repeat-
edly that without an intellectual community that "serves as a means of
social control, as a discipling agent ... [and] common point of reference"
intellectuals quickly "accommodate themselves to the policies of those
who make strategic decisions" (Coser 1965, pp. 184-85). And Kissinger
is equally direct about how the "desire to be helpful" frequently compels
intellectuals to sacrifice "what should be [their] greatest contribution to
society: [their] creativity" (Kissinger 1959, p. 33).
The work of the midcentury writers is, to be sure, partly a reflection of
conditions that encouraged a sense of independence, influence, and coher-
ence in the intellectual community. But it is also a study of how those
conditions are created and maintained. Because of that, the achievement
of the midcentury Americans is more than what can be added to our
scientific storehouse from their studies of the sociology of intellectuals; it
is also to be found between the lines in what they suggest to us about the
Steven Brint 67

creation of politically effective intellectual communities. This kind of

political inquiry is especially timely today as a counterweight to the
dependence and incoherence that are so evident-and are even rather
energetically cultivated-all around us.
What do the midcentury writers tell us that is relevant to any future efforts
to renew the independent political significance of the intellectual community?
They remind us, first of all, of an old conservative truth-that intellectuals in
power have avoided a disastrous loss of humanity in the service of theory only
when they have maintained strong ties to the practical and emotional life of
their times (see, e.g., Curti 1956, pp. 9-18). They also tell us that policy
intellectuals should not hope to maintain their independence without the help
of a group of like-minded colleagues, who are, at least to a large degree,
independent from government employment, and who exercise a vigorous
form of social control on one another in the service of programmatic aims
(Coser 1965, pp. 171-80). Perhaps most of all, they tell us that intellectuals
must be loyal, in a Durkheimian sense, to the particular "totem" of their tribe.
Ultimately, qualities of mind and spirit are essential. If intellectuals are to
find effective community and if they are to sustain it, they must cultivate again
an attitude quite like reverence for the efforts of the most profound and
graceful thinkers of many different stripes.


1. These include Annan's work (1955) on British intellectuals, Curti's (1956) discussion
of the American founding fathers as intellectuals, Williams's (1958) discussion of intellec-
tual reactions to the industrial and democratic revolutions, the Summer 1959 issue of
Daedalus with comments by five American intellectuals (Schlesinger et al. 1959) on the
status of American intellectuals, Bell's (1960) The End ofIdeology, de Huszar's (1960) edited
collection on intellectuals, Apter's (1964) edited collection on ideology and discontent, and
Lasswell and Lerner's (1965) collection on world revolutionary elites.
2. When the difference is defined, as Hofstadter observes, "it becomes easier to under-
stand why we sometimes say that a mind of admittedly penetrating intelligence is relatively
unintellectual; and why, by the same token, we see among minds that are unmistakably
intellectual a considerable range of intelligence" (Hofstadter 1963, p. 25).
3. In terms of linguistic history, there is, of course, much to commend this view. Both
the original use of the term intellectual during the Dreyfus affair in France and the original
use of the term intelligentsia in mid-19th century Russia were meant to denote groups critical
of and alienated from society (see Coser 1965, pp. 215-26, and Bell 1976).
4. It is a short jump from this definition to the stereotypical view of intellectuals as
maintaining a critical distance between themselves and the everyday life of their societies, and
thus, implicitly, as standing very often in a relation of alienation from their societies. Indeed,
Coser himself makes this short jump: "Intellectuals, then," he writes, "are not only puzzling but
upsetting to the run of ordinary citizens .... Without intellectuals to challenge the established
routines the traditions of the eternal yesterday-s-even while they maintain standards and

articulate demands-our culture would soon be a dead culture" (Coser 1965, p. x). And
elsewhere: "They consider themselves," Coser writes, "jealous guardians of moral standards
that are too often ignored in the marketplace and the houses of power" (Coser 1965, p. viii).
5. The emphasis on intellectual judgments and intellectual commitments are not univer-
sal. Lipset (1960), at times, favors a variation of the sociological argument based on status
inconsistency. He argues that intellectuals feel a sense of deprivation in the gap between their
cultural accomplishments and their perceived but middling status in the society.
6. When I say class, 1 obviously do not mean to use the term in the Marxist sense, but
rather in the less exacting sense of a self-conscious group of similarly situated people. In the
American context, some conservative writers quite rightly discerned a change of temper, an
increasing sense of self-consciousness and separation, in the way these matters were being
talked about. Kirk (1956), for example, noted that while Europe had long known a self-con-
scious intellectual class, the sense of intellectuals as a distinct class in society had only
recently attained a corresponding influence in the English-speaking world, largely because
of the relatively widespread diffusion of traditional liberal learning and the Protestant-dem-
ocratic concept of a free and dignified personality in all walks of life.
7. Of course, this emphasis was not entirely missing in the work of midcentury Ameri-
cans. It is mentioned at least peripherally by all of the writers. But it is at the center only in
Mills's (1951) vision of "Brains, Inc." Mills writes, for example,

Perhaps, in due course, intellectuals have at all times been drawn into line with either
popular mentality or ruling class, and away from the urge to be detached; but now in the
middle of the twentieth century, the recoil from detachment and the falling into line seem
more organized, more solidly rooted in the centralization of power and its rationalization
of modern society as a whole. (Mills 1951, p. 155)

The difficulty with Mills's view is that it is presented in a rather melodramatic way, full of
expressionistic exaggerations of the brute forces and the evil of the modern world.
8. This neutral usage of the term ideology contrasts with the connotations that ideology
has gained with bias, inflexibility, and special pleading. Geertz (1964) aptly summarizes one
prevailing view: "I have a social philosophy, you have political opinions, and he has an
ideology." For a defense of a nonpejorative use of the term, see Apter (1964), Geertz (1964),
and Gouldner (1976).
9. The sense of intellectual ferment in the Reagan White House is well conveyed in a
recollection (by a presidential speechwriter) of lunch at the White House mess during the
final months of Reagan's first term: "This is like a tutorial, I thought. These people are the
bright, intense people in college who couldn't stop talking about Hawthorne and Tennyson
and the Lake poets. Only the topic was public policy .... It was the authentic sound of the
Reagan revolution buzzing around me" (Noonan 1990, p. 53).
10. They are in fact mentioned as an aside in Coser's discussion as one of two other
possibilities for intellectuals, the other being "withdrawing entirely." See Coser 1965, p. 136.
11. As Kissinger himself noted, "Since individuals who challenge the presuppositions of
the bureaucracy rarely keep their positions as advisers, great pressures are created to
elaborate on familiar themes rather than risk new departures that may both fail and prove
unacceptable" (Kissinger 1959, p. 33).
12. Survey evidence from this period suggests that significant changes in policy-related
racial attitudes were occurring at precisely this time. See, for example, Schuman et al. (1985).
13. The dissatisfaction of conservative ideologists with the "timid" Reagan Administra-
tion is evident as early as January of 1983. Irving Kristol, the neoconservative theorist, for
example, despairs in "The Emergence of Two Republican Parties" of the possibility of a truly
Steven Brint 69

activist conservative government emerging out of the "pragmatic" Reagan Administration.

This dissatisfaction ofthe more ideologically driven Reaganites reaches a crescendo in David
Stockman's The Triumph of Politics. written in 1986. Lipset and Dobson (1972) note that
ideologically oriented liberal intellectuals were similarly unhappy with the caution of the
putatively liberal, but cautious Kennedy Administration as well.
14. It is indeed something of an irony that leftist intellectuals should live to see a time
when ideals clearly mattered in the constitution of new political coalitions, as they had hoped
they would all along, and to have those ideals represent conservative viewpoints. But this
has been the case in the United States and also in many European countries. It is a further,
and perhaps more telling, irony that the intellectuals of the right, who are schooled on Burke
and de Tocqueville when out of power. often have been disappointed with "overly practical"
or "unimaginative" politicians, who failed to press conservative ideas strenuously enough.
See, for example, Kristol 1983; and Stockman 1987.
15. Shils's (1958) original ideas about the intellectuals and the powers are also relevant to
the issue of intellectual leftism. Although intellectuals are-for Shils-to be defined exclusively
by the nature of their concerns, they do also have traditions that can, under some (unspecified)
circumstances. encourage them to take adversarial positions. Shils stresses the traditions of
scientism, romanticism, millenarianism, and populism. While I would tend to agree with him
about the importance of intellectual traditions, I am inclined to de-emphasize scientism and
millenarianism as important oppositional forces in the contemporary world, and to add early
20th century progressivism and humanism as other important traditions in American society.


Aberbach, Ronald, Robert D. Putnam, and Bert A. Rockman. 1981. Bureaucrats and Politi-
cians in Western Democracies. Cambridge, MA: Harvard University Press.
Anderson. Martin. 1988. Revolution. New York: Harcourt-Brace.
Annan, Noel H. 1955. "The Intellectual Aristocracy." Pp. 243-87 in Studies in Social History.
edited by J. H. Plumb. London: Longmans, Green.
Apter, David E., ed. 1964. Ideology and Discontent. London: Free Press of Glencoe.
Bell, Daniel. 1960. The End of Ideology. Glencoe, IL: Free Press.
- - - . 1976. "The 'Intelligentsia' in American Society." In Tomorrow's American: The
Will Lectures of 1976. New York: Oxford University Press.
Brint, Steven. Forthcoming. Retainers. Merchants. and Priests: Politics and Culture in the
Professional Middle Class in America. Princeton: Princeton University Press.
Bruce-Briggs, B. 1979."Conclusion: Notes Toward the Delineation of the New Class." pp. 191-216
in The New Class? edited by B. Bruce-Briggs. New Brunswick. NJ: Transaction Books.
Cameron, David R. 1982. "On the Limits of the Public Economy." Annals of the American
Academy of Political and Social Science 459:46-62.
Coser, Lewis. 1965. Men of Ideas. New York: Free Press.
Curti, Merle. 1956. American Paradox: The Conflict ofThought and Action. New Brunswick,
NJ: Rutgers University Press.
De Huszar, George B. 1960. The Intellectuals: A Controversial Portrait. New York: Free Press.
Gagnon, Alain G., ed. 1987. Intellectuals in Liberal Democracies. New York: Praeger.
Glazer, Nathan. 1975. Affirmative Discrimination: Ethnic Inequality and Public Policy. New
York: Basic Books.
Geertz, Clifford. 1964. "Ideology as a Cultural System." In Ideology and Discontent. edited
by David E. Apter. London: Free Press of Glencoe.
Gouldner, Alvin W. 1976. The Dialectic ofIdeology and Technology. New York: Seabury Press.

Heclo, Hugh and Mendrik Madsen. 1987. Policy and Politics in Sweden. Philadelphia:
Temple University Press.
Hirsch, E. D. 1987. Cultural Literacy. Boston: Houghton-Mifflin.
Hochschild, Arlie. 1989. The Second Shift: Working Parents and the Revolution at Home.
New York: Viking.
Hodgson, Godfrey. 1976. America in Our Time. Garden City, NY: Doubleday.
Hofstadter, Richard. 1963. Anti-Intellectualism in American Life. New York: Knopf.
Jacoby, Russell. 1987. The Last Intellectuals: American Culture in the Age ofAcademe. New
York: Basic Books.
Kadushin, Charles. 1974. The American Intellectual Elite. Boston: Little, Brown.
Kennedy, Paul. 1987. The Rise and Fall of Great Powers. New York: Random House.
Keohane, Robert O. 1984. "The World Political Economy and the Crisis of Embedded
Liberalism." Pp. 15-38 in Order and Conflict in Contemporary Capitalism. edited by
John H. Goldthorpe. New York: Oxford University Press.
Kirk, Russell. 1956. Beyond the Dream ofAvarice. New York: Henry Regnery.
Kissinger, Henry. 1959. "The Policy-Maker and the Intellectual." The Reporter 29(March
Kreisi, Hanspeter. 1989. "New Social Movements and the New Class in the Netherlands."
American Journal of Sociology 94: 1078-1116.
Kristol, Irving. 1983. Reflections of a NeoConservative. New York: Basic Books.
Lasswell, Harold D. and Daniel Lerner. 1965. World Revolutionary Elites: Studies in Coer-
cive Ideological Movements. Cambridge: MIT Press.
Lipset, Seymour Martin. 1960. Political Man: The Social Bases ofPolitics. Garden City, NY:
- - - . 1963. The First New Nation: The United States in Comparative Perspective. New
York: Norton.
Lipset, Seymour Martin and Richard B. Dobson. 1972. "The Intellectual as Critic and Rebel."
Daedalus 101:137-98.
Mills, C. Wright. 1951. White Collar. New York: Oxford University Press.
Nettl, J. P. 1969. "Ideas, Intellectuals, and Structures of Dissent." Pp. 53-124 in On Intellec-
tuals, edited by Philip Rieff. Garden City, NY: Doubleday.
Noonan, Peggy. 1990. What I Saw at the Revolution. New York: Random House.
Parsons, Talcott. 1969. "The Intellectual: A Social Role Category." Pp. 3-24 in On Intellec-
tuals, edited by Philip Reiff. Garden City, NY: Doubleday.
Pinard, Maurice and Richard F. Hamilton. 1989. "Intellectuals and the Leadership of Social
Movements," Research ill Social Movements. Conflicts. and Change 11:73-108.
Schlesinger, Arthur R., Jr., et al. 1959. "Comments on 'American Intellectuals.' " Daedalus
Schuman, Howard, et aL 1985. Racial Attitudes in America: Trends and Interpretations.
Cambridge, MA: Harvard University Press.
Schumpeter, Joseph. 1942. Capitalism. Socialism. and Democracy. New York: Harper & Row.
Shils, Edward A. 1955. "The End of Ideology?" Encounter 5(November):52-58.
- - - . 1958. "The Intellectuals and the Powers." Comparative Studies in Society and
History 1:5-23.
- - - . 1968. "Intellectuals," International Encyclopedia of the Social Sciences 7:399-414.
Stockman, David A. 1986. The Triumph of Politics: Why the Reagan Revolution Failed. New
York: Harper & Row.
Suleiman, Ezra N., ed. 1984. Bureaucrats and Policy-Making. New York: Holmes & Meier.
Trilling, Lionel. 1965. Beyond Culture. New York: Viking.
Williams, Raymond. 1958. Cultureand Society. 1780-1950.New York: Columbia University Press.
Wright, Vincent. 1989. The Government and Politics ofFrance. New York: Holmes & Meier.
Part I

The Politics of Citizenship

Migrants Into Citizens?
Traditions of Nationhood and Politics of
Citizenship in France and Germany

Rogers Brubaker'

Immigration of unprecedented magnitude and diversity has posed a fun-

damental challenge to the nation-states of northwestern Europe. It has
compelled these countries to reinterpret traditions, reshape institutions,
and rethink the meaning of citizenship and membership. The taken-for-
granted link between citizenship and nationhood, axiomatic to politics in
the nation-state, has become problematic-both for those contesting it in
the name of a postnational multiculturalism and for those reaffirming it in
the name of a new nationalism. National identity has been revived as a
political theme, nativism as a political program.
This chapter addresses in historical and comparative perspective one
aspect of the new politics of citizenship and nationhood. Its point of
departure is a sharp contrast between the policies and politics of citizen-
ship of the two major nation-states of continental Europe. While birth and
subsequent residence in France automatically transform second-genera-
tion immigrants into citizens, birth in the Federal Republic of Germany
has no bearing on German citizenship. The French citizenry is defined
expansively, as a territorial community, the German citizenry-except in
the special case of ethnic German immigrants-restrictively, as a commu-
nity of descent. Naturalization policies, moreover, are more liberal in
France than in Germany, and naturalization rates are 4 to 5 times higher.
The overall rate of civic incorporation for immigrants is 10 times higher
in France than in Germany. The gap is even greater for second- and
third-generation immigrants. A generation of young Franco-Portuguese,


Franco-Algerians, and Franco-Moroccans is emerging, claiming and exercis-

ing the rights of French citizenship. In Germany, by contrast, more than 111.2
million Turks-including more than 400,000 who were born in Germany-
remain outside the community ofcitizens. Yet at the same time, newly arrived
ethnic German immigrants from Eastern Europe and the former Soviet
Union-more than a million in 1988-1991-are legally defined as Germans
and automatically granted full civic and political rights.
The expansiveness of French citizenship vis-a-vis immigrants, this
chapter argues, reflects a state-centered, assimilationist, essentially polit-
ical understanding of nationhood, while the restrictiveness of German
citizenship reflects an ethnocultural understanding of nationhood as prior
to and independent of the state. After noting the thoroughgoing similarities
in migration processes, immigrant populations, and immigration policies
in France and Germany, and outlining the striking differences in the
policies and politics of citizenship, the chapter analyzes the genesis and
development of distinctive French and German traditions of nationhood
and concludes by showing how they have shaped and sustained sharply
differing definitions of citizenship and patterns of civic incorporation.



The divergence in the policies and politics of citizenship is particularly

striking in view of the similar French and German experiences with
migrant labor in the last quarter-century (Rist 1979; Manfrass 1984). In
both countries, foreign workers were recruited in large numbers in the
1960s and early 1970s in response to labor shortages. Organized recruit-
ment was suspended in 1973-1974, partly in response to the oil shock and
ensuing recession, partly in response to the growing concern about the
social and political consequences of large-scale immigration. Nonetheless,
populations of immigrant origin have continued to grow in both countries,
largely through processes of family reunification.
Immigrants in both countries have become dramatically more visible in
everyday life during the last two decades. During the 1950s and early
1960s, most foreign workers were either single or separated from their
families. Many lived in isolated workers' hostels. Outside the workplace,
they were largely invisible, participating little in the social, cultural, or
political life of the host society. In the last two decades, however, the
sojourners have become settled. Single workers were joined by their
families, or formed new families. Immigrants became neighbors, school-
mates, and joint users of public spaces. An increasingly vocal second
Rogers Brubaker 75

generation emerged, tenuously rooted in the culture of the parents' gener-

ation, yet economically and socially marginalized in the country of resi-
dence. Groups marked by dress, language, religion, and custom as "cul-
turally distant" comprised the fastest growing segment of the immigrant
community. Immigrants in both countries have clustered in particular
regions and, within cities, in particular neighborhoods. All these develop-
ments made immigrants much more visible.
In both countries, immigrants comprise a substantial fraction of the manual
working class and are overrepresented in dirty, dangerous, unpleasant, ill-
paid, and menial occupations. They are also overrepresented among the
unemployed. As a relatively young group, with comparatively high fertility
rates, immigrants playa similar demographic role in France and Germany,
which share concerns about low fertility and aging populations. This has
implications for the labor market, the social security system, and (in the longer
run) for military conscription-if indeed peacetime conscription survives the
great geopolitical reconfiguration now under way.
Discourse about immigration and immigrants follows similar patterns
in both countries. There is an inclusionary discourse that stresses the
economic and cultural contribution of immigrants to the host society and
the values of tolerance and diversity. And there is a counterdiscourse
stressing the unassimilability of immigrants, the dangers of excessive
cultural heterogeneity, the social strains and economic costs of immigra-
tion, and the prospect of Islamic fundamentalism and interethnic strife.
Finally, there are striking similarities in immigration policies. Since the
mid-1970s, all French and German governments, left and right, have
pursued the same threefold policy, seeking to impose strict limits on
further immigration, to encourage voluntary return migration, and to
facilitate the integration of second-generation immigrants.i
There are, of course, significant differences between French and Ger-
man experiences with immigration. Many immigrants to France have come
from former French colonies and protectorates, while immigration to
Germany has lacked this colonial connection. Both countries have been
concerned with undocumented immigration and with an upsurge in the
number of persons seeking political asylum, but the French have been
particularly preoccupied with the former, the Germans with the latter. Jean
Marie Le Pen's National Front has fared much better in France than any
far-right xenophobic party in Germany, although radical right parties have
made recent gains in Germany as well. France is particularly concerned
about migration from the south, from the Maghreb and sub-Saharan Af-
rica, Germany about migration from the east, especially from Poland and
the former Soviet Union. And since the massive exodus 30 years ago of
colonial settlers from postindependence Algeria, there has been no French

analogue to the great migration of ethnic Germans from Eastern Europe

and the Soviet Union, which brought more than a million immigrants to
Germany between 1988 and 1991.


Despite these differences, the overall picture is one of similar migration

processes, comparable immigrant populations, and converging immigra-
tion policies. In the context of these thoroughgoing similarities, the sharply
differing policies and politics of citizenship stand out as a striking anom-
aly. In the first place, German naturalization policies, although recently
liberalized, remain more restrictive than those in France. Ten years'
residence is ordinarily required in Germany, five years in France. More
important, candidates for naturalization must ordinarily renounce their
original citizenship in Germany, but not in France. Besides these specific
differences in requirements, there are more general differences in attitudes
toward naturalization. This is clearly expressed in the administrative
regulations governing naturalization, which state unambiguously that "the
Federal Republic is not a country of immigration [and] does not strive to
increase the number of its citizens through naturalization" (quoted in
Groth 1984, p. 88). In countries of immigration like the United States and
Canada naturalization is expected of immigrants; the failure to naturalize
is anomalous. In France too, which alone in continental Europe has a
tradition of immigration for purposes of permanent settlement, naturaliza-
tion has been considered the normal and desirable outcome of permanent
settlement. In German self-understanding, by contrast, one cannot join the
nation-state by voluntary adhesion (the North American model) or state-
sponsored assimilation (the French model).
Immigrants' attitudes toward naturalization, moreover, differe in France
and Germany. In 1985 only 6% of German migrant workers and family
members, and 9% ofthose aged 15-24, intended to naturalize (Forschungs-
institut der Friedrich-Ebert-Stiftung 1986, pp. 484, 486), while about a
quarter of young foreigners in France intend to beome citizens (Dubet
1989, p. 85). The very low propensity to naturalize among German immi-
grants, many of whom would clearly qualify for naturalization, reflects a
desire to retain one's original citizenship (Forschungsinstitut der Friedrich-
Ebert-Stiftung 1986, pp. 485-87). Beyond this, though, the differential
interest in naturalization may reflect different understandings of what
naturalization means in France and Germany. To a greater extent in
Germany than in France, it appears, naturalization is perceived as involving
Rogers Brubaker 77

not only a change in legal status, but a change in nature, a change in

political and cultural identity, a social transubstantiation that immigrants
have difficulty imagining, let alone desiring. Evidence of this blurring, in
the minds of immigrants, between legal citizenship and a richer, more
diffuse notion of ethnocultural nationality can be found in France as well.
In France, however, a larger fraction of the immigrant population seems
to have adopted a more instrumental, "desacralized" understanding of
citizenship, seems to have divorced the legal question of citizenship from
broader questions of political loyalty and cultural belonging (Sayad 1982,
These differences in policies and attitudes toward naturalization are
reflected in naturalization rates that are 4 to 5 times higher in France than
in Germany for the main groups of migrant workers and their dependents."
Italians naturalize at rates roughly 5 times higher, Spanish at rates 10 times
higher in France than in Germany. Tunisians and Moroccans in France
naturalize at rates nearly 10 times those of Turks in Germany. Of the 1.5
million Turks in Germany, over 1 million of whom have resided 10 or more
years in Germany and more than 400,000 of whom were born there, only
about 1,000 acquire German citizenship each year. Even if rates increased
tenfold, naturalizations would still be far outweighed by the 25,000-
30,000 new Turkish citizens born each year in the Federal Republic.
The German government has been saying since the mid-1980s that it favors
the naturalization of second-generation immigrants, for "no state can in the
long run accept that a significant part of its population remain outside the
political community.t''' In 1990, the legal provisions governing naturalization
were liberalized for persons brought up in Germany and educated in German
schools, as well as for persons having resided more than 15 years in Germany.
Over time, a modest increase in naturalization rates is to be expected as the
immigrant population becomes increasingly settled. Yet patterns of natural-
ization are unlikely to change dramatically. The most important obstacle to
naturalization-the requirement that candidates give up their original citizen-
ship--was not touched by the 1990 reform. Moreover, the barriers to natural-
ization lie not only in the restrictiveness of legal provisions but equally in the
political culture of naturalization, embodied in attitudes of Germans and
immigrants alike. Without a changed understanding of what it is to be--or to
become-German, this liberalization in naturalization policy will not produce
a dramatic surge in naturalization. .
Naturalization rates, then, are 4 to 5 times higher in France than in
Germany. But patterns of civic incorporation diverge even more sharply
than this suggests. It is not enough to consider the voluntary acquisition
of citizenship by naturalization or individual declaration. One must con-
sider also the attribution of citizenship by the state. Naturalization patterns

and policies must be understood in conjunction with the rules specifying

whom states unilaterally define as citizens. Working invisibly and auto-
matically, independently of the will-and sometimes even the know1-
edge--of the persons concerned, the rules governing the ascription of
citizenship have been all but ignored by the meager literature on immigra-
tion and citizenship. Yet they are more important than naturalization rules
in shaping patterns of civic incorporation in France and Germany. Ascrip-
tion constitutes and perpetually reconstitutes the citizenry; naturalization
reshapes it at the margins. The striking difference in the civic incorpora-
tion of immigrants in France and Germany is chiefly a consequence of
diverging rules of ascription. Differing naturalization rules and rates
reinforce this difference but are not its fundamental source.
The central difference between French and German ascription rules
turns on the significance attached to birth and prolonged residence in the
territory. While French citizenship is ascribed, at birth or majority, to most
persons born on French territory of foreign parents, German citizenship is
ascribed only on the basis of descent. Birth and prolonged residence in
Germany have no bearing on citizenship status. French citizenship law
automatically transforms most second- and third-generation immigrants
into citizens; German citizenship law allows immigrants and their descen-
dants to remain foreigners indefinitely.
In both France and Germany, to be sure, as throughout continental Europe,
citizenship is ascribed to children of citizens, following the principle of jus
sanguinis. In Britain and the Americas, by contrast, citizenship is ascribed to
all persons born in the territory, following the principle of jus soli. What I
want to highlight here is sharp difference in the extent to which France and
Germany, sharing the same basic principle of jus sanguinis, supplement this
principle with elements ofjus soli. France and Germany represent polar cases
here: French citizenship law includes a substantial territorial component,
while German citizenship law includes none at all. Most other western
European jus sanguinis countries include some complementary elements of
jus soli, without going as far as France. 6
Although based on jus sanguinis. French citizenship law incorporates
substantial elements of jus soli. Thus French citizenship is attributed at
birth to a child born in France if at least one parent was also born in
France-including Algeria and other colonies and territories before their
independence. This means that the large majority of the roughly 400,000
children born in France since 1963 of Algerian parents are French citizens.
Moreover, citizenship is acquired automatically at age 18 by all children
born in France of foreign parents, provided they have resided in France
for the last 5 years and have not been the object of certain criminal
condemnations. By this means roughly 300,000 persons became French
Rogers Brubaker 79

between 1973 and 1991. More than a million foreign residents are under
age 18. At least two-thirds of them were born in France and are destined
to become French at age 18. Thus although the citizenship law of the
United States is based on jus soli, while that of France is based on jus
sanguinis, the result-as far as second-generation immigrants are con-
cerned-is similar: almost all persons born in France and residing there at
majority have French citizenship. German citizenship law, in contrast, is
based exclusively on jus sanguinis. Birth in the territory, even coupled with
prolonged residence, has no bearing on citizenship. Second-generation
and even third-generation immigrants can acquire German citizenship
solely by way of naturalization.
We are now in a position to consider the combined effects of different
rules of ascription and rates of naturalization on the civic incorporation of
the major immigrant groups (excluding ethnic German immigrants to
Germany). Of the nearly 3 million foreign residents from the core immi-
grant groups in Germany-Italians, Spaniards, Greeks, Yugoslavs, and
Turks-fewer than 5,000 acquire German citizenship each year, and nearly
half of these are Yugoslavs. France, on the other hand, gains more than
53,000 new citizens each year from a slightly smaller core immigrant
population of Italians, Spaniards, Portuguese, Algerians, Tunisians, and
Moroccans. This includes about 16,600 persons who are defined as French
at birth each year by virtue of birth in France in conjunction with the birth
of at least one parent in preindependence Algeria, and another 12,700
persons born in France and defined as French on attaining legal majority.
The overall rate of civic incorporation for these core immigrant groups is
thus more than 10 times higher in France than in Germany,"
For second- and third-generation immigrants, the difference in rates of
civic incorporation is even greater. In both France and Germany, new
immigration declined precipitously after 1973. As a result, a steadily
increasing fraction of the population of immigrant origin consists of
persons born in France or Germany. In France, almost all of these persons
are either defined as French at birth or programmed to become French
automatically at age 18. In Germany, which lacks any mechanism of
automatic civic incorporation, second- and third-generation immigrants
will have to naturalize if they want to become citizens. And there is no
indication that they will do so in large numbers.
One further peculiarity of German citizenship law should be noted.
While the citizenry is defined restrictively vis-a-vis non-German immi-
grants, it is defined expansively vis-a-vis ethnic Germans. This ethnic
inclusiveness has two aspects. First, the citizenry recognized by the Fed-
eral Republic of Germany always included the citizens of the German
Democratic Republic. As far as citizenship law is concerned, the division

of Germany never happened. Or rather it happened only from the East

German side. Between 1967 and 1990, there was a separate East German
citizenship, but there never was a separate West German citizenship. Not
wanting to validate the division of Germany, the West German authorities,
throughout the postwar period, insisted on the continued validity of a
single German citizenship. As the two Germanies consolidated their sep-
arate statehoods, the continued West German insistence on a single citi-
zenship came to seem increasingly anomalous. Yet it took on dramatic new
meaning in fall 1989 and spring 1990; for it was the common German
citizenship that guaranteed every East German, as a German citizen, the
constitutional right to enter, reside, and work in West Germany. Common
citizenship paved the way for common statehood.
The second aspect of the ethnic inclusiveness of German citizenship
pertains to the ethnic German immigrants from Eastern Europe and the
former Soviet Union. These immigrants are treated very differently from
non-German immigrants. They are legally defined as Germans and imme-
diately accorded all the rights of citizenship.f The liberalization of emi-
gration and travel policies in Eastern Europe and the Soviet Union has
engendered a great exodus of ethnic Germans from this region, particularly
from Poland, Romania, and Russia, reversing the centuries-old Drang nach
Osten of Germans into Slavic lands. More than a million ethnic Germans
arrived in Germany between 1988 and 1991; at this writing, the flow
continues unabated.
The policies and politics of citizenship, then, are strikingly different in
France and Germany. Naturalization policies and practices are more lib-
eral, and naturalization rates 4 to 5 times higher, in France. French
citizenship, moreover, is automatically attributed to French-born children
of immigrants at their majority, while German citizenship is based solely
on descent. As a result, the rate of civic incorporation for migrant workers
and their families is more than 10 times higher in France than in Germany.
Yet while German citizenship is closed toward non-German immigrants,
it is remarkably open towards ethnic German immigrants from Eastern
Europe and the former Soviet Union.


How can one explain this striking divergence-pregnant with conse-

quences for the future-in the policies and politics of citizenship? A
familiar strategy of comparative analysis, known since John Stuart Mill as
the method of difference, seeks to provide a crude substitute for experi-
mental or statistical control by choosing cases differing in outcome and in
Rogers Brubaker 81

one key contextual respect-the suspected cause-and maximally similar

with respect to other possible causes (Mill [1843] 1949, p. 255ff). This
strategy-as Mill himself forcefully pointed out-should not be mistaken
for a rigorous method, particularly when one is dealing with entities as
complex as national societies ([1843] 1949, p. 575). Still, it is a useful
orienting strategy, for the stubborn dissimilarity in the politics of citizen-
ship in the face of converging immigration policies and similar immigrant
populations does pose an analytical problem. And France and Germany do
differ in one respect that is at least plausibly connected to the politics of
citizenship. For two centuries, locked together in a fateful position at the
center of state- and nation-building in Europe, France and Germany have
been constructing, elaborating, and furnishing to other states distinctive,
even antagonistic models of nationhood and national self-understanding,
grounded in differing historical paths to nation-statehood.
In the French tradition, the nation has been conceived in relation to the
institutional and territorial frame of the state. Revolutionary and Republi-
can definitions of nationhood and citizenship-militantly unitarist, uni-
versalist, and secular-reinforced what was already in the ancien regime
an essentially political understanding of nationhood. Yet while French
nationhood is constituted by political unity, it is centrally expressed in the
striving for cultural unity. Political inclusion has entailed cultural assim-
ilation, for regional cultural minorities and immigrants alike.
If the French understanding of nationhood has been state-centered and
assimilationist, the German understanding has been Yolk-centered and
differentialist. Since national feeling developed before the nation-state,
the German idea of the nation was not originally political, nor was it linked
to the abstract idea of citizenship. This prepolitical German nation, this
nation in search of a state, was conceived not as the bearer of universal
political values, but as an organic cultural, linguistic, or racial commu-
nity-as an irreducibly particular Volksgemeinschaft. On this understand-
ing, nationhood is an ethnocultural, not a political fact.
I should emphasize that my argument is not purely cultural. If the category
of the nation has been more firmly linked to the territorial and institutional
frame ofthe state in France than in Germany, this reflects differences in the
timing of state-building and in the scale of the emerging territorial states. A
state roughly coextensive with what would come to be conceived as the nation
was in place before the development of national consciousness in France but
not in Germany: national consciousness therefore developed within the terri-
torial and institutional frame of the state in France but outside and against the
territorial and institutional frame of existing states in Germany. The wider
reach of territorial state-building in France than in Germany, in turn, reflects
a deep difference in economic, cultural, and political geography between what

Rokkan and Urwin have called monocephalic and polycephalic zones of

Europe. Polycephalic Europe, a legacy of medieval overland trade pat-
terns, consists of the broad north-south belt of closely spaced cities
stretching from Italy to the North Sea and the Baltic, and running through
the heart of western Germany. Here the density of cities and ecclesiastical
principalities inhibited the early consolidation and expansion of territorial
states. Large territorial states developed earlier on the fringes of this city
belt, where contending centers (such as the tie de France) faced less
competition and enjoyed more room for expansion (Rokkan and Urwin
1983, pp. 7-12, 27, 35-39).
The scale of political authority in central Europe during early modern
times, then, made it impossible to identify the German nation with the
institutional and territorial frame of a state. The medieval and early
modern Empire came closest to providing such a frame. But it was supra-
national in scope and in self-understanding; more important, it was not a
state but a loosely structured set of increasingly irrelevant institutions.
Unlike the early modern French state, it lacked the integrative power of a
centralizing bureaucratic administration and could not shape a firmly
state-anchored national consciousness. In Germany, the "conceived order"
(Lepsius 1985) or "imagined community" (Anderson 1983) of nationhood
and the institutional realities of statehood were sharply distinct; in France
they were fused. In Germany, nationhood was an ethnocultural fact; in
France, it was a political fact.
A second, closely related difference in elite habits of national self-un-
derstanding is also rooted in political and cultural geography. The French
understanding of nationhood has been assimilationist, the German under-
standing "differentialist." The gradual formation ofthe French nation-state
around a single political and cultural center has long supplied to the French
elite a model of effective and, by and large, legitimate assimilation that the
conglomerative pattern of state-building in polycentric, biconfessional," even
(in Prussia) binational Germany could not provide.i'' The vehicle for the
concentric, assimilative expansion of nationhood in France was the grad-
ually increasing penetration into the periphery of the instruments and
networks of the central state (school, army, administration, and networks
of transportation and communication) (Weber 1976). In Germany, the
development of the Prussian state comes closest to this model of the
assimilationist state-nation. Yet it was the geopolitical fate of Prussia to
become, in the late 18th century, a binational state; and Prussia failed,
decisively, to assimilate its large Polish population. The French state did
not fully assimilate Bretons, Basques, Corsicans, and Alsatians, but its
failure was neither so complete, so evident by the turn of the century, nor
so consequential for national self-understanding.
Rogers Brubaker 83

The ethnocultural frontier between Germans and Slavs, not only in

eastern Prussia but throughout the zone of mixed settlement in east central
Europe, has been central to German self-understanding. This frontier has
no parallel in the French case. I I Massive eastward migration of Germans
in the high middle ages and again in the early modem period had created
numerous pockets of German settlement in Slavic lands. Much assimila-
tion in both directions occurred in these borderlands over the centuries.
Yet the decisive fact, for national self-understanding, was the assimilation
that did not occur. 12 The preservation of German language, culture, and
national identity over centuries in boundary-maintaining enclaves and
outposts in the Slavic East, and the success of the Poles in preserving their
language, culture, and national identity in eastern Prussia, furnished to the
German elite a differentialist, bounded model of nationhood, a feeling for
the tenacious maintenance of distinctive ethnonational identities in zones
of ethnoculturally mixed populations. Germany defined itself as a frontier
state, with reference to the German-Slav borderlands, in a way thathas no
parallel in France.


Distinctive patterns of national self-understanding, then, have deep

roots in the political and cultural geography of medieval and early modem
Europe. This holds both for the state-centered, assimilationist strand in the
French understanding of nationhood-which antedates, although it was
powerfully reinforced by, the French Revolution-and for the ethnocultu-
ral, differentialist strand in the German understanding. To be sure, French
and German understandings of nationhood have not been fixed and immu-
table. At the time of the Dreyfus Affair, during the Vichy regime, and
again in recent years, the prevailing French idiom of nationhood-state-
centered and assimilationist-has been challenged by a more ethnocultu-
ral counteridiom, represented today by Jean-Marie Le Pen. And in Ger-
many, the ethnocultural idiom of nationhood has represented only one
strand of a more complex national self-understanding.
For several centuries, nonetheless, the prevailing French and German
idioms of nationhood have differed markedly; and they continue to differ
today. These distinctive understandings of nationhood are embodied and
expressed in sharply differing definitions of citizenship. The expansive,
assimilationist citizenship law of France, which automatically transforms
second-generation immigrants into citizens, reflects the state-centered,
assimilationist self-understanding of the French. And the German definition

of the citizenry as a community of descent, restrictive toward non-German

immigrants yet remarkably expansive toward ethnic Germans from East-
ern Europe and the Soviet Union, reflects the pronounced ethnocultural
inflection in German self-understanding. .
It is one thing, of course, to point out affinities between understandings
of nationhood and definitions of citizenship; it is another to demonstrate
in detail the concrete bearing of national self-understandings on the shap-
ing and reshaping of citizenship law. The latter is undertaken elsewhere
(Brubaker 1992); here I can give only a sketch of the argument. The
sharply differing ways of defining the citizenry in France and Germany
crystallized in the decades before World War I, in 1889 and 1913 respec-
tively. The German population, during these decades, was larger than the
French and growing much more rapidly. One might think, then, that
differing demographic and military interests led the French and German
states to adopt differing definitions of citizenship. I do not accept this
instrumentalist explanation. To be sure, the French were increasingly
concerned about demographic stagnation after the Franco-Prussian war.
Yet in the 1880s, the state did not need new citizens as soldiers. Now that
conscription was defined as a universal obligation of citizenship, the state
disposed of too many, not too few, potential soldiers. Since military
budgets did not permit the training of all fit and eligible French citizens,
there was no military interest in enlarging further the pool of citizens by
redefining second-generation immigrants as citizens (Mitchell 1984, p. 20;
Challener 1965, pp. 54-55).
There was, however, a political interest in an expansive definition of
citizenship. Republican civic ideology, which emphasized universal and
equal military service, made second-generation immigrants' exemption
from military service ideologically scandalous and politically intolera-
ble--especially since second-generation immigrants were not considered
true foreigners at all, but rather persons who were French in fact though
not in law. One legislator denounced them significantly, as "pretendus
etrangers," as "would-be foreigners" (Journal Officiel 1887, p. 81). The
prevailing characterization of second-generation immigrants as socially
and culturally French was made possible by an assimilationist understand-
ing of nationhood. Deeply rooted in political and cultural geography, this
assimilationist self-understanding was powerfully reinforced during the
1880s by Republican reforms of school and army. Primary education,
under Jules Ferry, was made free, compulsory, secular, and intensely
nationalistic; and primary schools become great engines of assimilation,
welding France for the first time into a unified nation. The army, too,
reorganized on the basis of universal conscription and conceived as the
"school of the nation," was an agent of assimilation. If schools and army
Rogers Brubaker 85

made "peasants into Frenchmen," as Eugen Weber (1976) has shown, they
made second-generation immigrants into Frenchmen in the same way. The
interest of the French state in an expansive definition of citizenship, then,
was not immediately given by demographic or military imperatives. Rather,
this interest was mediated-indeed constituted-by a certain way of think-
ing and talking about membership of the French nation-state.
Nor can the distinctiveness of the German definition of citizenship-
markedly restrictive toward non-German immigrants, yet markedly expan-
sive toward ethnic Germans from Eastern Europe and the Soviet Union-be
interpreted in instrumental terms. In Wilhelmine Germany as in Republican
France, understandings of nationhood shaped appraisals of state interests.
Yet while the French understanding of nationhood-state-centered and
robustly assimilationist-engendered an interest in the civic incorporation
of second-generation immigrants, the German understanding of nation-
hood engendered an interest in their civic exclusion.
Migrant labor was economically indispensable in eastern Prussia in the
Wilhelmine era. Yet immigrants---ethnic Poles from Russia and Austria-
were not wanted as citizens, for no one believed that they could be made
into Germans. In part, this was the legacy of a traditionally less assimila-
tionist, more ethnocultural understanding of nationhood. Yet just as the
French assimilationist self-understanding was powerfully reinforced in
the 1880s, so too the German ethnocultural, differentialist self-under-
standing was powerfully reinforced in the Wilhelmine era by the increas-
ingly evident failure of attempts to assimilate indigenous Poles in the
Prussian east (Broszat 1972, pp. 143, 157; Blanke 1981, p. 60). Having
failed to secure the political loyalty of Poles to the German state, and
having failed to assimilate them to German language and culture, Pruss ian
and German policy toward the indigenous Poles became increasingly
"dissimilationist." The state openly discriminated by ethnic nationality,
treating ethnic Germans and ethnic Poles differently in an effort to
"strengthen Germandom" in frontier districts (Wehler 1979, pp. 193-95;
Tims 1941, pp. 151-88). Since the state had failed to assimilate indigenous
Poles in the Pruss ian east, there was no reason to believe that it would
succeed in assimilating immigrant Poles. An ethnocultural, differentialist
way of thinking and talking about membership of the German nation-state
thus supported an interest in a restrictive definition of citizenship. An
expansive citizenship law like that of France, automatically transforming
second-generation immigrants into citizens, presupposed confidence in
their effective assimilation. The French elite possessed that confidence:
the German elite did not.
In rejecting an instrumentalist account of French and German citizen-
ship policies and practices, then, I do not wish to offer a naively culturalist

account in its place. Instead, I argue that particular cultural idioms-ways

of thinking and talking about nationhood that have been state-centered and
assimilationist in France, and more ethnocultural and differentialist in
Germany-were reinforced and activated in specific historical and insti-
tutional settings; and that, once reinforced and activated, these cultural
idioms framed and shaped judgments of what was politically imperative,
of what was in the interest of the state. Understandings of nationhood and
interests of state are not antithetical categories. State interests in an
expansive or restrictive citizenry are not immediately given by economic,
demographic, or military considerations. Rather, judgments of what is in
the interest of the state are mediated by self-understandings, by cultural
idioms, by ways of thinking and talking about nationhood.
The more general analytical point is that cultural idioms are not neutral
vehicles for the expression of preexisting interests: cultural idioms consti-
tute interests as much as they express them. These culturally mediated and
thereby culturally constituted interests do not exist prior to, or indepen-
dently of, the cultural idioms in which they are expressed. As Gareth
Stedman Jones has argued,

We cannot ... decode political language to reach a primal and material expres-
sion of interest since it is the discursive structure of political language which
conceives and defines interest in the first place. What we must therefore do is to
study the production of interest, identification, grievance and aspiration within
political languages themselves. (Stedman Jones 1983, p. 22)

I do not subscribe to Stedman Jones's purely culturalist perspective.

Idioms of nationhood, as I have suggested above, are ultimately rooted in
political and cultural geography; and they are proximately rooted in, and
reinforced by, experiences and practices that, while linguistically medi-
ated, are not reducible to speech acts. If it is necessary to "study the
production of interest ... within political languages," it is also necessary
to study the social production and reproduction of political languages
themselves. Yet Stedman Jones provides a powerful argument for attend-
ing to the way in which cultural idioms constitute rather than merely
express interests.
"Not ideas," wrote Weber, "but interests-material and ideal---directly
govern men's conduct. Yet very frequently the 'world images' that have
been created by 'ideas' have, like switchmen, determined the tracks along
which action has been pushed by the dynamic of interest" (1946, p. 280,
translation modified; 1922, p. 252). Differences in citizenship policies and
practices are not produced exclusively or immediately by differing under-
standings of nationhood. Of course definitions of citizenship are condi-
Rogers Brubaker 87

tioned by state interests. Yet conceptions of nationhood, to adopt the terms

of Weber's metaphor, have determined the tracks along which the politics
of citizenship has been driven by the dynamic of interests.


Other things being equal, one would expect French and German defini-
tions of citizenship to converge. A Europe without internal frontiers
requires a common immigration policy; it might be thought to require a
common citizenship policy as well. Convergence would seem especially
likely in view of the similar migration processes, comparable immigrant
populations, and converging immigration policies of France and Germany.
There is no doubt, moreover, that the postwar migrations have placed
consideration strain on both French and German self-understandings. The
French tradition of assimilation has been challenged both by the multi-
culturalist left, arguing that immigrants should not be assimilated, and by
the exclusionary right, arguing that they cannot be assimilated. In Ger-
many, on the other hand, even the present conservative government has
had to acknowledge that large numbers of Turkish migrants have become
permanent immigrants, and has proclaimed a public interest in the natu-
ralization of second-generation immigrants.
Yet citizenship policies remain sharply opposed. A French campaign
for a more restrictive politics of citizenship, taken up by the center right
under pressure from the far right, encountered strong resistance, which
ultimately led the government of Jacques Chirac to withdraw its proposed
reform of citizenship law from the legislative agenda in December 1986.
Critics of the proposed reform included the august Council of State, which
characterized it as "contrary to Republican tradition and principles." And
the commission appointed by the government to study the issue proposed
enlarging rather than restricting access to citizenship (Commission de la
Nationalite 1988). In Germany, naturalization policies were liberalized in
1990 for second generation immigrants. But the French system of jus soli
was not even considered: The automatic transformation of immigrants into
citizens is simply unthinkable in Germany-except in the case of ethnic
German immigrants. 13
Why have citizenship policies escaped the convergence to which immi-
gration policies have been subjected? What is special about citizenship?
An important piece of the answer has to do with what is at stake in debates
about citizenship law. Citizenship confers not only political rights but the
unconditional right to enter and reside in the country, complete access to
the labor market, and eligibility for the full range of welfare benefits. In

a world structured by enormous and increasing differences between states

in labor and consumer markets, welfare systems, and public goods such as
order, security, and environmental quality, the rights conferred by citizen-
ship decisively shape life chances (Schuck and Smith 1985, p. 109). For
peaceful and prosperous states, on the other hand, it is vital to limit the
number of persons possessing these powerful rights of access-especially
since demographic, economic, and political differentials on the one hand
and increasingly dense networks of communication, transportation, and
migration on the other have generated unprecedented and increasing mi-
gratory pressures. The question of access to the territories, labor markets,
and welfare systems of the world's favored states is decisive for persons
and states alike (Zolberg 1983).
Yet these weighty questions of access are not at stake in the French and
German debates about the citizenship status of postwar labor migrants and
their families. These debates have concerned the citizenship status of
actual, not potential immigrants-of persons already in the territory, the
labor market, and the welfare system. For these immigrants, and for the
French and German states, the material interests at stake are relatively
minor. Most of the immigrants already enjoy a secure residence status and
broad economic and social rights that differ only at the margins from those
of citizens. Citizenship confers complete protection against expulsion,
access to public sector employment, and eligibility for those few social
services and benefits that are limited to citizens. While not negligible,
these marginal advantages conferred by citizenship over and above those
conferred by the status of long-term foreign resident are of modest import
(Schuck 1989; Brubaker 1989). From the point of view of the immigrants
concerned, then, citizenship status as such does not decisively shape life
Nor does the citizenship status of immigrants seriously engage vital
state interests. The modern territorial welfare state, to be sure, has a vital
interest in the legal, political, and administrative capacity to control access
to its territory, labor market, and welfare system, and the control of access
pivots on the institution of citizenship (Zolberg 1983; Brubaker 1992,
chap. 1). Noncitizens are routinely excludable; citizens are not. The
citizenship status of potential immigrants therefore matters a great deal to
the state. The citizenship status of actual immigrants, however, matters
much less. Noncitizens can routinely be refused admission to the territory;
and noncitizens admitted to the territory on a strictly temporary basis-as
tourists, students, or short-term workers with fixed-duration contracts-
can routinely be refused admission to immigrant status. In both cases,
citizenship is a decisive instrument of closure. But noncitizens who have
been admitted to immigrant status-particularly those who have lived 10,
Rogers Brubaker 89

IS, even 20 years in the country- can no longer be excluded routinely

from the territory,labor market, or welfare system.l" They have politically
and legally protected claims to membership, if not to full citizenship. With
respect to these immigrants, citizenship is no longer a decisive instrument
of closure. Closure based on citizenship remains crucial in the political
domain. But in the social and economic domain-and in the crucial
question of access to and residence in the territory-noncitizen immi-
grants can be excluded only in marginal ways or in exceptional circum-
stances (Brubaker 1989). And since citizenship status is no longer the axis
of routine exclusion, it no longer matters to the state in the same way
whether or not immigrants have citizenship. The state retains marginally
greater freedom of action vis-a-vis noncitizen immigrants than vis-a-vis
citizens, and can still expel them in exceptional circumstances. But this
marginal gain in freedom of action is not of decisive importance to the
state, just as the marginal improvements in legal position are not of
decisive significance for the life-chances of individual immigrants.
It is not enough, therefore, to consider citizenship in its "functional"
context, in terms of its contribution to the life chances of immigrants or
the exclusionary capacities ofthe state. One must also examine citizenship
in its political-cultural context. For if the material stakes of citizenship law
are relatively minor, the "moral" or symbolic stakes are considerable. Citi-
zenship, in a nation-state, is inevitably bound up with nationhood and national
identity, membership of the state with membership of the nation. Proposals
to redefine the legal criteria of citizenship raise large and ideologically
charged questions of nationhood and national belonging. Debates about
citizenship in France and Germany are debates about what it means to belong
to the nation-state. The politics of citizenship today is first and foremost a
politics of nationhood. As such, it is a politics of identity (Cohen 1985), not
a politics of interest (in the restricted, materialist sense). It pivots more on
self-understanding than on self-interest. The "interests" informing the politics
of citizenship are "ideal" rather than material. The central question is not
"who gets what?" but rather "who is what?"
This was particularly clear in the French debate about citizenship law in
1986-1987. Debate centered on what it meant, and what it ought to mean, to
be or become French. It pivoted on self-understanding, not on group or state
interests. Both sides attempted to articulate and mobilize around "a certain
idea of France." For advocates of citizenship law reform, becoming French
was an honor, not a bureaucratic convenience; it was necessary to restore
"will," "value," and "dignity" to citizenship by restricting access to persons
desiring and deserving to become French. France was a nation to be loved
and served, not merely a state dispensing benefits; one should admit as
citizens only persons who were "French at heart," excluding those unwilling

or unable to assimilate. For its opponents, the reform marked a dangerous

departure from the long French tradition of assimilation and inclusion, a
departure inspired by a discredited conception of the nation as a commu-
nity of descent. Advocates of reform, after all, had criticized only the
"automaticity" inherent in jus soli, not that inherent in jus sanguinis; they
singled out for criticism the automatic attribution of citizenship to second-
generation immigrants who were born and raised in France, while endorsing
the automatic attribution of citizenship to children of French citizens, even
those born and raised abroad. To opponents, this preference for the ties of
blood over those of milieu was inconsistent with the assimilationist tradition
of France and distastefully reminiscent of Vichy.
Recent German discussions about immigration and citizenship have also
centered more on identity and self-understanding than on interest (in the
narrow sense). Defenders of existing citizenship law argue that the system
of pure jus sanguinis properly reflects the fact that Germany is not, in a
deep political-cultural sense, a country of immigration. They reject jus soli
in any form as foreign to the German legal and political tradition. They
argue further that Germany cannot be or become a Yielvolkerstaat, a
multinational state; and that the massi ve transformation of immigrants into
citizens would be a dangerous step in that direction (Uhlitz 1986; Stocker
1989; p. 82; Quaritsch 1988). Individual immigrants, in their view, can
become citizens, provided that they are willing to become German; but
naturalization must remain an essentially individual process, not a form of
collective incorporation. Underlying this view is a marked skepticism about
the eventual social and cultural assimilation of immigrants. But this skepti-
cism is shared by all parties to the debate---even those most strongly commit-
ted to the civic inclusion of immigrants. They too find it difficult to imagine
immigrants being or becoming German en masse. Thus while they have
criticized the restrictiveness of German citizenship law, and proposed that
immigrants wishing to acquire citizenship should have the right to do so, they
have not placed great importance on this. For naturalization has assimilation-
ist connotations: To acquire German citizenship, in their minds-and in the
minds of immigrants-is to become German in something more than a merely
legal sense. And they tend to find this both implausible and illegitimate.
Instead of strongly urging the incorporation of immigrants as formal citizens,
inclusionists have sought to articulate a new, postnational political formula
that would allow immigrants to be citizens in Germany without being German
citizens. A system of automatic civic incorporation like that of France has not
been seriously considered in Germany, Many of those committed to granting
immigrants full political and civil rights have hesitated to propose attributing
German citizenship to them: for this, in their view, would constitute a form
of symbolic or cultural violence against immigrants.
Rogers Brubaker 91

The politics of citizenship vis-a-vis immigrants is similar in form in

France and Germany but sharply different in content. In both cases it
pivots on national self-understanding, not on state or group interests; but
the prevailing elite self-understandings are very different. 15 The French
understand their nation as the creation of their state, the Germans their
nation as the basis of their state. There is a strong assimilationist strain in
the prevailing French self-understanding that is lacking in the prevailing
German self-understanding. France is not a classical country of immigra-
tion, but it is a classical country, perhaps the classical country, of assimi-
lation.i'' And schemes of self-understanding referring originally to the
assimilation of the French periphery by the Parisian center have been
easily and in a sense automatically and unconsciously transferred to the
assimilation of immigrants. 17 In the last two decades, to be sure, both the
desirability and the possibility of assimilating immigrants have been
contested, as has the legitimacy of the Jacobin-Republican model of
internal assimilation. Yet the prevailing elite understanding of nationhood
remains more assimilationist in France than in Germany; indeed assimila-
tionist rhetoric has enjoyed a revival in France in recent years. 18 And the
idea of North African immigrants being or becoming French remains much
more plausible and natural than the idea of Turkish immigrants being or
becoming German.
Despite similar immigrant populations and immigration policies, then,
French and German citizenship policies vis-a-vis immigrants remain sharply
opposed. In part, this reflects the absence of compelling state or group
interests in altering definitions of the citizenry. But it also reflects the fact
that existing definitions of the citizenry-expansively combining jus soli
and jus sanguinis in France, restrictively reflecting pure jus sanguinis in
Germany-embody and express deeply rooted national self-understand-
ings, more state-centered and assimilationist in France, more ethnocultural
in Germany. This affinity between definitions of citizenship and concep-
tions of nationhood makes it difficult to change the former in fundamental
ways. In France, the center-right government headed by Jacques Chirac
was unable, in 1986-1987, to adopt even a mildly restrictive reform of
citizenship law, in part because its opponents were able to mobilize
effectively by appealing to the prevailing elite national self-understand-
ing. In Germany, naturalization policies were liberalized in 1990. But
there is no chance that the French system of jus soli will be adopted: the
automatic transformation of immigrants into citizens is simply unthink-
able in Germany. And liberalized naturalization rules alone will do little
to further the civic incorporation of immigrants. Immigrants as well as
Germans continue to associate the legal fact of naturalization with the
social and cultural fact of assimilation, yet neither German political

culture in general nor the specific social, political, and cultural context of
the postwar immigration is favorable to assimilation. Add to this the fact
that dual citizenship is permitted only in exceptional cases, and it seems
likely that naturalization rates will remain quite low, and that the citizen-
ship status and chances of immigrants in France and Germany will con-
tinue to diverge.


1. This paper draws on material appearing in my book Citizenship and Nationhood in

France and Germany (Cambridge. MA: Harvard University Press. 1992) and reprinted here
by kind permission of the publisher. Research for that book was generously supported by
grants from the Graduate School of Arts and Sciences of Columbia University; the French
Government; the Institute for the Study of World Politics. with funds provided by the
Compton Foundation; and the Joint Committee on Western Europe of the American Council
of Learned Societies and the Social Science Research Council. with funds provided by the
Ford Foundation. the William and Flora Hewlett Foundation. and the French-American
Foundation. For comments on preliminary versions of this chapter. I would like to thank
Allan Silver. Klaus Bade. Zsuzsa Berend. Jiirgen Fijalkowski. Klaus-Martin Groth. Bill
Heffernan. Lutz Hoffmann. James Hollifield. Jiirgen Kocka, Jean Leca, Mark Miller. Lliszl6
Nemenyi, Joseph Rothschild. Peter Sahlins, Rudolf von Thadden, Dietrich Thrahardt, and
Aristide Zolberg.
2. On broader patterns of convergence among European labor-importing countries. see
Hammar 1985. pp. 292-304; Castles 1984. p. 9; Miller 1981. pp. 7.15.
3. French officials sometimes stress the need to combat illegal immigration in place of
the need to encourage return migration.
4. "Naturalization" here includes not only discretionary grants of citizenship by the state
but also. in France. the declarative acquisition on citizenship as a matter of right by spouses
of citizens and French-born children of foreign parents. The naturalization data are reported
in greater detail in Brubaker (1992). chapter 4.
5. This formulation recurs in various official statements. for example in the Interior
Ministry's "Aufzeichnung zur Ausliinderpolitik und zum Ausliinderrecht ill der Bundes-
republik Deutschland." February 1988.
6. Thus. for example. Belgium and the Netherlands attribute nationality to third-gener-
ation immigrants. Austria, Belgium. Finland. the Netherlands. Norway. and Sweden permit
persons born in the territory and residing there for a certain length of time to acquire
nationality by simple declaration at or around the age of majority. (Residence alone. not birth
in the territory, is required in Sweden.) In Italy. persons born in the territory and residing
there for 10 years before majority acquire Italian nationality automatically at their majority.
Among European countries that have experienced substantial immigration during the postwar
period. only Switzerland. apart from Germany, has a nationality law based exclusively on
jus sanguinis.
7. The data in this and the preceding paragraph are drawn from Brubaker. 1992.
8. Technically, in order to be legally defined as a German, these immigrants must not
only be ethnic Germans (of German Yolkszugehorigkeity; they must also qualify as Ver-
triebene, that is, as persons "driven out" of Eastern Europe and the Soviet Union because of
their German Yolkszugehiirigkeit, Originally, this category referred to the ethnic Germans
Rogers Brubaker 93

who were physically driven out of Eastern Europe and the Soviet Union in the immediate
aftermath of the Second World War. But German administrative practice has been to consider
virtually all ethnic German immigrants from Eastern Europe and the Soviet Union as
Yertriebene, without inquiring into the actual circumstances of their emigration. Given the
very large numbers of persons who have taken advantage of the status of Yenriebene to gain
entry and citizenship in Germany. the practice of automatically granting the status of
Vertriebene is currently being reconsidered. Even if this practice is modified, however, it
appears likely that the migration of ethnic Germans will continue. On German policies in
this sphere, see Otto 1990.
9. The 1648 Peace of Westphalia. ending the Thirty Years' War, imposed religious
pluralism on Germany and provided a model for the coexistence of distinct cultural commu-
nities. This model, to be sure, permitted assimilation on the level of the individual principality
(according to the formula cuius regio, eius religio J, but on the scale of the nation it was
pluralist and differentialist.
10. See Rokkan and Urwin 1983, esp. pp. 27, 35-38.74-76,96, for suggestive comments
on the geoeconomic and geocultural conditions underlying different patterns of state-build-
ing in France and western Germany. On the roots of the more assimilationist self-understand-
ing of France, see also von Thadden 1987; Elias 1978; and, building on Elias, Noiriel1988,
11. The French-Spanish and French-German borderlands are interesting in their own right
(Sahlins 1989), but they are marginal to national self-understanding. Since Alsace-Lorraine
belonged to Germany during the crucial phase of nation-building in late 19th and early 20th
century France, we cannot know how vigorously assimilationist policies would have been
pursued then. But before and after this interlude, French tolerance for bilingualism shows
that this cultural borderland posed no threat to national identity.
12. The second wave of settlers, from the 16th century on, especially the more numerous
Protestants. tended to preserve their language, religious culture. and national identity (Hagen
1980, pp. 1-9).
13. The recent French and German debates about immigration and citizenship are ana-
lyzed in detail in Brubaker (1992). chapters 7 and 8.
14. By "immigrants" 1 mean persons enjoying a status permitting them, ordinarily, to
remain indefinitely in the country and, outside of the political domain, to participate in social
and economic life on virtually the same terms as citizens. Persons may be admitted to the
territory as immigrants, as is the case of persons who enter the United States as permanent
resident aliens, or they may be admitted to the territory with some other status and later
become immigrants. as in the case of persons who entered France and Germany with
short-term residence and work permits and only later graduated to immigrant status. The large
majority of resident foreigners in European states are "immigrants" in this sense.
15. Popular understandings of nationhood may be much more similar. But the politics of
citizenship depends on elite self-understandings, for formal citizenship-unlike, say, immi-
gration-is not a salient popular issue.
16. The French political and intellectual elite is itself a remarkable product of assimila-
tion. The habitual schemes of thought and expression deployed by professionals in the
representation of the social and political world-politicians, journalists, high civil servants.
intellectuals, and so on-are products of the labor of "continuous normalization" imposed
by centralized institutes such as the Ecole Normale Superieure, the Ecole National
d' Administration, or the Institut d'Etudes Politiques (Bourdieu 1981).
17. On the transfer of schemes of interpretation and understanding from one domain to
another. see Bourdieu 1977, pp. 82-83; and Bourdieu 1984. pp. 170-75.

18. On the left, the idea of assimilation, and even the word itself, has been revived in
response to the rise of Le Pen and to the right's assertions of the unassimilability of
immigrants (Noirie11988, p. 341; Pinto 1988, p. 96; Schlegel 1985).


Anderson, Benedict. 1983. Imagined Communities: Reflections on the Origin and Spread of
Nationalism. London: Verso.
Blanke, Richard. 1981. Prussian Poland in the German Empire. Boulder, CO: East European
Bourdieu, Pierre. 1977. Outline of a Theory of Practice, translated by Richard Nice. Cam-
bridge: Cambridge University Press.
- - - . 1981. "La representation politi que. Elements pour une theorie du champ politique."
Actes de la recherche en sciences sociales 36-37:3-24.
---.1984. Distinction: A Social Critique ofthe Judgement ofTaste, translated by Richard
Nice. Cambridge, MA: Harvard University Press.
Broszat, Martin. 1972. Zweihundert Jahre deutsche Polenpolitik. Rev. and enlarged ed.
Frankfurt: Suhrkamp.
Brubaker, [William] Rogers. 1989. "Citizenship and Naturalization: Policies and Politics."
Pp. 99-127 in Immigration and the Politics ofCitizenship in Europe and North America,
edited by William Rogers Brubaker. Lanham, MD: German Marshall Fund of the
United States and University Press of America.
- - - . 1992. Citizenship and Nationhood in France and Germany. Cambridge, MA: Harvard
University Press.
Castles, Stephen (with Heather Booth and Tina Wallace). 1984. Here for Good: Western
Europe's New Ethnic Minorities. London: Pluto Press.
Challener, Richard D. 1965. The French Theory ofthe Nation in Arms 1866-1939. New York:
Russell & Russell.
Cohen, Jean. 1985. "Strategy or Identity: New Theoretical Paradigms and Contemporary
Social Movements." Social Research 52:663-716.
Commission de la Nationalite. 1988. Etre Francais aujourd'hui et demain, vol. 1. Paris: La
Documentation Francaise.
Dubet, Franeois. 1989. Immigrations: Qu' en savons-nous? Paris: La Documentation Francaise.
Elias, Norbert. 1978. The Civilizing Process. Vol. 1: The History of Manners, translated by
Edmund Jephcott. New York: Urizen Books.
Forschungsinstitut der Friedrich-Ebert-Stiftung, Arbeitsgruppe Auslanderforschung und Aus-
landerpolitik. 1986. Situation der ausldndischen Arbeitnehmer und ihrer Familien-
angehorigen in der Bundesrepublik Deutschland. Bonn: Bundesminister fUr Arbeit und
Groth, Klaus-Martin. 1984. Einbiirgerungsratgeber. Frankfurt: Alfred Mekner.
Hagen, William W. 1980. Germans, Poles, and Jews: The Nationality Conflict in the Prussian
East, /772-1914. Chicago: University of Chicago Press.
Hammar, Tomas. 1985. "Comparative Analysis." Pp. 237-304 in European Immigration
Policy, edited by Tomas Hammar. Cambridge: Cambridge University Press.
JournalOfficiel. 1887. Senat, Debars Parlementaires, Compte Rendu, February 3, 1887.
Lepsius, M. Rainer. 1985. "The Nation and Nationalism in Germany." Social Research
Rogers Brubaker 95

Manfrass, Klaus. 1984. Auslanderproblematik in europliischen Industrielander: Ein Vergleich

Frankreich-Bundesrepublik Deutschland." pp. 758-83 in Auswanderer-Wanderarbeiter-
Gastarbeiter. Vol. 2, edited by Klaus J. Bade. Ostfildern, Germany: Scripta Mercaturae.
Mill, John Stuart. [1984]1949. A System of Logic. London: Longmans, Green.
Miller, Mark J. 1981. Foreign Workers in Western Europe: An Emerging Political Force.
New York: Praeger.
Mitchell, Allan. 1984. Victors and Vanquished: The German Influence ofArmy and Church
in France after 1870. Chapel Hill: University of North Carolina Press.
Noiriel, Gerard. 1988. Le creuset francais. Histoire de I'immigration XIXe-XXe siecles.
Paris: Seuil.
Otto, Karl A., ed. 1990. Westwiirts-Heimwiirts? Aussiedlerpolitik zwischen "Deutschttimelei"
und "Yerfassungsauftrag." Bielefeld: AJZ.
Pinto, Diana. 1988. "Immigration: L'ambiguite de la rU~rence americaine." Pouvoirs 47:93-
Quaritsch, Helmut. 1988. "EinbUrgerungspolitik als Auslanderpolitik?" Der Staat 27:481-
Risl, Ray. 1979. "Migration and Marginality: Guestworkers in Germany and France." Daedalus
Rokkan, Stein and Derek Urwin. 1983. Economy, Territory, Identity: Politics of West
European Peripheries. London: Sage.
Sahlins, Peter. 1989. Boundaries, The Making ofFrance and Spain in the Pyrenees. Berkeley:
University of California Press.
Sayad, Abdelmalek. 1982. "La naturalisation, ses conditions sociales et sa signification chez
les immigres Algeriens." Part II. Greco 13: Recherches sur les migrations internatio-
nales 4-5: 1-51.
Schlegel, Jean-Louis. 1985. "Comment parler de !'immigration?" L' Esprit 102:82-88.
Schuck, Peter H. 1989. "Membership in the Liberal Polity: The Devaluation of American
Citizenship." Pp. 67-79 in Immigration and the Politics of Citizenship in Europe and
North America. edited by William Rogers Brubaker. Lanham, Md.: German Marshall
Fund of the United States and University Press of America.
Schuck, Peter H. and Rogers M. Smith. 1985. Citizenship without Consent: Illegal Aliens in
the American Polity. New Haven, CT: Yale University Press.
Stedman Jones, Gareth. 1983. Languages of Class: Studies in English Working Class History
1832-1982. Cambridge: Cambridge University Press.
Stocker, Hans A. 1989. "Nationales Selbstbestimmungsrecht und Auslanderwahlrecht.' Der
Staat 28:71-90.
Thadden, Rudolf von. 1987. "Umgang mit Minderheiten.' Unpublished paper.
Tims, Richard Wonser. 1941. Germanizing Prussian Poland: The H-K-T Society and the
Struggle for the Eastern Marches in the German Empire, 1894-1919. New York:
Columbia University Press.
Uhlitz, Ono. 1986. "Deutches Volk oder 'Multikulturelle Gesellschaft'? Von den ver-
fassungsrechtlichen Grenzen der Auslander- und Einburgerungspolitik.' Recht und
Politik 3/86:143-52.
Weber, Eugen. 1976. Peasants into Frenchmen: The Modernization of Rural France 1870-
1914. Stanford, CA: Stanford University Press.
Weber, Max. 1922. Gesammelte Aufsiitze zur Religionssoziologie, 2nd. ed., vol. 1. Tilbingen:
- - - . 1946. "The Social Psychology of the World Religions." Pp. 267·301 in From Max
Weber: Essays in Sociology, translated, edited, and with an introduction by H. H. Gerth
and C. Wright Mills. New York: Oxford University Press.

Wehler, Hans-Utrich, 1979. "Polenpolitik im Deutschen Kaiserreich." Pp. 184-202 in Krisen-

herde des Kaiserreichs. 2nd. ed., edited by Hans-Ulrich Wehler. GOttingen: Vandenhoeck
& Ruprecht.
Zolberg, Aristide R. 1983. "Contemporary Transnational Migrations in Historical Perspec-
tive." Pp. 15-51 in U.S. Immigration and Refugee Policy, edited by Mary M. Kritz.
Lexington, MA: Lexington Books.
Citizenship and Welfare: Social Democratic
and Liberal Perspectives

J. Donald Moon

The welfare state, like any concrete form of political life, was built over
generations by different groups, institutions, and individuals, responding
to a variety of problems and seeking different ends. It would be a theorist's
conceit, then, to suppose that we could offer a single theory of the welfare
state, explaining its characteristic policies and institutions in terms of a
delimited range of basic principles and values. 1 Actual welfare states are
based on various and sometimes incompatible values and beliefs, and their
institutions and practices are similarly heterogeneous. Nonetheless, the
arguments of their critics and advocates often express more or less coher-
ent theoretical positions, which frequently enable us to illuminate impor-
tant political issues. The kinds of proposals that are presented, the criti-
cisms they engender, and the dilemmas that policymakers and citizens face
often reflect a particular interpretation of, or set of aspirations for, the
welfare state.
In this chapter I will present two interpretations of the welfare state,
which I will call the "liberal" or "individualist" conception, and the
"social-democratic" or "collectivist" conception.i I offer these with some
hesitation since even at the level of theory we are dealing with consider-
ations that are often shared by different theorists. What I present as
systematic, integrated positions may better be thought of as differences in
emphasis and concern. But by presenting these differences in a somewhat
sharper and more integrated way, I hope to shed light on some of the
characteristic problems and dilemmas we face, and to explicate some
important aspects of the debate over the direction of social policy.




Ideals and Values

It may seem odd to speak of liberal theories of the welfare state at all,
given the historical association of liberalism with theories of limited
government. But the radical individualist, free-market position is a minor-
ity view within the liberal camp. The dominant schools of liberalism have
been broadly supportive of the welfare state, and liberal theorists and
politicians have been among its most important architects.
There are many routes by which liberals come to support an active,
interventionist state that is committed, at least as a matter of practice if
not of constitutional doctrine, to providing a more or less extensive set of
"social rights of citizenship." Perhaps the most characteristic liberal value
is its commitment to autonomy and to moral equality. I have argued
elsewhere that liberalism can best be understood as a response to the
problem of moral pluralism, to the development within society of deep
conflicts among individuals and groups regarding their basic moral and
religious commitments.' By sharply separating the "public" sphere, to be
regulated by universal rules of justice accepted by all citizens, and the
private sphere, within which individuals are free to live in accordance with
their particular moral beliefs, liberalism seeks to find a way for a society
to exist as a moral community, based upon a shared conception of justice,
while at the same time accommodating moral pluralism.
Central to the liberal project, then, is a conception of freedom as
negative liberty, and what might be called "agency rights." Moral plural-
ism is accommodated by providing an extensive sphere of individual
liberty within which people are permitted to pursue their particular visions
of a good or worthwhile life. For public purposes citizens must be seen as
autonomous agents who are capable of directing their own lives in accor-
dance with their own judgments, and so must be accorded the rights to act
as agents. Agency rights include the familiar rights of freedom of con-
science, liberty, property in one's person, and so forth. These "option"
rights delimit areas within which individuals are their own masters, where
they are free to act on their own purposes without interference from others.
These rights are often (but misleadingly) called "negative" rights because
they are commonly formulated as injunctions against interference. But, as
19th century liberals increasingly came to recognize, option rights are by
themselves inadequate and must be supplemented with welfare rights if
citizens are to exercise their capacities for self-direction.
J. Donald Moon 99

Welfare rights are required, in the first place, to make option rights
effective. The right to due process of law may be rendered nugatory for
someone who cannot afford to hire a lawyer; as a result, the "positive" right
to free legal counsel comes to be part of the meaning of due process itself. In
a similar vein, welfare rights can be defended on the grounds that the
commitment to equal liberty requires measures to equalize the "worth" or
value of liberty to different individuals. 4 But the liberal theory of the welfare
state goes beyond these considerations, arguing that people become individ-
uals and develop the capacities for autonomy and self-direction only within
particular forms of society. Values such as equality of respect and human
dignity can be realized in actual, on-going relationships only for people who
are full members of their society, who are able to share in what Tawney called
"a common heritage of civilization." Equality cannot be achieved among
"abstract individuals," but only among people who not only share a way of
life in which conceptions of equality and individuality are central, but who
also enjoy a range of common opportunities and "life chances." Liberal
society is premised on what Marshall ([1950] 1977, p. 101) revealingly calls
the ideal of "equal social worth," which requires that citizens be guaranteed
certain social rights as well as the traditional civil and political rights.
Finally, liberals are led to embrace the welfare state because it is necessary
to overcome the moral shortcomings of the market. As I have argued else-
where (1988), market society is premised upon an ideal of equality and
reciprocity, in which individuals are expected to meet their own needs
(mediated through their membership in families) by offering services or other
goods they can provide in exchange for the services or goods of others. The
centrality of markets in liberal society results from its commitment to auton-
omy and liberty, which require that the individual have property in his or her
own person, and which place a premium on the right and ability of individuals
to shape their own ways of life through the choices they make regarding
(among other things) work and consumption. However, the normal operation
of markets is such that individuals may find themselves in situations in which
they are unable to meet their needs, as a result of unemployment, illness,
disability, or old age, through no fault of their own. The welfare state must
fill this gap in market society in a way that is consistent with its basic norms
and so with the self-respect of its members.
For these reasons liberals are led to support the welfare state. The ideal
of the 19th century liberal state, which provides only a legal order enshrin-
ing the principle of juridical equality, has never described any political
reality. In my account of liberalism, however, it could not even be an ideal
because the minimal state fails in one of the primary tasks of the liberal
state, which is to provide the necessary means for everyone to attain full
membership in society, since that is a condition of our understanding

ourselves as morally equal persons. The liberal state must be a democratic

welfare state in which everyone enjoys a common status of citizenship.
Exactly what citizenship requires (beyond equal civil and political
rights) has been a contentious issue for more than a century. One of the
most obvious requirements is universal education, freely available to a
level considered "normal" for the society, and the provision of public
education is always among the first social rights that liberal governments
provide. It also requires regulation of the market to ameliorate the socially
adverse effects of essentially self-interested exchanges. And it requires the
provision of a level of material well-being that is sufficient to enable
people to live at least minimally "respectable" lives. What this standard
requires changes over time and is subject to conflicting interpretations.
Deciding what constitutes the social rights of citizens and designing
policies to realize them have been central political issues in liberal demo-
cratic countries for more than 50 years.

Liberal Policies and Problems

Liberal welfare states must cope with two dilemmas in formulating
social policy-the dilemma of neutrality, and the dilemma of provision.
The first arises from the liberal commitment to negative liberty and its
response to the existence of cultural and moral pluralism. Although it is
an exaggeration to say that the liberal state is committed to an ideal of
"neutrality" among the different conceptions of the human good, or the
ends and purposes of life, that is a good starting point for considering the
dilemmas of the liberal welfare state. Ideally, liberal governments provide
a framework of laws that regulate the interactions of individuals who
pursue their own ideals and values. Much has been made of the metaphor
of the "night watchman" state, but a better image might be the "traffic cop"
state. Like the traffic cop, the purpose of the state is to regulate people's
movements in such a way that they can reach their destinations safely, not
to tell people where they should be going.
Even as a description of the "minimal state" of classical liberalism, this
picture is misleading. As an account of the liberal welfare state, it is
unsustainable. By defining a set of welfare rights, or social rights of
citizenship, the state implicitly sets up certain ends as worthy of pursuit
and disvalues others. By basing policy on certain views of what kinds of
needs and opportunities citizens must have in order to be full members of
the society, leading lives of dignity and self-respect, the state establishes
certain values as worthy of promotion. Rather than simply respecting the
choices of individual citizens, it necessarily defines a certain range of
choices as appropriate ends for them to pursue. The welfare state, then,
J. Donald Moon 101

involves at least an implicit commitment to some idea of positive freedom.

But this leads to a deep tension with its commitment to negative liberty
and the agency rights of citizens.
In America today there are many examples of this conflict. For example,
many argue that welfare policy should be based on an ideal of personal
responsibility, in which individuals are expected and encouraged to take jobs
in the paid work force, and to provide for their children. Welfare policy is not,
nor could it be, neutral with regard to expectations surrounding gender, sexual
behavior, and parenthood. Similarly, in the area of medical care, we have to
decide what kinds of services must be provided, and in some cases to rank
different kinds of procedures, in order to determine which will be made
available through public provision. Such choices, however, could hardly fail
to reflect our judgments as to the proper ends of life. In the case of abortion
and family planning, these conflicts are all too evident.
The second dilemma the liberal welfare state faces is the dilemma of
provision. Not only must the state provide certain welfare rights to its
members, but it must do so in a way that is consistent with their maintaining
self-respect, consistent with the values and norms characteristic of a liberal
and market society. I alluded to this dilemma above when I discussed the
moral shortcomings of the market. Because individuals in liberal society are
expected to meet their own needs, social provision always runs the risk of
stigmatizing its beneficiaries by marking them as "failures," as people who
are unable to discharge the normal expectations the society holds of its
members. The liberal welfare state's response to this dilemma is to use the
principles of social insurance and universal provision of certain goods (such
as education and medical care) to provide for the social rights of citizenship.
Social insurance, which provides benefits to individuals who face such
contingencies as illness, old age, or disability, is consistent with the norm of
reciprocity, because benefits are "earned" through contributions. Similarly,
universal provision is acceptable because it does not single out any particular
group for special treatment. Although these mechanisms enable the liberal
state to provide for an important range of needs, they are not adequate to
satisfy all of the social rights of citizenship.f


Ideals and Values

To speak of "the" social democratic conception of the welfare state is
even more problematic than to speak of the liberal conception. Social

democracy is much younger and less well defined than liberalism. At the
edges, it blurs either into liberalism or into a more full-blooded socialism,
raising doubts as to whether it is a distinctive political position at all. Once
again, I would emphasize that my interest is not in discovering "essences" of
different political tendencies, but in highlighting clusters of related ideas that
can help us to understand conflicts over the design of social policy.
In the preceding section I suggested that liberals come to the welfare
state because of their commitment to agency rights and as a result of the
moral failures of markets, a central institution of liberal societies. In their
vision, the welfare state plays a limited, even a "residual" role, in correct-
ing the problems arising from core liberal institutions and principles. For
social democrats this relationship is reversed. They begin from a concern
with social solidarity and need, rather than individual rights and reciproc-
ity. For them, public provision is not a supplement to market society, but
the core function of the state. Although they may allow market transactions
to coordinate the bulk of economic activity, they view markets as distinctly
subordinate to the realization of collective purposes and to nonmarket
forms of organization and provision. In the revealing title of Esping-
Andersen's book on social democracy (1985b), for social democrats it is
"politics against markets." At an extreme, some social democrats are
unremittingly hostile to markets; their ideal is a society in which all
activities are organized through voluntary, cooperative efforts, and in
which individuals are oriented principally toward common needs and
aspirations. Titmuss (1972) extols the "gift relationship" and Harris (1987)
speaks of the family as a model for social life. More concretely, Offe
(1984) and Esping-Andersen (1985b) hope that the growth of collective
consumption and public control will eventually displace capitalism, lead-
ing to a socialist order of society.
This commitment to decommodification has a number of sources, but
two of the most important are equality and citizenship. Once again, social
democrats and liberals are not completely opposed with respect to these
values. But where liberals focus on juridical equality-equality of rights
and equal respect-and see the need for a limited amount of redistribution
in order to make this ideal effectively part of our social experience, social
democrats see equality of life chances as a good in itself. Conceiving of
"society" as responsible for distributing goods and opportunities among
its members, and recognizing that the institutions through which distribu-
tions occur are human constructions, not natural necessities, social demo-
crats hold that institutions should equalize the life chances of different
individuals. Since needs differ, and since different individuals have dif-
ferent natural endowments, actual distributions of goods and services
would not usually be equal, but the aim should be to equalize the circum-
J. Donald Moon 103

stances of all citizens. Where liberals worry that social policies may go
too far in weakening the linkage between individual performance and
outcome, social democrats seek to sever that tie as much as possible. They
bemoan the failure of welfare states to effect a greater redistribution of
income; they are distressed by the ability of privileged groups to utilize
its programs more effectively than the less privileged; and they object to
the way in which many programs provide income and benefits that are tied
to earnings, thereby reinforcing patterns of Inequality."
This concern with equality has many sources, but one of the most
common is the commitment to equal citizenship, which is sometimes
described in terms of solidarity or membership. This is a major theme in
Michael Walzer's (1983) account of the welfare state; indeed, he argues that
the purpose of all political associations is to meet the needs of their members.
Of course, different societies accept different accounts of human needs,
but once something is recognized as a need, then the goods necessary to
satisfy it should be distributed in accordance with the extent of need and
in a way that recognizes and upholds the status of its members. In a society
committed to equal citizenship, the necessary goods and services should
be universally provided in a way that equally satisfies everyone's needs
and without stigmatizing any citizen (1983, chap. 3 generally).
In a similar vein, David Harris argues that "full membership" in a
society requires that each person be able to enjoy "a certain style of life"
and "certain life chances" (1987, p. 147). Although he recognizes that
modem societies include a plurality of different groups, he insists that
there are more or less common standards of what one must be able to do
and how one must be able to live if one is not to be excluded or socially
marginalized. These standards determine the needs of members of that
society, and should be equally available to all citizens as a matter of right.
The provision of welfare rights involves a corresponding obligation on the
part of the society as a whole, and of its various members, to provide the
necessary goods and services (or to support those institutions which can
provide them [Plant 1988, p. 73]). But why do we have such obligations, and
why are they more urgent than other goals we might pursue individually or
collectively with the resources at our disposal? To answer these questions,
Harris ultimately appeals to the idea of membership itself, drawing an analogy
between political society and the family. Just as we have obligations to
members of our family, irrespective of what they may have done for us
individually, so we have obligations to our fellow citizens.
It is easy to see why family relationships may give rise to obligations-
given the centrality of such relationships to one's identity, and the ties of
affection and shared histories. But it is far from obvious that common
citizenship gives rise to such obligations. Harris concedes that support for

an extensive welfare state depends upon the development of "certain

sentiments of community" (p. 163), but that doesn't go very far toward
justifying the social democratic model. Granted, if we have such feelings,
we will support an extensive welfare state. But why should we have or
undertake to develop such feelings?
One of the main weaknesses of many social democratic theories of the
welfare state is their failure to answer (or even to confront) this question.
The appeal to equality based on solidarity or fellow feeling often is made
in highly moralistic terms, or rests upon misplaced analogies to families
or to other genuine communities. Ironically, welfare states have a system-
atic tendency to undermine the very communitarian sentiments and rela-
tionships that would support the values of solidarity and equality. Harris
(1987), for example, argues that the universal provision of certain services
is expressive of, and may contribute to, a sense of community and equal
citizenship. But more often than not what we actually experience is being
reduced to the status of a client, attempting to meet our needs through an
impersonal and unresponsive bureaucracy. Far from contributing to a
sense of community, public provision (which is almost inevitably bureau-
cratic provision) disrupts the communitarian forms through which needs
may have been met in the past, replacing personal relationships that
engender obligations and mutual identification with legally prescribed
associations of strangers.'
Social democrats have an answer to this critique: democracy. Many
social democrats stress the social in social democracy, conceiving of
citizens as clients and consumers of services to which they have "rights,"
and spend a great deal of energy trying to explain where these rights come
from and what they require.f But a more powerful response is to stress the
democracy in social democracy, conceiving of citizens not as consumers
or clients, but as active participants in shaping the conditions of their lives
by self-consciously controlling and directing the processes of social pro-
duction and reproduction. As consumers of welfare services, our relation-
ships to our fellow citizens are too passive to sustain the obligations of the
social democratic welfare state. But as participants actively working with
others to shape and direct our common lives, we are (at least potentially)
bound to them by important ties and experiences, giving rise to mutual
identification and a sense of a shared fate-sentiments of solidarity and
equal membership that the social democratic model requires.f
Michael Walzer has stressed the importance of democracy in the welfare
state, arguing that social provision should itself be "socialized" through
expanded citizen participation in decision making and "in the actual
delivery of welfare services" (1988, p. 20). Drawing an analogy to the
disappointments that have accompanied socialist programs of nationalizing
J. Donald Moon 105

industry, Walzer points out that the nationalization of consumption through

the welfare state has also led to "centralized control, bureaucracy, [and]
uniform regulation" (p. 14). Just as nationalization by itself leads to the
enhancement of state power, "not the power of ordinary people" (p. 16),
so the socialization of distribution should involve "collective help-which
is the work not only of bureaucrats spending taxpayers' money but also of
citizens 'spending' their own time and energy" (p. 17). It is not just the
provision of welfare services that should be democratized, but the society
as a whole. All spheres within which power is exercised should be trans-
formed to make its use accountable to "those who most immediately
experience its effects" (1983, p. 285). For Walzer, this includes not only
governmental power, but also power in the workplace. The importance of
economic democracy is also emphasized by Esping-Andersen, who sees it
as the goal of a properly constituted social democratic strategy: " 'Political
citizenship' must precede 'social citizenship,' and these are in turn indis-
pensable for the third stage, 'economic citizenship.' Workers must be eman-
cipated from social insecurity before they can partake effectively in economic
democracy" (1985b, p. 22). Indeed, Esping-Andersen argues that without a
concerted push for economic democracy in successful social-democratic
regimes, "the trend toward social democratic party decomposition will con-
tinue" (p. 324), perhaps imperiling its previous successes.
The importance of democratization is reflected not only in social dem-
ocratic theory, but also in the practices of the Scandinavian countries
where social democracy has achieved its greatest success. Hernes de-
scribes "social democratic citizenship" as an "activist, participatory, egal-
itarian ideal" in which "participatory rights and material entitlements"
have become an aspect of "almost all relationships between the state and
the individual within a great variety of institutional settings" (1988, pp.
,200, 199). This stress on participation, Hernes observes, is fairly recent,
as the focus of "social democratic practice and legislation" had previously
been "the desire to secure material rights and entitlements" (p. 201).
However, the very success of this program created "the potential for
paternalism" and "the dangers of clientilization," which are being combat-
ted by the addition of "participatory rights ... to various public roles as
client, consumer and worker" (p. 203). This has taken place against the
background of a "corporate system" in which significant powers have been
devolved to decentralized settings. It is important to note that these roles
are not confined to the state sector as such, but include "the corporate
system, the public sector and the amalgam of social movements and
political parties of democratic (partly electoral-based) institutions" (p.
208). The pattern, then, is one of "segmented citizenship" in which citizens
face participatory opportunities in a variety of settings.

Social Democratic Policies and Problems

Given the goals and principles described above-decommodification,
equality, equal citizenship or solidarity, and the democratization of soci-
ety-social democrats tend to support certain kinds of programs and to
weight certain wogram characteristics differently from the way liberals
evaluate them. 0 In particular, social democrats favor social provision
through universal programs providing specific services (Le., in-kind rather
than cash), with limited or no opportunities for exit (i.e., state monopoly
of provision). Social insurance, involving contributions and benefits tied
to individual performance, is de-emphasized, and means-tested programs
directed toward the relief of poverty is strongly devalued. The range of
services provided through the state should be greater than in liberal
systems, where state provision is secondary or even residual, and "self-
help" provisions should be minimized. What is critical in this model is the
collectivization of benefits under conditions of equality. By avoiding
divisions among recipients, by breaking links between benefits and con-
tributions, by structuring programs in such a way as to dissociate individ-
ual performances from individual outcomes, by denying opportunities for
indi viduals to meet their needs outside of these programs, social democrats
seek to reinforce a sense of shared fate and social solidarity, and to realize
equality among citizens. By combining these programs with extensive
opportunities for participation, social democrats hope to create the condi-
tions for citizens collectively to control significant aspects of their lives.
As with the liberals, these commitments to programs and principles give
rise to certain moral dilemmas. II Perhaps the most obvious is what we
might call the dilemma of democratization, which arises in part from the
tension between bureaucracy and participation. The pressures for bureau-
cratization result from the need for uniformity if services are to be deliv-
ered as a matter of right, and the necessity of expertise in many social
welfare programs. Walzer (1988) speaks of the need to "balance" the
conflicting imperatives of bureaucracy and participation, and Heroes (1987)
points out that "participatory rights" are "not necessarily decision-making
rights but rights to information and to be heard" (p. 203). The right to be
heard is certainly important, but it is not clear that it can carry the kind of
weight the social democratic model imposes on it.
The tension between bureaucratization and democratization also arises
from the range of settings in which citizen participation is called for.
Hernes describes a "segmented and fragmented citizenship," and ex-
presses concern about the willingness and ability of citizens to "live up to
such high participatory ideals" (p. 203). And with the decentralization and
J. Donald Moon 107

fragmentation necessary to make participation possible, there is the further

question of coordination among various institutions and processes, and the
ability of any agency to make significant and effective decisions. Rather than
providing for effective participation, the process of devolution may involve
"a loss of power on all parts, and certainly on the part of [electorally based]
political institutions" (1987, p. 207). To the extent that occurs, the effort to
broaden democracy may have the paradoxical effect of weakening it. 12
The second dilemma social democracy encounters is the dilemma of
autonomy and empowerment. Social democrats are committed to empow-
erment, to enabling everyone to participate effectively in public affairs.
Such participation requires a strong sense of individual efficacy and a
capacity for dealing with cognitive complexity. These abilities are fos-
tered by the experience of individual autonomy and choice, and so rest on
the protection of option rights. However, the social-democratic model
restricts option rights in order to enhance security and equality, thereby
posing a threat to autonomy and, ultimately, empowerment itself. One
obvious manifestation of this is the often expressed concern about the
growth of clientelization and passivity resulting from the pervasiveness of
large bureaucratic organizations. Collective empowerment requires indi-
vidual autonomy, and so rests upon--even while limiting-a significant
sphere of negative liberty.13
Perhaps the most difficult dilemma for social democracy is the tension
between its commitments to decommodification and solidarity. Decommodify-
ing labor is supposed to free the individual from subservience to the
market. But where this has been achieved, notably in Sweden, "market
necessity" has been replaced by "political necessity," as receiving unem-
ployment benefits is contingent upon the worker's accepting a job (or a
place in a training scheme), thereby reinforcing his or her ties to the labor
market. Similarly, the commitment to solidarity requires the use of uni-
versal programs, but the existence of differentiated needs, resulting from
differences in the income that different workers earn, means that higher
paid workers would not be satisfied with the benefits available to all. This
has resulted in the creation of programs-particularly pensions-that tie
benefits to income, thereby reproducing the status differentials of the
market within the supposedly solidaristic and universalistic programs of
the welfare state. As Esping-Andersen argues, this may avoid dualistic
systems in which some rely primarily upon public provision while others
rely upon private plans to supplement the public program, but it does so
by building unequal provision into the public plan itself (1990, p. 48ff.).


For both liberal and social democratic theory targeting social benefits
on the basis of income is problematic, but for different reasons. Ideally,
both theories call for social benefits to be provided to people through
universal programs on the basis of their meeting certain criteria of entitle-
ment, including being retired, ill, disabled, unemployed, having the care
of a child, or requiring education. By providing benefits only through
programs that meet needs arising from widely or universally shared life
contingencies, the welfare state is supposed to secure the social rights of
citizenship in a way that treats all citizens equally. Unfortunately, realiz-
ing this ideal is very difficult. All welfare states, even the Scandinavian
systems long dominated by Social Democratic parties, rely to some degree
on means-tested programs of social assistance to meet the needs of people
whose benefits under universal programs are inadequate. In the past 15
years of slow economic growth, the gaps in universal programs have
increased, and more and more people have been forced to rely upon social
assistance. Moreover, the characteristics of those who require social assis-
tance are similar in all societies-elderly women, ethnic minority groups, and
various categories of seasonal and low-paid workers; in the past decade,
more people have fallen into this category, particularly with the rise of
single-parent households and growing long-term unemployment.!"
Targeting is a particular problem for social democrats, as it directly
contradicts their commitment to solidarity. By reinforcing market-based
distinctions among people-between those who have successfully formed
stable attachments to the labor force and those who have not, or whose
wages and working conditions are poor-social assistance undermines the
sense of common interests necessary to sustain programs of decommod-
ification and democratization. Since it creates an identifiable class of
recipients whose benefits are made possible by the taxes paid by others, it
sets up an opposition of interests at the heart of the welfare state, generating
political pressure to cut such benefits and to dilute the quality of services
provided. In recent years it has led to renewed attacks on the moral failings
of recipients, particularly regarding their willingness to work, culminating in
demands that they be forced to work as a condition of receiving benefits (an
issue to be discussed in the next section). By engendering conflict and
legitimating market-based distinctions, social assistance is a primary culprit
in the process of "destructuration" of the welfare state that Offe (1988)
describes, undermining popular support for moves toward the full democra-
tization of society, including economic democracy. IS
J. Donald Moon 109

The problem of stigmatization is a serious concern for liberal theory,

but for somewhat different reasons than for social democrats. 16 As I have
argued elsewhere (1988), means-tested income maintenance programs
violate the norm of reciprocity that is central to liberal society. Further-
more, to the extent that such programs lead to political pressures that cause
the level of support to be inadequate to meet the social rights of citizen-
ship, liberals have instrumental reasons to oppose means-testing. But
liberals also have reason to support means-tested programs. In the first
place, their commitment to autonomy and individual choice militates
against universal programs, particularly those providing in-kind benefits,
because they restrict the range of choices available. This is particularly
true when universality is combined with a prohibition on private provision,
making the state the monopoly provider of a particular service. While such
restrictions on choice may only directly harm those who can afford to
make private provision, it is still a "cost" in terms of liberal or individualist
values. 17 Moreover, in certain areas such as housing or education, a large
proportion of the population can be affected.
Second, in a liberal framework, a program will be stigmatizing only
when it in some way singles out its recipients as having failed to meet
some duty or expectation, for it is this failure that constitutes the reproach
or disgrace that stigmatization involves. But not all targeted programs are
ofthat sort. To take an obvious example, need-based support for university
students should involve no disgrace for the student or his or her family.
Not only is there no expectation that students (or their families) should be
able to provide higher education for themselves, but the value attached to
"equality of opportunity" specifically licenses support for educational
programs of all sorts.
The problem of stigmatization is most severe in programs of general
income support for able-bodied people who are not covered by social insur-
ance. Programs designed to target this group run a serious risk of branding
people as failures, given the norms of reciprocity and the expectation that
healthy adults will support themselves. Welfare states can minimize this
problem, but it is extremely difficult to eliminate it entirely, and none has
done SO.18 But doing so will be more difficult for liberal welfare states than
for social democratic ones, because liberal systems will seek to reduce as
much as possible the total proportion of society's income that is channeled
through public programs, the opposite of the social democratic objective, and
because the liberal system will have a preference for cash rather than in-kind
aid. Because it pursues multiple objectives, it will have to balance concerns
with stigmatization against other values.

The Problem of Obligation

One of the first things students learn in introductory classes in political
theory is that rights and duties are correlative. Not surprisingly, accounts
of the obligations that are correlative with the social rights of citizenship
vary dramatically with ideological position. Those on the left concentrate
on the responsibility of advantaged individuals to provide for the needs of
the disadvantaged, while conservatives concentrate on the responsibilities
of the "poor" to work, attend school, and provide for their families. 19 In
this section I will examine the ways in which obligations appear in the
framework of liberal and social democratic theories. As we shall see,
obligations present difficult problems for both liberals and social demo-
crats because they heighten the dilemmas of both positions.
One of the best known discussions of obligation in the context of the
welfare state is Mead's (1986) Beyond Entitlement: The Social Obligations
of Citizenship. Although there are elements in Mead's work that are broadly
compatible with liberal or social democratic conceptions, the dominant
themes of his work are opposed to both of these theories. For liberals, this is
particularly true of one of the main accounts that he gives of the relationship
between citizenship and obligation. Mead frequently speaks of the "social
obligations of citizenship" as a set of norms that defines adequate social
functioning, to which citizens must conform. One of these norms is to work,
in the sense of holding gainful employment, but the range of obligations is
much wider, including supporting one's family, being literate in the official
language of one's country, learning enough in school to be employable, and
showing respect for the rights of others (pp. 242-43).
Mead frequently claims that these obligations should be enforced on
paternalistic grounds, for the good of the individual involved, because
without such enforcement some of the recipients of state welfare will fail
to "function" adequately. However, there is no real evidence offered for
this claim. He does not explain what human functioning involves, nor does
he show how observance of these norms would enable individuals to
realize the human good, to achieve adequate human "functioning." (In-
deed, he does not even suggest that independently wealthy people be
required to work, nor does he show any concern about their low level of
human functioning resulting from their failure to work!) Moreover, he
offers no argument to show that enforcement of these norms would actu-
ally increase functioning except in the trivial sense that it would increase
participation in work. But would such forced participation actually be
"good" for people? Would it in fact increase their "functioning"? Without
addressing these issues, Mead's paternalistic argument for enforcing the
social obligations of citizenship fails.
J. DonaldMoon 111

Even if Mead had answered these questions, this line of argument would
be unacceptable to liberals. As I argued above, a fundamental tenet of
liberalism is to respect the different choices that individuals make about
the ends and directions of their lives. It thus denies the legitimacy of
government policies expressly designed to promote certain conceptions of
proper human functioning, not to mention paternalistic attempts to enforce
them. While social democrats may not object as strenuously to Mead's
paternalism as liberals do, they are unlikely to find the grounds on which
he would defend his proposals palatable.
But Mead (1986) also argues that working (and observing other norms)
is required by considerations of fairness. Since taxpayers have to work,
why not everyone who is able to do so? Indeed, since liberals view the
welfare state as an adjunct to the market, ameliorating its moral defects,
they are committed only to providing material support to those who are
either unable to work (because of age, disability, etc.) or to those who are
unable to find adequate employment. Thus programs tying social provi-
sion to work are in principle acceptable within liberal theory, at least so
long as this is done in a way that respects individual dignity and does not
force people to take jobs that involve conditions of degradation or exploi-
tation that deprive them of effective citizenship. As I will argue below, at
least some programs of welfare reform involving work requirements are
acceptable in terms of liberal political theory.
Social democrats also would find an obligation to work based on an
appeal to fairness to be plausible. Given the social democratic opposition
to the commodification of labor, some kind of obligation to work would
seem to be essential. In Sweden, accepting a job or a place in a training
program is a condition for the receipt of unemployment benefits, and
Hernes notes that one of the duties of social democratic citizenship is "to
accept ajob even if this means moving home and family" (1988, p. 201).20
Unlike Mead, however, social democrats would qualify this obligation in
a number of crucial ways. In particular, they would insist that the obliga-
tion to work not only be nonpunitive, but it should also be incorporated
into genuinely universalistic programs, such as unemployment compensa-
tion. "Workfare" tied to social assistance or means-tested programs is
objectionable because it reinforces the separation between those who
supposedly "deserve" social support (due to disability, age, etc.) and those
who do not, and who must therefore be compelled to work. As Katz puts
it, "Workfare is a nonresponse to the structural sources of poverty in
America. It addresses the politics of poverty, not its roots" (1989, p. 232).
Second, social democrats would insist that enforcing work obligations
is acceptable only when the government pursues economic policies de-
signed to bring about full employment and to secure a high minimum

wage. 2 1 Social democratic commitments to decommodification, to the

democratization of many nongovernmental areas of social life, and to
equality and solidarity, put the social democrat in something of a quan-
dary. On the one hand, solidarity and equality require that everyone be
integrated into the major activities and structures of society-including,
in advanced economies, paid work. On the other hand, social democrats
wish to restrict the power of capital and the scope of competitive, individ-
ualistic-oriented activities. In the absence of strong unions, tight labor
markets, and high minimum wages, the expansion in the labor supply that
would result from enforcing work obligations would be counterproductive
to these goals. Thus we can expect social democrats to oppose policies
establishing and enforcing obligations to work except in a highly restricted
set of circumstances.



In concluding this chapter I would like to analyze one carefully framed,

integrated proposal for addressing the key issues involved in welfare
reform, that offered by David Ellwood (1988) in Poor Support. I will argue
that Ellwood's proposal fits squarely within the liberal tradition, and go a
long way to overcoming the dilemmas of provision that I sketched above.
Because the policies he recommends strongly reinforce liberal values, I
will show that they are problematic from a social-democratic point of view
(although most social democrats would see them as a distinct improvement
over what we have now).
Ellwood proposes to abolish welfare, in the sense of permanent income
support to able-bodied individuals who do not work, and to replace it with
programs to provide medical care, child support, jobs, and income supple-
ments enabling all who work to have adequate incomes to escape poverty. In
addition, his program would give all citizens the right to receive "transitional
assistance" for a limited period of time, including cash benefits for income
maintenance, job training or education, and certain social services. Many of
these benefits would be provided through the tax system by changing deduc-
tions for dependents and for child-care expenses into refundable tax credits,
so that they will have the same cash value to low-income people as to
high-income people. Similarly, Ellwood would dramatically expand the
"earned income tax credit," making it function like a negative income tax, but
restricted to those who work. These measures would enable all employed,
two-parent families to escape poverty through their earnings, supplemented
by transfers through the tax system.
J. Donald Moon 113

Single-parent families would receive assurance of child support through

a system that would automatically deduct child support payments from the
noncustodial parent's wages, and make payments to the custodial parent.
A minimum payment, financed from general revenues, would assure all
families of child support in the event that the noncustodial parent does not
have adequate earnings, or is disabled in some way.
The key to Ellwood's program is the emphasis it places on work and
family responsibility. His aim is to design a program that will, as much as
possible, both express the basic values and commitments of American
society, and reinforce those values. His critical assumption is that, with a
strong program of transitional assistance enabling people to acquire skills
and prepare for work, and with income-support programs that "make work
pay," jobs can be found for virtually all who are capable of working. This
is not an unreasonable assumption, even in an economy in which 5%-6%
of the work force is unemployed. Although it is impossible to eliminate
joblessness in a market economy, it is at least in principle possible for
people who lose their jobs to find new ones within a reasonable period of
time-say 3 to 6 months. With an adequate program of unemployment
insurance, such transitional unemployment is compatible with the moral
imperatives of liberal theory. As long as the queue of jobseekers keeps
moving, so that everyone can reach the head of the line without waiting
too long, Ellwood's work-oriented program combined with unemployment
insurance should enable everyone to realize the social rights of citizenship.
The problem with this proposal is that there will be some who will not
be able to find work in normal labor markets. In some cases, such as severe
recessions or regionally depressed economies, there may not be enough
jobs to keep unemployment at manageable levels. But even apart from
such cases, there will be some individuals whose skills and/or motivation
are so poor that they will be unable to find and hold regular jobs. For them
Ellwood proposes a program of guaranteed jobs, with the government
serving as the employer of last resort. It is easy to see why such a program
is essential, but hard to see how it would work (except in cases of general
or regional economic downturns). If guaranteed jobs are real jobs, jobs
that we want to have done, then administrators will seek qualified and
motivated workers, the kind of workers who are likely to be able to find
employment on the regular labor market. If this program is supposed to
provide goods and services that meet real needs, then a certain level of
performance will be required, and those failing to meet it must be subject
to dismissal. But how can we fire people from the jobs that are essential
to their survival? On the other hand, if the standards for these jobs are set
sufficiently low, then it is hard to see how that would avoid the kind of
stigmatization that currently attaches to welfare or social assistance.

It is important not to exaggerate this problem. A great deal depends on

the number of people who would be served by the program, and how long
it would take them to move from a government guaranteed job into the
regular job market. Moreover, even if the prognosis on these points is
fairly bleak, this program would represent a significant improvement over
our current system. Nonetheless, the problem of residual joblessness and
the guaranteed jobs program represents a potentially serious difficulty
from both liberal and social democratic perspectives.
Ellwood's program goes a long way to realizing the liberal ideal of the
welfare state at least with regard to income maintenance. It does involve
a high level of targeting, and a strong political commitment to the idea that
citizens have obligations to work and to discharge their family responsi-
bilities. However, the use of the tax system to provide income support for
working families and families with children is not stigmatizing, since these
reforms can be justified on the grounds that they make the tax treatment
of child-care expenses and the allowance for dependents equal for all
citizens. Under the current system, the value of these deductions is greater
for higher income taxpayers than for lower income individuals. While the
expansion of the earned income tax credit involves a more direct element
of redistribution, it does not violate the expectation of reciprocity on which
liberal society is based. While individuals are expected to support them-
selves by offering their services on the market, it is widely recognized that
market outcomes are distorted in various ways, and that any individual's
"return" will be, to a significant degree, a matter of luck. If what is
important to the ideal of reciprocity is that one provide services in return
for the goods and services provided by others, and not the precise terms
of exchange, Ellwood's proposal, unlike a negative income tax or guaran-
teed annual income, is fully compatible with liberal values. I might also
add that this use of the tax system has the further advantage of avoiding
unnecessary stigmatization by keeping the identities of people aided under
the program private.
The strong commitment to social obligations of citizenship in Ellwood's
proposal is more problematic from a liberal perspective. This is particu-
larly true of the paternalistic justifications for enforcing obligations that
Ellwood offers at a number of points. To the extent that policies are
intended to promote the good of the individual, they run foul of the
"dilemma of neutrality" outlined above. Nonetheless, obligations to work
(and to support one's children) can be justified on grounds of equity and
need not rest on paternalistic considerations.
Many of the considerations that make Ellwood's proposals so satisfac-
tory from a liberal perspective have the opposite effect when considered
from a social democratic perspective. The strong targeting his program
J. Donald Moon 115

involves, although superior to the present system, still divides the popula-
tion into a large group of contributors and a minority of recipients. This
builds a tension into the heart of the program, in which the interests of
some will be to expand the size of the redistributive transfers whereas
others seek to limit them. The forging of alliances based on equality and
solidarity of interests may therefore be inhibited. This problem is exacer-
bated by the proposal to limit transitional assistance, to abolish all perma-
nent income support, to create a guaranteed jobs program, and the strong
use made of cash transfers through the tax system. Although much would
depend upon exactly how these steps are implemented, social democrats
would object to the reinforcement of the commodification of labor and
consumption that these measures involve. Indeed, Ellwood's system ties
social rights of citizenship to participation in the market even more closely
than does the current system. Under Ellwood's proposal, one is eligible
for benefits only as a worker or as a potential worker, and as consumers,
since benefits will be provided principally in cash rather than through
universally provided social services. Finally, although social democrats
can accept strong obligations regarding work, they would be concerned
that the means through which the proposals would be implemented and the
context in which it would function, could serve to reinforce patterns of
labor market segmentation, confining a significant group of workers in
exploitive jobs that they are forced to take, and that they can afford to keep
only because the state in effect provides a significant part of their wages.
Far from creating the conditions for the gradual displacement of capitalism
in a system of economic democracy, Ellwood's proposals, both symboli-
cally and in practice, would serve to stabilize the existing structures of the
Because social democrats and liberals hold different principles and
objectives, they come to support different kinds of policies and to evaluate
proposed changes in different ways. Although the exigencies of power and
political compromise may lead to convergence, we will be able to under-
stand the controversies and processes of compromise better by attending
to the underlying conceptual structures within which conflicting interpre-
tations and judgments are formed.


1. A conceit, 1 must confess, 1 have shared, in my 1988.

2. My use of these terms is similarto Esping-Andersen's (1985a, 1985b, 1990), on whom
1 have drawn liberally, particularly for my account of the social democratic conception of
the welfare state. My account differs from his in attempting to specify the normative visions

of these theories more fully and in the account of the liberal position. While I emphasize the
differences between these two models of the welfare state, I do not wish to deny the important
similarities. Indeed, many critics hold that the development of social policy has been quite
similar in all industrial societies, responding to the imperatives of industrialization and
modernization. One important similarity, which I have not been able to discuss in this chapter,
has to do with the ways in which both liberal and social democratic models are premised on
certain assumptions regarding gender, work, family, and so on, with men as the principal
wage-earners, and women as principally responsible for child care and household work. See
Hemes (1987) and Gordon (1990) for analyses of the welfare state focusing on questions of
gender, an issue that also receives some attention in Esping-Andersen (1990).
3. See Moon forthcoming. There is a vast literature on the foundations of liberalism, and
considerable controversy regarding the kind of position advanced here. My view is similar
to Rawls (1985); for a critique, see Raz (1990).
4. For a discussion of the worth of liberty, see Rawls 1971, pp. 204-5. Holmes (1988)
and Home (1988, 1990) argue convincingly that welfare programs are natural extensions of
traditional liberal values such as security or property.
5. This point is developed in my 1988, from which much of this discussion is drawn.
6. See, for example, Goodin and LeGrand 1987, and the contrasting views of the welfare
state in Barry 1990, and Goodin 1990. For a discussion of the welfare state couched almost
entirely in terms of equality, see Weale 1990.
7. This is an important theme in Wolfe's analysis and critique of state provision. See
Wolfe (1989), especially chapters 4 and 5.
8. This seems to me to be characteristic of much British theorizing on the welfare state,
including Titmuss, Harris, Townsend, and others.
9. That the commitment to equality can sometimes sit uneasily with the commitment to
democracy can be illustrated by Weale's argument for earnings-related welfare state schemes
on the grounds that, by increasing the total volume of government transfers, they lead to
greater "egalitarian effectiveness." Weale explains this egalitarian effectiveness in part as

Of course, there is no necessary incentive to redistribute savings in the public eamings-

related system, but equally there is little practical opportunity to resist any modest
redistribution that managers of the public scheme determine. Denied the "exit" option of
shopping around ... , the typical citizen is confronted merely with the costly "voice"
option of changing the terms of the public scheme. Since people are often highly ignorant
of the details of pension schemes, participation to change their terms is extremely costly.
(1990, p. 481)

In short, because democratic control is difficult, popular opposition to redistribution will be

ineffective, allowing elites to achieve "egalitarian effectiveness:' At least until a Reagan or
Thatcher gets elected.
10. My account of the characteristic structures of social democratic programs follows
II. I focus only on the moral dilemmas. There are obviously difficult technical questions
as to whether the social democratic formula is sustainable on economic grounds. See, for
example, Lundberg 1985. Recently, Esping-Andersen (1990) also questions the viability of
the social-democratic model. See particularly pp. 184-88.
12. See the discussion of this point in Dahl 1982.
13. Hemes (1988) points to an increasing interest in individual rights in Scandinavia in
recent years (p. 201). Robert Lane has explored the (ambivalent) connections between
J. Donald Moon 117

markets and the development of the cognitive and emotional capacities that contribute to
effective citizenship in a series of essays. See, for example, his 1978.
14. See Room et al, 1989, especially pp. 173-74.
15. Targeted or means-tested programs do contribute toward one social democratic
objective, equality, since they concentrate benefits on those most in need and so bring about
a greater degree of equalization than more universal programs can achieve.
16. Esping-Andersen (1990) argues that reliance upon means-testing is characteristic of
liberal programs, but he fails to recognize how problematic this is in terms of liberal theory.
Moreover, his measures of this reliance exaggerate the difference between liberal and
social-democratic regimes by excluding "normal income-tested schemes" as opposed to
"poor relief expenditure," although both make eligibility for benefits conditional upon need.
The distinction here, and its use in classifying particular programs, appears arbitrary. See his
discussion on p. 78.
17. To the extent that the quality of services decline as a result of monopolistic provision,
everyone may be harmed.
18. For a discussion of this difficulty, see my 1988.
19. Some critics of the welfare state disdain any talk of obligation at all, as they view the
welfare state as a set of disciplinary institutions functioning to control and manage its clients.
20. See also Burtless 1987, p. 199.
21. Esping-Andersen (1990) provides a useful account ofthe way in which Sweden, taken
as a model of the social-democratic welfare state, has developed a distinctive "labor-market


Barry, Brian. 1990. "The Welfare State and the Relief of Poverty." Ethics 100:503-29.
Burtless, Gary. 1987. "Taxes, Transfers, and Swedish Labor Supply." Pp. 185-249 in The
Swedish Economy, edited by B. Bosworth and A. Rivlin. Washington, DC: Brookings
Dahl, Robert A. 1982. Dilemmas ofPluralist Democracy. New Haven, CT: Yale University Press.
Ellwood, David. 1988. Poor Support. New York: Basic Books.
Esping-Andersen, G. 1985a. Politics Against Markets. Princeton, NJ: Princeton University
- - - . 1985b. "Power and Distributional Regimes." Politics and Society 14:223-56.
---.1990. The Three Worlds of Welfare Capitalism. Princeton, NJ: Princeton University
Goodin, Robert. 1990. "Stabilizing Expectations." Ethics 100:530-54.
Goodin, Robert and LeGrand, J., eds. 1987. Not Only the Poor. London: Allen & Unwin.
Gordon, Linda, ed. 1990. Women, the State, and Welfare. Madison: University of Wisconsin
Harris, David. 1987. Justifying State Welfare. Oxford: Basil Blackwell.
Hernes, Helga. 1987. Welfare State and Woman Power: Essays in State Feminism. Oslo:
Norwegian University Press.
- - - . 1988. "Scandinavian Citizenship." Acta Sociologica 31: 199-216.
Holmes, Stephen. 1988. "Liberal Guilt." In Responsibility, Rights, and Welfare, edited by J.
Donald Moon. Boulder, CO: Westview.
Horne, Thomas. 1988. "Welfare Rights as Property Rights." pp. 107-32 in Responsibility,
Rights. and Welfare, edited by J. Donald Moon. Boulder, CO: Westview.

- - - . 1990. Property Rights and Poverty. Chapel Hill: University of North Carolina Press.
Katz, Michael. 1989. The Undeserving Poor. New York: Pantheon.
Lane, Robert E. 1978. "Autonomy, Felicity, Futility: The Effects of the Market Economy on
Political Personality." Journal of Politics 40:2-24.
Lundberg, Erik. 1985. ''The Rise and Fall of the Swedish ModeL" Journal of Economic
Literature 23: 1-36.
Marshall, T. H. [1950]1977. "Citizenship and Social Class." In his Class, Citizenship, and
Social Development. Chicago: University of Chicago Press.
Mead, Lawrence. 1986. Beyond Entitlement. New York: Free Press.
Moon, J. Donald. 1988. "The Moral Basis of the Democratic Welfare State." Pp. 27-52 in
Democracy and the Welfare State, edited by A. Gutmann. Princeton, NJ: Princeton
University Press.
- - - . Forthcoming. Thin Selves, Rich Lives, Tragic Conflicts: Moral Pluralism and the
Creation of Political Community. Princeton, NJ: Princeton University Press.
Offe, Claus. 1984. Contradictions of the Welfare State. Cambridge: MIT Press.
- - - . 1988. "Democracy Against the Welfare State?" Pp. 189-228 in Responsibility,
Rights, and Welfare, edited by J. Donald Moon. Boulder, CO: Westview.
Plant, R. 1988. "Needs, Agency, and Welfare Rights." pp. 55-74 in Responsibility, Rights,
and Welfare, edited by J. Donald Moon. Boulder, CO: Westview.
Rawls, John. 1971. A Theory ofJustice. Cambridge, MA: Harvard University Press.
- - - . 1985. "Justice as Fairness: Political Not Metaphysical." Philosophy and Public
Affairs 14:223-51.
Raz, J. 1990. "Facing Diversity." Philosophy and Public Affair 19:3-46.
Room, G., R. Lawson, and F. Laczko. 1989. "'New Prosperity' in the European Community."
Policy and Politics 17: 165-76.
Titmuss, R. 1972. The Gift Relationship. New York: Vintage.
Walzer, M. 1983. Spheres of Justice. New York: Basic Books.
- - - . 1988. "Socializing the Welfare State." pp. 13-26 in Democracy and the Welfare
State, edited by A. Gutmann. Princeton, NJ: Princeton University Press.
Weale, Albert. 1990. "Equality, Social Solidarity, and the Welfare State." Ethics 100:473-88.
Wolfe, Alan. 1989. Whose Keeper? Social Science and Moral Obligation. Berkeley: Univer-
sity of California Press.
"Social Citizenship," Work, and Social Solidarity:
Historical Comparisons
Between Britain and Sweden

Roger Lawson

In this chapter I want to draw upon comparisons between Britain and

Sweden to examine an aspect of the development of "social citizenship"
discussed by T. H. Marshall (1963) in his classic account of citizenship
and social class. It concerns the relationship between the evolution of
social rights and "industrial citizenship"; rights and responsibilities relat-
ing to work, developments in industrial relations, and the labor market. In
Marshall's general theory of citizenship-with its emphasis on the inter-
action between civil, political, and social rights-this relationship played
a minor role; and indeed, as we will see, rested anomalously in his
historical scheme of things. I want to suggest, however, that more attention
should be paid to it if we are to understand important cross-national
variations in the social dimensions of citizenship and inequality or if we
are to appreciate the different ways in which social policy developments
have intersected historically with social relations.
More specifically, the chapter will argue that a prerequisite for the kind
of social democratic strategies pursued in Sweden for much of the period
since the 1930s has been, as Korpi has put it, "that the gap between
people's roles as wage-earners and their roles as citizens is bridged, i.e.
that in the citizen role action is influenced by the interests in the wage-
earner role" (1978, p. 270). In Sweden historically close ties between
industrial relations, labor market developments, and social policy have led
to welfare state programs that have gone further than other countries in
fusing work and welfare. In the process the Swedes have made a distinctive


effort to resolve one of the central dilemmas implicit in "social citizenship,"

that of reconciling communitarian and egalitarian goals with the values and
motivations present within economic activity. By contrast, British develop-
ments have long been characterized by a sharp bifurcation-underestimated
in Marshall's analysis-between the trends in "labor politics" and "social
politics." This has discouraged an overlapping of worker and citizen roles and
has served to circumscribe the realm of social rights and of labor rights. This
in tum has had the effect of restricting how problems of poverty and inequality
have been understood; by, for example, separating "poverty" policy from
broader labor market and economic issues.

T. H. Marshall and Social Citizenship

The best way to approach these issues is by considering briefly Marshall's
theory of citizenship, and particularly his emphasis on its three main
components-civil, political, and social rights. Marshall portrayed these
rights as evolving in Britain over the past 200 years or so through a process
of fission and fusion. Fission was evident when the initial triumphs of
modern capitalism generated institutions favoring the primacy of civil or
legal rights, the rights of individual liberty, property ownership, and
formal equality before the law. For much of the 19th century, the core of
citizenship, including the property-based franchise, was composed of
these civil rights that, as Marshall put it, "were in origin intensely individ-
ual, and ... harmonized with the individualistic phase of capitalism"
(1963, p. 97). These were rights that served initially to consolidate the
sharp class inequalities of liberal capitalism, partly by detaching even the
minimal social rights of the Poor Law from the status of citizenship.
Marshall was chiefly concerned, however, with the processes of fusion
by which political and social rights came to be incorporated into the status
of citizenship. As Turner (1986) has shown, Marshall's view of these
processes was predicated on assumptions about the importance of struggle
and conflicts of interest between different classes and collectivities. Thus
he argued that political rights were gradually extended down the social
scale in the late 19th and early 20th centuries both as a response to the
emerging organizational strength of the working classes and a means of
drawing the teeth from the conflicts inherent in capitalist development.
But political rights, unlike civil rights, were full of potential danger to the
capitalist system. They led to a transfer of disputes over distribution and
status from civil society and the market to politics, by channeling the
aspirations of the organized working class into a drive for social or welfare
rights. To Marshall, in other words, working class political mobilization
and parliamentary power were the crucial forces behind the emergence of
Roger Lawson 121

the modern welfare state, though this did not of course preclude other
important equalizing influences.
Marshall developed his theory of citizenship at the end of the 1940s, only
a few months after the enactment, under the Labour Government, of much of
the legislation associated with the wartime Beveridge Report. The thesis
expressed the confidence at the time about the new sense of national solidarity
and will to achieve a more just and equal society embodied in these reforms.
The aim of the new social rights, Marshall claimed, is no longer merely an
attempt "to raise the floor-level in the basement of the social edifice, leaving
the superstructure as it was. It has begun to remodel the whole building, and
might even end by converting the skyscraper into a bungalow" (1963, p. 100).
Nevertheless, Marshall's analysis also represented the postwar settlement as
a sort of truce between citizenship and social class. He by no means ruled out
the resurgence of class and other social tensions that might challenge, or
destabilize, "social citizenship." On the contrary, his writings were soon to
contain ominous warnings of the fragile character of the welfare state. Writing
in the late 1950s, he was highly critical of the new atmosphere of self-enrich-
ment and competitive consumption that encouraged the middle classes and
more prosperous workers to think of the welfare state as conceived in the
1940s as "a back number" (1963, p. 287).
This is an important point that I want to develop in this chapter. In
commentaries on the more recent "crises" facing Britain's welfare state,
there is a tendency to neglect the threats to its stability evident well before
the economic recession and rise of the new right in the 1970s, and not
merely among the middle classes and rising white-collar strata. Indeed, as
the comparisons with Sweden suggest, more critical in the long run was
the failure in Britain to consolidate the wartime spirit of universalism and
solidarity among the mass of the work force, but especially within organ-
ized labor. However, Marshall's general thesis provides us with too little
insight into why this was so, and mainly because it rested on assumptions
about the influence of the British labor movement on social rights that
need to be reappraised. But this also raises broader questions about
Marshall's historical schema, particularly about the way he handled the
impact both on citizenship rights and class relations of trade unionism and
related developments in the "civil sphere of citizenship."
Marshall acknowledged that these issues complicated his thesis when
he introduced the additional category of industrial citizenship-the right
to form trade unions, to strike, and so on. His treatment of these rights
points, however, to weaknesses in his historical analysis. He claimed, for
example, that they developed initially as an extension of civil rights,
although they were clearly quite different from the legal rights of individual
freedom that he saw as buttressing liberal capitalism (cf. Giddens 1982). But,

more significantly, his account implied that trade unionism and develop-
ments in work and industry generally were essentially of secondary im-
portance in the interaction between citizenship and class. Trade unionism
created a "secondary system of industrial citizenship, which naturally
became imbued with the spirit appropriate to an institution of citizenship"
(1963, p. 116). Before the full development of political rights, this was
important in establishing the claims of workers that they, as citizens, were
entitled to certain social rights. In the 20th century, however, the exercise
of political power had become, in effect, "the normal method" of estab-
lishing social rights, and the role of trade unions in shaping citizenship
was changing. "In the past," wrote Marshall, "trade unionism had to assert
social rights by attacks delivered from outside the system in which power
resided. Today (i.e. the 1940s) it defends them from inside, in co-operation
with government" (1963, p. 117). But he also injected a note of skepticism
when he added that the traditions built up when trade unions were fighting
for their existence still made it difficult for them to accept "a lively sense
of responsibility towards the welfare of the community."
I want to argue that the relationship between trade unionism and class
politics, democracy, and citizenship has been more complex than this
suggests; and, moreover, a key factor in the social construction of ideas of
"social citizenship." To illustrate this, the chapter makes comparisons
between Britain and Sweden in the late 19th and early 20th centuries, the
critical period in Marshall's analysis when the franchise was being opened
up to members of the working class. But it includes also a review of the
contrasting conceptions of social rights in the two countries since then,
and of their effects in terms of broader welfare state strategies and social


The Development of "Industrial" and "Social" Rights

In Britain recent historical research has stressed the ambiguities in
workers' attitudes to state welfare between the 1880s and 1920s when
political rights were extended step by step to working class men and
women. Indeed, Pelling has claimed that

the extension of the power of the state at the beginning of this century, which is
generallyregardedas havinglaid the foundations of the welfare state, was by no
meanswelcomed by members of the workingclass, was indeed undertakenover
the critical hostility of many of them, perhaps most of them. (1968, p. 2).
Roger Lawson 123

Other studies have qualified this picture by suggesting that there was popular
support for certain collective solutions to social problems, such as old age
pensions and housing reforms. However, as Thane has concluded:

There can be little serious doubt that most trade unionists would have preferred
adequate wagesand full employment, advancedthrough voluntary organization,
with the resulting sense of independenceand equality with other classes ... to
dependenceon publicly funded benefits. These were seen as undesirable for the
sense of dependence which they created; and, often also, as employer-inspired
alternatives to ... further development of an independent assertive working
class movement. (1985, p. 185)

Whiteside, too, has observed how, in the labor movement, any consensus
on social reforms "collapsed abruptly in areas where industrial bargaining
and potential state welfare overlapped" (1987, p. 213).
Such tensions between industrial and social rights point to certain
peculiarities of British capitalism in the 19th and early 20th centuries. As
Price (1986) has shown, Britain differed from other European countries in
the degree of autonomy over day-to-day affairs that the more organized
sections of the working class were able to prize from the liberal state,
industrialists, and the middle classes. This was not, however, as Marshall
appeared to suggest, a consequence mainly of the primacy of civil rights
in this period. Rather, its origins were in more informal processes of
negotiation and of containing class conflicts, which characterized Britain's
industrial revolution but which became especially important following the
failure of Chartism in the 1840s. Price shows, for example, how much
control over the supply and organization of labor was quite deliberately
evacuated by management in many mid-Victorian British firms. And, as
Perkin (1989, p. 112) observes, "the tradition amongst craftsmen and most
factory workers was that they rather than the employer controlled the pace of
production and, to a surprising extent, the way in which the work was done."
This pattern of negotiated space and autonomy for labor in British
society was critical in shaping class consciousness when a mass labor
movement emerged between the 1870s and the First World War. At one
stage, in the 1880s and early 1890s, there were possibilities that an alliance
between socialists and "new unionists" could radicalize the labor move-
ment. However, the main trends of this period "represented not a victory
for socialists, but the effective containment of the socialist impulse within
older traditions of trade union politics and class collaboration" (Hinton
1982, p. 21). Thus, despite labor's growing political presence, especially
after the formation of the Labour Party in 1900, the prime strength of
British workers was their ability to secure gains in trade union rights and

to consolidate their workplace power and nonstatutory custom and practice.

This, at least, was true of the broadening stratum of moderately well off
workers, including the rapidly growing semiskilled workers among whom
trade unions made their greatest gains. With the decline of new unionism in
the 1890s, the mass of the urban poor remained impermeable to trade union-
ism. The effects as far as the drive for "citizenship" rights was concerned were
profound. Preoccupied with the role of defending and advancing where
possible in its bargaining with employers, much of British organized labor
viewed these rights primarily in terms of union immunities and freedom from
constraint by government and the judiciary, and pressed only secondarily for
state welfare, and then in specific areas that were mostly consistent with the
autonomy of labor. Even the extension of political rights was not a central
issue of working class politics, despite the fact that before the First World
War more than two thirds of the working class, male and female, were
excluded from the franchise (Hinton 1982, p. 37).
The bifurcation here between the concerns with "industrial" and "social"
rights was also influenced by the reactions of government and industry.
British governments were generally restrained in their approach to unions and
willing to encourage the voluntarism or "collective laisser-faire" in the labor
market that kept the unions away from politics. British employers, too, usually
proved less intransigent in the face of demands for collective bargaining and
related labor rights than Continental employers. This was partly because
many employers welcomed trade union discipline and found formal collective
bargaining more attractive than irregular disruptions of work. But the over-
riding concern of most British firms at the time was also to maintain autonomy
and freedom from legal constraint and state interference. They, too, had a
deep-rooted distrust of the political stage, preferring to confine their links
with government to lobbying in private. This was different from the situation
developing in Sweden and many other European countries, where social and
economic affairs were more closely regulated by powerful state bureaucra-
cies, and by a growing interarticulation between the state and collective
organizations representing industry and capital.
Similar processes were evident in the growth during the 19th century of
self-governing working-class benefit funds and self-improvement associ-
ations, run by friendly societies, trade unions, cooperatives, and other
fraternal clubs. These were again more wide-ranging and influential in
shaping working-class culture and attitudes to state welfare than such
institutions elsewhere in Europe. Here, the demarcation of an autonomous
sphere of influence for labor involved initially a convergence of themes
that were common to working-class and middle-class ideology and culture.
As Price (1986, p. 64) observes, both were suspicious of government and
bureaucratic power, defined liberty against the state, and expressed a
Roger Lawson 125

deeply ingrained belief in self-improvement. For the trade unions these

benefits served also to reinforce the process of industrial bargaining, by
providing union branches with a means to police trade agreements and
discipline members who transgressed them. It should be added, too, that,
like British trade unionism generally, the solidarities on which these
actions were based emphasized sectional, localized, and male, interests
among workers. And, by distancing the broad mass of "respectable"
workers from "the grey, faceless, lower third of the working class" (Gil-
bert 1966), they posed intractable problems for the Labour Party when it
sought to formulate policies on state welfare and poverty.
The effects of these divisions of interest within the working class were
clearly evident in the social reforms of the Liberal Government in the decade
before the First World War. The reforms, which instituted old age pensions,
health and unemployment insurance, and other labor market measures, are
commonly regarded as a major advance away from Victorian perceptions of
government toward universalist notions of social citizenship. They undoubt-
edly reflected a new responsiveness within governing circles to the "needs"
of the citizen-voter; and to concerns with Britain's "national efficiency" in
the face of increasing economic competition and the growing military threat
in Europe. Potentially, too, the administrative machinery established by the
reforms was capable of bridging the gap between industrial and social rights.
For example, the measures to deal with unemployment made possible, as
Beatrice Webb noted, "the increased control of the employer and the wage-
earner by the state" (quoted in Harris 1972, p. 365). Nevertheless, when
compared with the launching of welfare states elsewhere in Europe, what
most characterized the Liberal reforms was the way these apparently new
perceptions of welfare were overpowered by a more traditional concern
among Britain's elites. A primary political consideration of the reforms was,
as Hennock (1981, p. 85) puts it, "the need to reassure the citizen-voter that
he would not be allowed, for reasons over which he had no control, to fall
into the non-citizen class of paupers."
Behind this lay another distinctive feature of Britain's early social
policy developments: the absorption of the energies of many social reform-
ers in the politics of poverty and poor relief, in contrast to the Continental
emphasis on the politics of class (the "worker's question"). As has been
implied, this difference can be attributed to the way in which many issues
surrounding the worker's question in Britain had been dealt with in the
civil sphere of society. Compared with their Continental counterparts,
Britain's policy-making elites and the intellectuals closely associated with
government at the time of the Liberal reforms were thus more insulated
from class struggles and the tensions of industrial affairs. Under less
organized or sustained pressure from outside to rethink British concepts

of the state and society, their attitudes toward social needs remained
strongly influenced by traditional elite preoccupations with the poor
(Ashford 1986). The unique way in which Britain's Poor Law gave na-
tional exposure to problems of poor relief also played an important role
here. The establishment in the 19th century of a central Poor Law Com-
mission, within the framework of the unitary British polity, meant "that
tensions generated by the workings of the local workhouses, asylums and
sick wards, and by the implementation of other policies to deal with the
poor, inevitably and recurrently generated pressures for national debates,
investigations and policy changes" (Orloff and Skocpol 1984). Underlying
the Liberal measures there was also the paradox that, although much of
the impetus behind the reforms came from revelations of the injustices and
inadequacies of the Poor Law, many of the new services nevertheless "had
their character moulded by the moral assumptions of the nineteenth cen-
tury" (Titmuss 1963). Poor law ideas about need and behavior were carried
into the new legislation.
Despite their potential for more far-reaching change, the Liberal re-
forms thus did little to bridge the gap between "industrial" and "social"
citizenship. On the contrary, they firmly implanted in British politics a
sharper differentiation between "welfare" rights and rights relating to
labor and the productive process than existed in other European countries.
Broadly speaking (and there were exceptions), the former were aimed
essentially at protecting the weak from want, whereas the latter appealed
more to the aspirations of the active and economically stronger sections
of the work force. This division was further reinforced during and after
the First World War by a disenchantment in industry with the wartime
controls over the labor market and by the rising power of labor in the
workplace. "Home Rule for Industry" was the slogan of unions and many
employers after the Armistice. The effects of this were to be readily
apparent in the way Britain responded to the Great Depression of the
1930s. In striking contrast to Sweden, Britain's policy heritage effectively
stifled innovative ideas, particularly "progress toward combining 'social'
and 'economic' interventions in the form of deficit-financed public works"
(Weir and SkocpoI1985). British politics were locked instead into intense
struggles over a massive public assistance program for the unemployed
that raised again the specter of the poor law.
These battles played an important role in shaping the Labour Party's
attitudes toward state welfare, and in turn in setting the scene for the devel-
opments of the 1940s. There were, however, also signs of changes within the
labor movement during the 1930s that might have altered more significantly
the direction of British social policy. Unemployment and the harsher climate
in industrial relations following the General Strike in 1926 led to a marked
Roger Lawson 127

diminution of union bargaining power and the scope for collective action.
Union leaders looked increasingly to reforms in state welfare to safeguard
their members' interests. The central Trades Union Congress (TUC) ad-
vocated a number of measures aimed at forging closer links between work
and welfare, including earnings-related social security schemes and work
promotion programs. However, unlike the outwardly similar trends in
Sweden at the time, these developments stemmed from union and Labour
Party weakness not strength; and had little political impact in the 1930s.

The Separation of the Worker and Citizen Roles After 1945

Events during and immediately following World War II seemed also at
the time to promise a new fusion of the trends of industrial and social
citizenship. As Marshall's analysis indicated, this appeared to be the
implication of the vastly expanded role that trade unions began to play in
formulating national economic and social policies during the 1940s, but
especially after Labour's electoral victory in 1945. As part of this new
role, the unions entered forcefully into the debates on the welfare state
stimulated by the publication of the Beveridge report in 1942; or at least
into the debates surrounding Beveridge's advocacy of full employment
and Keynesian demand management as essential preconditions for ad-
vances in welfare. On the details of welfare reform they made little
specific contribution, despite the plans they had drawn up during the
1930s. More generally, the sense that a distinct shift in orientation had
occurred within the union movement stemmed from the apparently en-
hanced authority of the TUC; and its activities in promoting wide-ranging
reconstruction plans designed to ensure that the economy would function
within the imperatives of "public control," with full employment and an
equitable distribution of income its essential goals. Moreover, by bringing
key industries into public ownership and extending state regulatory power
over the private sector, the Labour Government clearly offered the unions
the possibility of enlarging their "corporatist" role in public policy.
Despite this, however, the most striking feature of Britain's postwar settle-
ment as far as the unions were concerned was the extraordinary tenacity of
older attitudes toward their autonomy in the labor market and toward the
division between industrial and social rights. In terms of their own organiza-
tion, the TUC and the union movement not only refused to tolerate state
interference into the traditionally voluntarist character of industrial relations.
They sought also to use the immediate postwar period to institutionalize more
firmly their freedom of action in the labor market in ways that gave them
"greater leverage over the Labour Party ... and substantial new power
to resist any Conservative turn towards the labor-repressive economic

strategies which, in the unions' eyes had caused so much misery in the
1930s" (Bomstein and Gourevitch 1984, p. 18). The unions went to
considerable lengths, too, to avoid compromising their freedom of action,
including (unlike most other European unions) withdrawing from any
direct role in the provision or administration of the postwar social services.
As this implies, this union stance meant in practice for most unionists
an even sharper focus than in the prewar period on their role in advancing
workers' interests in wage bargaining and workplace rights. At the same
time, the realm of the new social rights was circumscribed in ways that
distinguished the British labor movement's contribution to the construc-
tion of the welfare state from that of the Swedish Social Democrats and
unions. A feature of Swedish policies, as mentioned earlier, has been a
bridging of the gap between people's roles as workers or wage earners and
citizens. In Britain, by contrast, such an overlapping of these roles was for
the most part discouraged by the social policy reforms of the 1940s, despite
the new commitment to full employment; and partly because of the con-
straints imposed by the unions' traditionalist approach to industrial rights.
There were, however, other important factors here. As Marshall's (esp.
1963, chap. XX) writings showed, the painfully acquired sense of national
community and solidarity in Britain during the war years encouraged a
view of social rights as promoting welfare for "the people" in ways that
were largely independent of, or tied only loosely to, people's concerns as
workers. This notion of social policy as representing a "higher authority,"
the interests and status of the citizen, became firmly entrenched in the
Labour Party's attitudes to welfare.
In the unique circumstances of the 1940s such ideas provided a powerful
stimulus for collectivist advances in a number of social services, most
notably in health care, and this seemed at the time to promise a more
far-reaching equalization of social status and resources. But there were
also contradictions and tensions underlying the British version of social
citizenship working against this, and that had much to do with the separa-
tion of the worker and citizen roles. Some of these have been highlighted
by Ashford in an interesting comparative analysis of the development of
welfare states. To Ashford, "one of the remarkable characteristics of the
British welfare state [was] how readily government intervened in all kinds
of social policies, but how reticent it was to suggest that wage and income
policies were politically and economically linked to social policy" (1986,
p. 18). The consequences were evident during the rapid welfare state
expansion between the 1950s and 1970s, when Britain was one of Europe's
least successful performers in articulating the interdependence of wages,
welfare, and productivity. Closely related to this, the key area of voca-
tional training and other matters relating to the promotion and recognition
Roger Lawson 129

of skills in the mass of the work force were left largely in postwar Britain,
as previously, to negotiations between unions and managers; and then to
highly decentralized bargaining (Bornstein and Gourevitch 1984, p. 26).
The separation of the worker and citizen roles contributed also to the
undermining of Britain's commitment to full employment well before the
recession of the 1970s. The abandonment of full employment as a primary
goal of public policy occurred in effect in 196611967, when Wilson's
Labour Government vainly attempted to maintain the international posi-
tion of the pound in the face of mounting pressures on the economy. The
pressures reflected British industry's difficulties in maintaining its com-
petitiveness, but behind them lay also the paradoxical and unique combi-
nation of the postwar settlement: full employment and expanding commit-
ments to welfare together with largely unrestrained, unregulated, and, by
the 1960s, increasingly unsettled industrial relations. Significantly, too,
any idea of developing a public labor market policy in cooperation with
the unions and industry to institutionalize the goal of full employment was
absent from the considerations of postwar Labour and Conservative Gov-
ernments. However one interprets this, the effects of the changes in the
1960s were quickly felt in a growing dualism in social policy. As unem-
ployment rose, trade unions successfully campaigned for improved redun-
dancy benefits. But among the poor and unorganized a growing hard core
of the unemployed came increasingly to depend on a last-resort public
assistance scheme still bearing many of the hallmarks of poor relief.
Moreover, as an Organization for Economic Cooperation and Development
(OECD) report observed, in Britain this section of the unemployed was still
"likely to be labelled comprehensively and pejoratively as 'deviant' ... and
subjects as much for social control as for social welfare" (Sinfield 1968).
There was evidence, the report noted, of trade union leaders decrying the
longer term unemployed as workshy or as types who "give the respectable
working man a bad name."
These attitudes reflected in turn other ways in which Britain's postwar
policies had departed from the egalitarian and solidaristic spirit of the
1940s. Again, however, they could be related to contradictions in the
strategy of social citizenship of the 1940s. In social security, for example,
the separation of the worker and citizen roles led in practice to a concep-
tion of welfare more directly related to the traditional act of charity and
to the older concerns with mitigating poverty than to the requirements of
the broad mass of the work force. Thus, the Beveridge-Labour reforms,
although introducing new principles of universal coverage and equal
treatment of citizens, were concerned primarily with ensuring that no one
should fall below a minimum subsistence standard. By tying social secu-
rity to such a basic conception of citizen's rights, the policies soon fostered

a new social divisiveness in welfare, with the middle classes and more
affluent workers relying increasingly on occupational schemes or private
insurance and regarding state benefits more as an expression of benevo-
lence toward the weak (Lawson 1987). Such developments were strongly
encouraged by postwar Conservative governments and by the 1960s had
created a formidable alternative to the welfare state: company-based
pensions and a range of other benefits acting as "concealed multipliers of
occupational success" (Titmuss 1963).
These divisions are crucially important in explaining the failure, after
the promise of the Beveridge reforms, to develop a coherent antipoverty
strategy and also the powerful survival in postwar Britain of older theories
attributing poverty to individual or group deficiencies. The growth of
occupational and private welfare, which has increasingly embraced blue-
collar workers, has served to undercut popular loyalties to key areas of
state welfare while drawing valuable resources away from the state sector
and enhancing resistance to improving benefits. The results have been
most evident in social security, where universal benefits surviving from
the 1940s, but still offering only minimal protection, have been supple-
mented to an ever-increasing extent by special means-tested programs for
the poor. Whereas in 1948 around 1.3 million persons were living in
households dependent on the basic national assistance scheme, by the early
1970s the number had already reached more than 4 million; and by the
mid-1980s it had doubled again to 8 million (Lawson 1987). Although this
has produced some policies that are benevolent in intent, it has also, as in
the United States, separated poverty policy from broader labor market and
economic issues and restricted the way in which the problem of "poverty"
has been understood. Thus, for example, in a survey of public perceptions
of poverty conducted in the mid-1970s twice as many Britons attributed
poverty to laziness or lack of willpower as did the Germans, French, or
Italians; and almost 4 times as many as the Danes and Dutch (European
Communities 1977).
The full implications of this became apparent when the economic crises
and dramatic rise in unemployment during the 1970s and early 1980s gave
a further boost to individual theories of poverty and fears about the
widespread abuse of benefits. More generally, the paradox of the British
version of social citizenship is that it became associated well before the
Thatcher years with policies that undermined the social solidarity it was
meant to foster by in effect splitting the working class and generating new
coalitions between the better-off workers and the middle class. By the
1980s this had created a large constituency, not only for antiwelfare and
tax backlash, but significantly also for a sustained attack on the power and
traditional autonomy of trade unions.
Roger Lawson 131


In turning to Sweden, I should perhaps first acknowledge a difference

in emphasis in my interpretation of the distinctive features of the Swedish
model of the welfare state from that of a number of other commentaries.
Discussions of Swedish social policy have tended to stress the strength of
entitlements to welfare based on citizens' rights that have released indi-
viduals from the force of work. Swedish policies, it is argued, have
involved a more sustained effort than policies in other Western democra-
cies to decommodify wage earners by reducing their reliance on their labor
power. I would certainly accept that such processes constitute an important
aspect of modern welfare state developments; and that their impact in
Sweden has been such that people do not bear the anxieties associated with
the "sale" of their labor power that have recently become more common-
place in Britain and certain other Western societies. My concern, however,
is that an understanding of these processes, and more generally of Sweden's
success, until recently, in pursuing egalitarian social citizenship objec-
tives, requires us also to consider a parallel set of developments through
which "wage labor has gradually strengthened its positions-as mode of
production, instrument of redistribution, social order and ideology"
(Christensen 1984). It is the distinctive work-welfare relationship result-
ing from this that is examined in this section.

Historical Background
How then does the kind of analysis applied to Britain help us to
understand these developments? Although the answers are to be found
mainly in the period of Social Democratic supremacy since the 1930s, we
need also to take account of earlier trends associated with the development
of citizenship rights, particularly during Sweden's rapid but belated indus-
trialization and modernization between the 1880s and 1920s.
By European standards, the country's transformation in this period was
relatively smooth. But it was nevertheless characterized by a sharp con-
frontation between a conservative state and religious establishment and
radical-democratic forces; and especially by struggles, only gradually
resolved in the first two decades of this century, over political rights and
the introduction of responsible parliamentary democracy. This was crucial
in shaping the traditions of Sweden's labor movement. In contrast to the
early pattern of developments in Britain, it led from the beginning to a
close and interdependent relationship between the movement's party and
union branches, in which the growth of unions was intertwined with
politics and with broader demands for social rights (Korpi 1978; Martin

1984). When the Social Democratic Party (SAP) was founded in 1889, at a
time when property and income qualifications restricted the franchise to only
about 5% of the population, it was to view unionization as the key to the
political mobilization of the working class. Party leaders stressed "the neces-
sity for trade unions to transcend immediate interest articulation regarding
wages and working conditions in order to participate in the general struggle
for equality and democracy" (Olsson 1990, p. 81). The unions were thus to
throw their weight behind the campaign for universal suffrage, by for example
mounting a 3-day general strike against suffrage restriction in 1902.
The importance attached to broader political and social goals by the unions
was reinforced by other developments. Although most Swedish employers
were willing to recognize workers' rights to organize and engage in collective
bargaining, their approach to labor's challenge was initially distinctly con-
frontational, and with some notable success. Unions were defeated in a series
of industrial conflicts culminating in a disastrous general strike in 1909, when
they temporarily lost half their members. Moreover, Swedish employers
differed from their British counterparts both in firmly insisting on managerial
prerogatives in most local workplace affairs and, perhaps more significantly,
in founding in 1902 a strong centralized confederation aimed at unifying
action against labor. The latter points also to a more general contrast with
Britain that shaped the reactions of Swedish labor: the high degree of concen-
tration and interlocking control in Swedish industry that encouraged an early
drive toward centralization of authority.
These were again important factors behind the premium placed within the
labor movement on a high level of integration, unity, and solidarity, and they
served to underline the unions' dependence on politics. In fact, in the years
surrounding World War I, which marked the democratic breakthrough in
Sweden, "the party's [SAP] position in the state arena, modest as it still was,
proved crucial in protecting the unions against the employers' efforts to use
the state to reinforce their power in the market arena" (Martin 1984, p. 197).
Significantly, too, this also provided a stimulus to Swedish trade unions to
rethink the relationship between "industrial" and "social" rights. Compared
with Britain, there was already considerably more overlap between the two
in the movement's early debates about the "workers' question." This was
taken further in the 1920s, when the central union confederation (LO) indi-
cated its preference for a national strategy of "wage solidarity" aimed at
diminishing wage differentials over conventional competitive wage bargain-
ing (Ashford 1986, p. 202). This was influenced by concerns at the effects of
unemployment in depressing wages and undercutting union strength. How-
ever, it marked also the first step toward the kind of agreement on wage policy
and wage earner solidarity that was soon seen to be crucial in achieving
the broader social objectives of social democracy.
Roger Lawson 133

There were other influences, too, on early trade union strategies that
stressed the overlapping character of the worker-citizen roles. The initial
upsurge of union activity, and of the labor movement more generally, was
part of a more wide-ranging and distinctively Swedish phenomenon of
"popular movements" (folkrorelser) that developed very rapidly in the
1880s and 1890s as a means of mobilizing the politically disenfranchised
and articulating popular resentment with the traditional social order. These
mass popular organizations embraced-often with overlapping member-
ships-religious dissenters and free-thinkers, a powerful temperance move-
ment, and consumer cooperatives, as well as trade unions. For the labor
movement, they were to play an important role in linking what were
perceived as working-class aspirations with more popular concerns, in-
cluding the concerns of Sweden's large class of peasant farmers and
smallholders. Moreover, both because of the early strength and resources
of the labor movement and because of other divisions within the folkrorelser,
Social Democracy became in the 20th century "the main carrier of the
popular movement tradition" (Therborn 1989, pp. 199-201).
This tradition was in turn to mold labor's attitudes toward social issues
and rights, and in two respects in particular. It meant, on the one hand, that
Swedish social democracy avoided the ghettoization on social issues
evident among Continental socialists. Instead, the Swedes were to develop
by the 1920s and early 1930s a clearer, if pragmatic, conception of
universalist "popular" welfare drawing upon the alliances associated with
the popular movements and appealing also to the rural population. On the
other hand, however, the disciplined worldview of the popular movements
meant that "from the start, Swedish Social Democracy had a restrictive if
not prohibitionist view on ... moral issues close to social policy" (Olsson
1990, pp. 75-76). As well as drinking, these moral issues included an
unusually deep distrust in laborite ideology of the workshy. Significantly,
however, the early associational world, which gave expression to these
values, was also to link a strong work ethic with activities accustoming
Swedes to the pursuit of cooperative and collectivist goals.
As Olsson's study of Sweden's early social policy discourse shows,
such values and concerns were found also among the bureaucracy and
liberal "welfare intelligentsia." Moreover, in sharp contrast to the British
liberals, Sweden's brief era of political liberalism in the first two decades
of this century was characterized in social policy by a playing down of
social insurance as a policy alternative to traditional relief, especially for
the able-bodied. The policy development with the most profound long-
term consequences was the establishment of a national Unemployment
Commission with wide-ranging powers to promote "workfare"; or what
was called the "work-line" as distinguished from the "cash-line" (Therb-

orn 1989, p. 218). First set up in 1914, the Commission soon enjoyed
considerable autonomy in developing an increasingly centralized system
of public relief works, special employment programs and training projects.
Cash relief was not excluded but the authorities had wide discretion to
deny this, and indeed to coax many of the younger unemployed into
nationwide projects building roads and power stations and in forestry and
land improvement. Understandably such measures were initially highly
controversial and during the 1920s two SAP minority governments fell
over disputes with the Commission. However, the disputes were mainly
over the Commission's insistence, backed by liberals and conservatives,
on below-market wage rates on relief projects. On the work-line itself a
broader consensus emerged during the 1920s: Hence at a time when British
labor was obsessed with the "dole," the energies of Swedish movement
were mostly devoted to plans for a more "civilized" model of workfare
(Weir and Skocpol 1985).
This raises also a more general historical point about Sweden's state
structures and political processes. Whereas the drive for social rights took
place initially against a background of confrontation with the state estab-
lishment, by the 1920s and 1930s Social Democratic strategies were
increasingly influenced by other processes deeply rooted in Swedish state
traditions that encouraged significant compromises between different po-
litical and socioeconomic interests. They included a tradition of investi-
gatory commissions and other public forms of deliberation and consulta-
tion, through which state officials had taken the initiative in drawing
"disparate groups into active and continuous participation in the policy
making system and did so early, before the participants became too firmly
locked into opposing positions" (Heidenheimer, Heclo, and Teich Adams
1982). This was to be important in several ways in the launching of the
"Swedish model" following the election of the first majority SAP govern-
ment in 1932. Prior to this, party and union leaders had already been
absorbed into the consultative processes; as, too, had been a new genera-
tion of economists and social policy experts that had begun to emerge
within the labor movement. But, more importantly, this helps to explain
"the unusual combination of political strength and reformist flexibility
exhibited by the Social Democrats" (Weir and Skocpol 1985) after 1932,
which led them into "historic compromises," first of all with the Agrarian
Party and then with industry in the celebrated Saltsjobaden accord of 1938.

The Swedish Model and Social Citizenship

These processes led to the formulation of a distinctive social democratic
variant of "social citizenship," embodying a different perception of the
Roger Lawson 135

expansion of the welfare state and of the links between work, welfare, and
social solidarity, to that found in the British labor movement. At a broad
ideological level, the Swedish strategy involved the notion of an historic
reform sequence beginning with political democracy and then leading through
the creation of a welfare state to greater economic democracy and to a more
fundamental social transformation. Thus social democratic theorists of the
1930s argued that Sweden was entering a second reformist "social citizen-
ship" phase in which the emphasis would be on raising the capacities and
resources of citizens and creating the broader solidarity on which more
far-reaching change could be based. As Wigforss put it, "welfare was to be
the immediate result. But behind it were the greater possibilities for workers
themselves to take over the production" (quoted in Korpi 1983).
More specifically, the priorities associated with the creation of the
Swedish welfare state differed in certain key respects from those in Britain
after World War II. Under the Swedish model far more than in Britain,
business interests were allowed favorable conditions for continuous ex-
pansion and investment, and indeed concentration, partly because, to
quote Wigforss again, "we had to try to hasten the development towards
big enterprises and thereby the need for economic planning which we had
assumed in our programme." SAP and union leaders also placed more
stress than their British counterparts on "making the pie larger in order
that there would be more to divide," and at the same time on laying a stable
basis for continuing Social Democratic governance by establishing firm
links in people's minds between greater social equality and social security
and rising productivity and economic efficiency. This entailed in turn a
revamping of policies promoting the "work-line" and restraining less
popular forms of "passive" or unproductive welfare spending. But it also
meant that, at the more general strategic level, "government welfare policy
was designed to be in harmony with labor-market policy, union wage-bar-
gaining practices, and government macroeconomic management. A defin-
ing characteristic ... [was], therefore, the absence of clear boundaries
between social policy and other policy domains" (Esping-Andersen &
Korpi 1984, p. 186).
Equally important to the Swedish model, however, were the links
between class politics and social rights. In contrast to the British attach-
ment to the moral appeal of "universal" social rights largely disassociated
from class relations, the SAP gave priority to measures aimed at raising
an awareness of "class" identities and interests among Swedish workers,
and hence mobilizing what was later described as a common "wage earners
front" committed to egalitarian policies. This again meant the forging of
close links between social rights and obligations and work, wages, and
training programs.

These considerations were clearly evident in the early policies pursued

by the SAP in the 1930s. Although these included improvements in
pensions and family support, it was the responses to the most pressing
issues of unemployment and economic recovery that gave the new Swe-
dish policies their distinctive character. At this stage, the old Unemploy-
ment Commission was left in operation, but wage rates on its schemes were
raised to prevailing market levels, the program of public works and
employment promotion was greatly extended, and the Commission was
increasingly staffed by officials sympathetic to the SAP. The trade unions
were also given primary responsibility for administering a new national
unemployment insurance scheme introduced in 1934. However, where
cash benefits were concerned, the scheme was designed to play only a
limited, residual role in social policy; and union administration was more
notable for efforts to encourage, and at times coerce, the unemployed into
joining relief works and training programs. Significantly, too, as the
economy revived in the 1930s, priority was given to the Saltsjobaden
accord on wage policy and industrial relations rather than to immediate
advances in welfare. Enormous political pressure was exerted on the union
and employer confederations to construct this accord, which inaugurated
what came to be known as "solidaristic wage policy," centralized collec-
tive bargaining aimed at gradually diminishing wage disparities and achiev-
ing "equal pay for equal work." With the agreement Sweden thus "inverted
the normal sequence of building social policies by first considering how
wage policies provide a foundation for social policy" (Ashford 1986, p. 18).
Swedish policies after 1945 were to develop these early initiatives in a
number of ways, but most strikingly by molding social security and other
welfare programs around a particular brand of "social Keynesianism." As
elaborated by SAP and union strategists in the 1950s and 1960s this had
the effect of firmly institutionalizing full employment and solidaristic
wage bargaining as the core components of social rights or the "first-line"
measures for reducing class and gender inequalities. At the same time, the
elaborate network of "active" labor market measures, which lay at the
heart of this strategy, was designed to raise the skill levels of Swedish
workers, steer them into expanding sectors of the economy, and to provide
Swedish industry with considerable latitude to invest and innovate. The
ways in which this strategy sought to confront the economic dilemmas of
full employment capitalism have often been commented on. Less attention
has been paid, however, to the sociological significance of these policies.
As well as building upon the earlier emphasis on the benefits of the "work
principle," Swedish social democrats appreciated also the potential of an
active labor market policy for the class awareness and action that they saw
as essential to achieving their goals. Thus the stress placed on labor
Roger Lawson 137

mobility, both geographical and occupational, can be explained only

partly in economic terms. It was also seen explicitly as a means of
"breaking down often competing solidarities of a communal, industrial or
craft kind ... [and] encouraging the development of a specifically class
identity among Swedish workers" (Goldthorpe 1990, p. 438).
These aspects of Swedish social policy have had a coercive and some-
times brutal side. Strict work enforcement procedures have been applied
by the Labor Market Board and unions and by Sweden's social assistance
authorities to minimize welfare dependence, but they have also had the
effect of uprooting families and of imposing powerful sanctions and
behavioral controls favoring wage labor and social conformity (Christensen
1980, 1984). Against this, however, it has been argued that the efforts to
contain "passive" welfare for the working-age population have played a role
in fostering consensus and support for the general expansion of social ser-
vices, and especially for the relatively generous benefits for pensioners and
the handicapped. There has been less evidence in Sweden than in many other
countries of public concern with welfare abuse, and this has been attributed
as at least partly due to "control measures which make it difficult for people
to obtain compensation for any length of time if they are genuinely available
for work" (Rehnberg 1984). Moreover, Sweden's work-oriented policies have
been far more effective than British (or U.S.) policies in preventing the
creation of a visible class of "welfare poor." Among families headed by solo
mothers, for example, public assistance benefits have constituted a relatively
minor income supplement to work income: more than 80% of solo mothers
are in paid employment and only around 10% of their family income has come
from welfare benefits, compared with more than half in Britain and the
United States (Rainwater, Rein, and Schwartz 1986).
In other respects, too, the mix of active labor market programs and
work-related social policies has been used with considerable effect by the
Swedes to reduce labor market disadvantages and redress social handi-
caps. In the mid-1980s about one third of the expenditure of the Labor
Market Board was devoted to disabled workers and one third to youth
measures (Standing 1988, p. 103). Sweden has achieved the world's
highest female labor force participation (more than 80%) via a unique
combination of measures facilitating labor supply, including elaborate
day-care facilities, with a social service expansion biased toward female
employment and welfare state transfers and taxes providing strong incen-
tives for women to work. Other incentives, such as the generous parental
and maternity leave programs, serve also to underline the Swedish fusion
of work and welfare as "the welfare state ... has taken upon itself to
permit employees to pursue non-work-related activities within the work
contract" (Esping-Andersen 1990, p. 155).

This. extension of the full employment commitment to women can be

attributed, at least in part, to the emphasis on wage earner unity and solidarity.
This was also to play a decisive role in the controversial pension reform
enacted in 1959 that had wide repercussions politically and for welfare state
strategies generally. The background to the reform was a growing dissatisfac-
tion with the modest benefits of the existing scheme that were leading better
paid employees to seek private insurance and occupational pensions to cater
for additional needs. Unlike the British governments at the time, the SAP and
LD responded by proposing a national earnings-related pension to supplement
basic minimum benefits, both to discourage private alternatives and to forge
a new "wage earners front" cutting across the divide between blue-collar and
white-collar workers. Politically, the events surrounding the reform marked
a watershed for the SAP that shifted the basis of its control over the state from
a parliamentary coalition with the Center Party and agrarian interests to an
electoral constituency based largely on the new alliance of wage earners
(Martin 1984). Moreover, the reform provided the impetus for further mea-
sures in other branches of social security and in housing policy aimed at
sustaining the wage earners front.
Compared with Britain's experience, Swedish reactions to the economic
recessions and welfare state crises of the 1970s and early 1980s revealed
the benefits of the distinctive work-welfare strategies. In this period
Sweden resisted serious "social disarmament" in the face of economic
difficulties that undermined its economy's competitiveness and led to
industrial profit and growth rates dipping even lower than in Britain. In
fact, Sweden experienced by far the largest increase in the number of
persons employed of all DECD countries. There was little evidence, too,
of the "casualization" process linked elsewhere to the growth of part-time,
temporary, and subcontracted work, and that has led to a serious erosion
of social rights (Standing 1988).
Nevertheless, recent events have raised fundamental questions about the
performance of the Swedish model. In the past this has depended on "the
presence of two crucial institutional conditions: centralized and solidaris-
tic trade union organization in political harmony with government policy;
and employer confidence and willingness to sustain high investment lev-
els" (Esping-Andersen 1990, p. 168). Both have been severely tested, and
not merely by the declining growth rates, rising inflation, and high taxa-
tion. One of the paradoxes of the 1970s recession was that its impact was
felt at a time when pressures were already building up within the Swedish
labor movement to move toward "economic democracy," particularly by
extending citizenship rights to control over capital through the introduc-
tion of "wage earner funds." In the context of the recession, this issue was
to lead to a bitter struggle between employers, unions, and political parties.
Roger Lawson 139

Its legacy, including the enactment in 1984 of a much diluted version of

the original Meidner plan, has been a more confrontational style of indus-
trial relations; but more significantly, a weakening of the much vaunted
wage earner unity and solidarity. Other factors have added to the tensions,
such as the shift from blue-collar to white-collar employment and a
tendency with this for wage determination to become more fragmented
and subject to wage drift and widening differentials (Standing 1988). More
generally, too, although, by comparison with Britain, Sweden's policies
have narrowed the scope for anti welfare backlash, it has undoubtedly
become more polarized on these issues (Marklund 1988). Whether this will
ultimately undermine Sweden's social democratic version of "social citi-
zenship" remains to be seen. But it serves to underline the main theme of
this chapter: That a nation's capacity for maintaining a broad consensus
and solidarity favoring wide-ranging welfare commitments depends cru-
cially on a close harmony of the spheres of work and industrial relations
and of social rights and obligations.


Ashford. D. E. 1986. The Emergence of the Welfare States. Oxford: Basil Blackwell.
Bornstein. S. and P. Gourevitch. 1984. "Unions in a Declining Economy: The Case of the
British TUC." pp. 13-89 in Unions and the Economic Crisis: Britain. West Germany
and Sweden. edited by P. Gourevitch et al. London: Allen & Unwin.
Christensen. A. 1980. "Disqualification from Unemployment Benefits: A Critical Study in
Swedish Social Security Law." Scandinavian Studies in Law 155-74.
- - - . 1984. Wage Labor as Social Order and Ideology. Stockholm: The Secretariat for
Future Studies.
Esping-Andersen, G. 1990. The Three Worlds of Welfare Capitalism. London: Polity Press.
Esping-Andersen, G. and W. Korpi. 1984. "Social Policy as Class Politics in Post-War
Capitalism: Scandinavia. Austria and Germany:' pp. 179-208 in Order and Conflict in
Contemporary Capitalism. edited by J. H. Goldthorpe. Oxford: Clarendon.
European Communities. 1977. The Perception of Poverty in Europe. Brussels: Commission
of the European Communities.
Giddens. A. 1982. Profiles and Critiques in Social Theory. London: Macmillan.
Gilbert. B. B. 1966. The Evolution of National Insurance in Great Britain: The Origins of
the Welfare State. London: Michael Joseph.
Goldthorpe, J. H. 1990. "A Response." In John H. Goldthorpe: Consensus and Controversy,
edited by J. Clark. London: Falmer.
Harris, J. 1972. Unemployment and Politics: A Study in English Social Policy 1886-1914.
Oxford: Clarendon.
Heidenheimer. A. J.• H. Heclo, and C. Teich Adams. 1982. Comparing Public Policies: The
Politics of Social Choice in Europe and America. London: Macmillan.
Hennock, P. E. 1981. "The Origins of British National Insurance and the German Precedent:'
pp. 86-104 in Emergence of the Welfare State in Britain and Germany. edited by W.
Mommsen. London: Croom Helm.

Hinton, J. 1982. "The Rise of a Mass Labour Movement: Growth and Limits." Pp. 20-46 in
A History of British Industrial Relations 1875-1914, edited by C. Wrigley. Brighton,
UK: Harvester.
Korpi, W. 1978. The Working Class in Welfare Capitalism. London: Routledge & Kegan
Korpi, W. 1983. The Democratic Class Struggle. London: Routledge & Kegan Paul.
Lawson, R. 1987. "Social Security and the Division of Welfare." Pp. 77-97 in Inside British
Society. edited by G. Causer. Brighton, UK: Wheatsheaf.
Marklund, S. 1988. Paradise Lost? The Nordic Welfare States and the Recession, 1975-1985.
Lund, Sweden: Arkiv forlag.
Marshall, T. H. 1963. Sociology at the Crossroads. London: Heinemann.
Martin, A. 1984. "Trade Unions in Sweden: Strategic Responses to Change and Crisis." Pp.
13-89 in Unions and the Economic Crisis: Britain, West Germany and Sweden, edited
by P. Gourevitch et al. London, Allen & Unwin.
Olsson, S. 1990. Social Policy and Welfare State in Sweden. Lund, Sweden: Arkiv forlag,
Orloff, A. and T. Skocpol. 1984. "Why Not Equal Protection? Explaining the Politics of
Public Social Spending in Britain, 1900-1911, and the United States, 1880s-1920."
American Sociological Review 49:726-50.
Pelling, H. 1968. Popular Politics and Society in Late Victorian Britain. London: Macmillan.
Perkin, H. 1989. The Rise of Professional Society. London: Routledge & Kegan Paul.
Price, R. 1986. Lobour in British Society. London: Croom Helm.
Rainwater, L., M. Rein, and J. Schwartz. 1986. Income Packaging in the Welfare State.
Oxford: Clarendon.
Rehnberg, B. 1984. "Labour Market Conditions." In OECD, High Unemployment: A Challenge
For Income Support Policies. Paris: Organization for Economic Cooperation and
Sinfield, A. 1968. The Long-Term Unemployed. Paris: Organization for Economic Coopera-
tion and Development.
Standing, G. 1988. Unemployment and Labour Market Flexibility: Sweden. Geneva: Inter-
national Labour Office.
Thane, P. 1985. "The Labour Party and State 'Welfare' " Pp. 183-216 in The First Labour
Party 1906-1914, edited by K. D. Brown. London: Croom Helm.
Therborn, G. 1989. " 'Pillarization' and 'Popular Movements.' Two Variants of Welfare
State Capitalism: the Netherlands and Sweden." In The Comparative History of Public
Policy, edited by F. G. Castles. London: Polity Press.
Titmuss, R. M. 1963. Essays on the Welfare State. London: Allen & Unwin.
Turner, B. S. 1986. Citizenship and Capitalism. London: Allen & Unwin.
Weir, M. and T. Skocpol. 1985. "State Structures and the Possibilities for 'Keynesian'
Responses to the Great Depression in Sweden, Britain and the United States." In
Bringing The State Back In, edited by P. Evans, D. Rueschmeyer, and T. Skocpol.
London: Cambridge University Press.
Whiteside, N. 1987. "Social Welfare and Industrial Relations 1914-39." Pp. 211-42 in A
History of British Industrial Relations, Vol. 2, 1914-1939, edited by C. J. Wrigley.
Brighton, UK: Harvester.
Part II

Organizations, Social Movements,

and Public Policy
Organizations as (Secondary) Citizens

Philippe C. Schmitter

Democratic theory has been extraordinarily varied in its definitions, as-

sumptions and prescriptions, but one element is virtually constant: the role
of citizenship. Regardless of the decision rules, the institutions of govern-
ment, the role assigned to deliberation and participation, the scope of
public policy, the impact upon the public-if there are no citizens, there
can be no democracy. They are a necessary, but not sufficient, component
of all forms of democracy-ancient and modem.
Moreover, in virtually all notions of citizenship, this is a property
exclusively assigned to individuals. 1 To democratic theorists it seemed
self-evident that only human beings could have the capability or the right
to perform this role. This has not always been the case. For example, in
many medieval cities, membership in the Burgertum or the Cive was
determined by one's status in an occupational community. Only by being
a member-in-good-standing of a recognized guild could a person enjoy the
full rights and bear the full obligations of living in the polity, and the
system of representation was set up accordingly so that each guild was
collectively entitled to so much participation and responsible for providing

AUTHOR'S NOTE: This is a segment from an unpublished paper titled, "Corporative

Democracy: Oxymoronic? Just Plain Moronic? Or a Promising Way Out of the Present
Impasse?" The themes in both these efforts emerged from lengthy discussions at the Center
for Advanced Studies in the Behavioral Sciences in Palo Alto during the 1987-1988 academic
year. I am particularly indebted to Claus Offe for his original reflections and relentless
criticisms. James Fishkin, Terry Karl, Wolf-Dieter Narr, Ulrich Preuss, and Gunther Teubner
also participated in these discussions at one time or another and I wish to thank them for
their contributions-and to absolve everyone from any respansibility for the subsequent
product. An earlier version ofthis chapter was presented at the Conference on "A Crisis ofthe
Welfare State Model, "Institute ofSocial Science, University of Tromse, August 21·23, 1989.


SO many resources (cf. Black 1984). Persons, of course, ultimately sat in the
municipal council and bore arms in defense of the city, but they did not do so
as individuals with their own will, preferences, or capacity for deliberation.
These city-states may not have been democratic by contemporary criteria and
they were soon displaced and disbanded by processes of state-building, but
they did leave an important and distinctive legacy to the Western political
tradition: that of representative and limited self-government.
Following upon the French Revolution's destruction of the remnants of
guild privilege, the advent of political liberalism definitively consecrated the
individualist basis of citizenship-first, restricting it to a few "worthy"
persons with sufficient socioeconomic independence to make autonomous
judgments and an adequate stake in the game to encourage them to be prudent
in making those judgments; and, later, extending it to less "worthy" persons
in exchange for their allegiance and conformity. With the exception of such
"oddities" as the Bavarian Senat or part of the Irish Seanad, formal represen-
tation in European parliaments came to rest exclusively on the votes of
individuals grouped according to territorial constituencies-at least, in those
countries with democratic regimes. Even the authoritarian regimes that ex-
perimented with various forms of corporatist representation in the interwar
period (and that persisted in Spain and Portugal until the mid-1970s), still
retained some system of individually defined, territorially based representa-
tion. They decried "individualism" in their rhetoric, but they found it difficult
to eradicate in their practice (Schmitter 1975).


What if we compare this standard tale of the inexorable (if hardly linear)
triumph of Western political individualism with the reality of contempo-
rary liberal democracies? Who actually participates in the making of
collective decisions and determines their outcomes? Certainly, not many
individual citizens. Even if we take into account the representative nature
of these polities and ask whether those who have been chosen in a
competitive struggle for the votes of individual citizens effectively control
the selection and implementation of policies, the answer has to be ambig-
uous. In some of the more "public" arenas where policies are standardized
and parliamentary review is still effective, it might be "yes," but not in
those where technical expertise, detailed information, and specific forms
of consent are required if policy is to met its designed ends. Put more
concretely, the expansion of the modern state into the realms of social
protection and economic promotion has shifted the balance against indi-
viduals playing the role of citizen-an ironic development, since it was in
Philippe C. Schmitter 145

the name of their welfare and security that these innovations were intro-
duced. Moreover, even where elected representatives are prominent ac-
tors, individuals are not the only "citizens" who determine the outcome of
the competitions that bring them to power.
Organizations or, better, associations have taken the place of individu-
als. 2 The extent of this displacement varies considerably from country to
country and sector to sector, but the trend is clear. Modern democracy is
increasingly becoming organized democracyr' Moreover, as associations
grow in number, extend their coverage, augment their resources, and
consolidate their access to public authorities, the making of binding collective
decisions becomes more and more the outcome of interorganizational bar-
gaining, less and less a struggle between organizations and individuals,
and even less a competition among individuals.


Democratic theorists have not made much of an effort to cope with this
sea change. There have long been complaints about "pressure groups,"
"lobbying," and "special interests." Policy measures have even been
designed to monitor and limit their intrusion into politics. But the super-
ficiality of the former and the ineffectiveness of the latter suggest that the
basic issue has not been joined: what difference does it make if many of
the de facto "citizens" of modern democracy are organizations and not
individuals? To pose this question properly would require a critical exam-
ination of the liberal foundations upon which Western democracy rests; to
answer it satisfactorily would require nothing less than a "postliberal"
theory of modern democracy." Strictly within the liberal corpus, two
theoretical arguments have been advanced to resolve the practical di-
lemma posed by the rise of organizations.

Organizations as Democrats
The first suggests that organizations pose no distinctive threat-factual
or normative-to democracy because they themselves are constituted
democratically. It is even argued that, quite the contrary, these new interest
intermediaries significantly enhance the democratic potential for individ-
uals by providing them with proximate arenas where they can tryout their
sense of personal efficacy, learn the rules of civility and deliberation, and
experiment with the practice of competition and compromise. The locus
classicus for this argument is Alexis de Tocqueville's De la Democratie
en Amerique ([1835 and 1840] 1969), which has been used repeatedly by

American pluralists to justify their optimistic conclusions.i It should be

pointed out that de Tocqueville never imagined that the associations
politiques he found so admirable in the United States of the 1830s would
become permanent organizations-least of all, that they would develop
large staffs and complex internal relations independent of their members.
Nor does de Tocqueville seem to have been aware of what every American
survey since has conclusively demonstrated, namely, that associational
membership is very unevenly distributed across the population and that a
relatively small proportion of individuals, already favored in wealth and
status, accumulate most of the leadership roles (Schlozman and Tierney
1986; Schlozman and Verba 1979). There is mounting evidence that many
American interest groups may have contributors, but no individual mem-
bers in the conventional sense (Salisbury 1984). One can hardly fault de
Tocqueville for not having anticipated Robert Michels's "Iron Law of
Oligarchy" (1962), but there is no excuse why subsequent students of
pluralist democracy have so persistently overlooked the prevalence of
nondemocratic practices within the internal processes of these associa-
tions (for an exception see Upset et al. 1956). Moreover, there is good
reason to believe that oligarchy in this realm is not accidental or incidental,
but functionally linked to the efficient performance of the primary task of
these organizations, that is, the defense and promotion of the specialized
interests of some social or economic category. To accomplish this under
modern conditions requires technical expertise, extensive data gathering,
full-time dedication, continuous access, and the development of long-term
trust among bargainers-all of which are antithetic to the sort of ad hoc,
spontaneous, amateurish, deliberative, and accountable associability that
de Tocqueville envisaged.

Organizations as Individuals
If associations cannot be plausibly described as "minidemocracies," the
next best solution may be to ignore their internal properties altogether and
to treat them as if they were no different than individuals. The participa-
tion, accessibility, accountability, and competition that one fails to find
consistently within specific associations can be shown to prevail between
associations representing different categories of interest. Just as each
individual in modern society is composed of a "bundle" of multiple,
overlapping, and, at times, competing interests, so the polity as a whole
can be conceptualized as a "bundle" of similarly structured interests, each
represented by a different organization or set of organizations. In this
version of democratic theory, the pluralist model is simply transposed
from one level of aggregation to another. Organizations are equated with
Philippe C. Schmitter 147

individuals and no apparent need arises to modify the liberal postulates

that underlie most versions of modern democratic theory (Truman 1962).
Aside from the general strictures against anthropomorphism in analyz-
ing social phenomena, there are at least five specific reasons why this
equation is potentially misleading:

(1) Inequality. For liberal democracy to function plausibly, it is impor-

tant that individuals can be said to be equal in some specific and significant
sense, for example, as voters or candidates, as defendants or plaintiffs, or
as producers or consumers. "Equality before the Law," "One Person One
Vote," and "Consumer Sovereignty" are among the most effective slogans.
Even if subsequent choices do not produce a substantive equality of
benefit, the process of arriving at them ensures a formal equality of
procedure-and that contributes to their democratic legitimation.
There is no way that associations can be described as equal. By gathering
contributions from many persons, firms, and social groups, they can accumu-
late "artificially" much greater disparities in resources than are likely to
prevail "naturally" among individuals. Moreover, if the liberal rule of volun-
tary choice applies to these contributions, we have every reason to expect that
differences that already exist in society due to the uneven distribution of
property and income (and which formal legal and political equality is sup-
posed to mitigate) will be augmented. Those groups that are "privileged" in
the sense of Mancur Olson (1968), that is, those that are few in number or
compact in distribution, will have a greater rational incentive to contribute in
the first place. They are also more likely to conform to the subsequent norm
of action because it is easier to monitor and, presumably, to control their
behavior. Large dispersed groups will be systematically underrepresented.
This could be labeled the "Paradox of Liberal Associability," namely, that
those who might benefit the most from the freedom of collective action will
be the least likely to engage in it.
Moreover, even where the rules require a formal parity of representation
between associations with differing material and human resources, the effect
of this is vitiated by the infrequency with which voting is used to make
decisions in such contexts. Instead, the norm seems to be the expression of
intensities, the exchange of threats, the elaboration of a compromise based on
proportionality, and, eventually, decision by mutual consent (or "concurrent
majority,,)6-none of which are intrinsically democratic devices.

(2) Amorality. Liberalism and, with it, liberal democracy are frequently
described as resting on the twin assumptions of self-interest and "non-
Tuism," that is, indifference to the benefit of others. It would be more
accurate to say that both of these are, in turn, contingent upon what Adam

Smith called "moral sentiments." Unless individuals in situations of con-

tract and conflict have been socialized into accepting certain prior rules
for treating each other with fairness, self-restraint, tolerance, and respect
for the system as a whole, it is difficult to imagine how there could exist
sufficient trust for momentary deals to be made, much less for establishing
long-term agreements, not to mention for respecting the victory of one's
opponent in an election. These "precontractual conditions for contract"
may be precapitalist in origin and they may be eroding rapidly under
contemporary social and economic conditions (Hirsch 1978), but they
remain essential for the viability of modern democracy.
Ethical reasoning and the inculcation of norms would seem to be exclusive
properties of individual human beings. Associations lack a single center for
gathering and internalizing these commands, and a mechanism for rendering
obedience to them compulsory. They may talk frequently about "codes of
conduct" for their respective classes, sectors, or professions, but these do not
have the same imperative status as a personal moral norm (and, moreover,
they apply to the behavior of associational members, not to the association
itself). Organizations can, of course, be held to certain norms, but this is
dependent either on the personal convictions of those individuals who mo-
mentarily compose them, or on prudential calculations about the likely
reaction of other organizations, especially state agencies. Their primary
functional task and, therefore, organizational responsibility rests with the
defense/promotion of member interests. This requires that they act with
instrumental, not purposive, rationality. For example, if an association were
to interject into the process of influencing policy some ethical norm that was
not strongly and uniformly held by its members, or that seemed imprudent in
the face of competing interests, it would risk being rejected by its members
or reviled by its interlocutors---even if such an action were in the "general
interest" of all. 7

(3) Immortality. The fact that humans are mortal poses a serious problem
for a rational interest-based conception of democracy. Why should any
one individual pursue a course of action many of whose benefits will only
accrue after his or her death? Why should existing persons be concerned
with the status of the polity or, even less, that of the entire earth when they
have left the scene?8 Part of the answer lies in the ethical socialization
stressed above, part in the temporal extension of oneself through family
and immediate community, but much depends on the immortality of
associations and other organizations. These "permanent persons" define a
continuous identity, establish standard operating procedures for promot-
ing it, and provide a living link between generations. Most of them
antedate the individuals that compose their membership at any given point
Philippe C. Schmitter 149

in time and most can be expected to persist long after their members have
died or moved on to other occupations or causesr' By their sheer exis-
tence-not to mention their deliberate efforts at publicization and indoc-
trination-associations provide many of the fixed symbols that orient
individuals in the selection of their social identities. They also layout a
complex set of incentives that can induce individuals to contribute to
collective action by offering them relatively immediate rewards. By link-
ing these incentives over time to successive groups of contributors, a basic
continuity of interest behavior is maintained that would be otherwise
difficult to explain on a purely individualistic basis.
Moreover, associability helps solve another, related, puzzle about lib-
eral interest politics. Individuals should have a very high rate of discount
for future benefits. They are compelled to act myopically. both by the stark
fact of their own mortality and by the sheer dynamics of the modern
society that surrounds them. And yet, no polity could persist without at
least some long-term contractual relations among major social actors. In
capitalist economies that entails a stable compromise between possessing
and nonpossessing classes. Ethical socialization may contribute to the
interpersonal trust that makes such agreements possible; political consti-
tutions can have the effect of binding future generations to a revered set
of rules; individual citizens may calculate that defection from an estab-
lished distribution of benefits may be too costly; but only associations and
other organizations can deliberately adopt and then effectively impose a
longer time perspective on their members/clients/suppliers. What gives them
this capability is their immortality, presided over by a permanent administra-
tive structure and a constantly renewed staff. Persons in these roles have an
incentive to calculate interests on a more future-regarding and other-regard-
ing basis, and to try to convince members to act accordingly. This shift from
an individualist to an organizational basis of negotiation does not ensure that
agreement will be reached among "social partners." Nor does it guarantee that
individual members will always come to accept a less myopic definition of
their interests and conform willingly to contracts made in their name. Never-
theless, it opens up the possibility for a style of conflict resolution that is
denied by participation on a predominantly individual basis.i''

(4) Omniscience. Individuals are strictly limited in the amount of atten-

tion they can give to politics. For one thing, they can only be in one place
at a time. For another, they can only be expected to keep up with a
relatively narrow range of concerns. More importantly, they must meet a
variety of claims on their resources-and there is abundant evidence that,
under modern conditions, most people prefer to spend their scarce time
and money on things other than public deliberation and decision making.

In the democracy of classical Greece, slaves freed their masters for

extensive political participation; servants performed much the same func-
tion in the democraties censitaires of the 19th century. Under contempo-
rary conditions, even the minimalist act of voting as an expression of
individual citizenship has shown some tendency to decline.
Associations are not so limited in their attention to politics-indeed,
they are singularly devoted to doing just that. They and other types of
organizations can be in several places at once and keep up with a wide
range of contentious topics. Within the limits established by their capacity
to extract resources from the environment, they can hire specialists and
technicians and organize them into complex divisions of labor that can
monitor events over a long period and broad scope. Their continuous
existence provides them with stockpiles of resources and elaborate mem-
ories that can be mobilized and brought to bear on new issues. They can
also be used as a potent force for the "routinizing" of existing ones.
In short, organizations not only have the potentiality of lengthening the
time span of political calculation due to their immortality, but also of
extending the span of policy attention and control through their omnisci-
ence. An individualistic model must assume a discontinuous and volatile
process with spurts of enthusiasm and cycles of "involvement," as persons
distribute their scarce resources and limited interests episodically (Hirsch-
man 1982). An organizational model of democracy can be expected to
reduce the variation and even out the attention between elections and
scandals-following issues quietly and persistently when the public at
large has lost interest or is distracted by other matters.

(5) Indispensability. Individualistic theories of liberal democracy must

assume a high degree of voluntarism, that is, that the decision to engage
in politics and to withdraw from politics depends on the autonomous will
of persons-and is relatively costless to make. Ease of "voice" and "exit"
are formally ensured by a broad range of constitutionally protected free-
doms: speech, petition, assembly, due process, and so forth, and by certain
critical separations: church and state, public and private, civil and military,
and so on. The behavior of public officials and private intermediaries can
be controlled by articulating complaints against them or by simply going
without their services-even by just threatening to do one or the other
(Hirschman 1970).
One of the reasons why students of democracy could for so long ignore
the effects of associability is because they assumed that, in such a plural-
istic order, exit or alternative voice was always possible and readily
available. If the leaders or staff of organizations promoted policies not in
agreement with the preferences of their members, the latter would simply
Philippe C. Schmitter 151

withdraw from the existing relationship. They could either organize a

dissenting faction or leave the organization altogether and take their
preferences, even their persons, elsewhere. This control mechanism (which
could even operate by anticipated reaction) ensured a strict congruence
between individual and collective action-and, coincidentally, freed the-
orists of any need to rethink the bases of liberal democracy.
As we saw above, persistent evidence about oligarchy has taken away
much of the compellingness of the argument that members can change
association policies by expressing "internal voice." Although by no means
impossible, member revolts are costly both to those who instigate them
and to the organization's longer term effectiveness. Exit would seem the
more attractive alternative. This, however, presumes that associations are
"dispensable" in one of two senses: (1) the member can do without the
services being offered; or (2) the member can find an alternative supplier.
Both have declined in plausibility. In some cases, membership has simply
been made involuntary, for example, the mandatory Bar Associations,
Ordres and Guilds for selected professions or the Chamber systems for
small businesses, artisans, agriculture---even for workers in Austria. In
others, the intermediaries provide state-mandated services that are either
compulsory, for example, professional licensing, or indispensable for the
welfare of members, for example, price subsidies, favorable credit, insur-
ance coverage. In still others, organizational competition in the past and
state recognition in the present have lead to situations where only a single
association represents a given category of interest. Under such monopo-
listic conditions the cost of exiting and creating an alternative voice is
prohibitive. By whatever stratagem, the associations have made them-
selves indispensable parts of the political process.
As the evidence from inequality, amorality, immortality, omniscience, and
indispensability accumulates, it becomes increasingly clear that democratic
theory cannot afford to equate organizations with individuals. It must modify
its basic assumptions and the way to accomplish this, I submit, is to take (and
to treat) organizations seriously as "citizens"-admittedly still secondary to
individuals-but nevertheless crucial elements both in understanding how
modern democracy works and how it can be improved.


Citizenship is a much discussed, if unresolved, concept. It seems to

begin with the acquisition of a status, protected by law, that grants general
equality of opportunity and treatment with respect to a (varied) bundle of
rights and obligations; becomes a process whereby those who have that

status make use of it, both to reduce their political uncertainty and to satisfy
their interests; and presumably terminates in a result that is the integration of
the citizen into the national culture of the polity that granted himlher/it the
status in the first place. Needless to say, in the real world, there are many
disjunctures in the scenario: the distribution of the status may be too narrow
or the content of the rights too restricted to protect citizens from arbitrary
treatment, social discrimination, or economic exploitation; the beneficiaries
of the process may exploit their favored status to increase the uncertainty of
others and to satisfy their interests at the expense of those who are excluded;
the outcome of the process may not always be to generate loyalty or a sense
of community, but to convince citizens that some other regime or polity can
offer them greater security or satisfaction. For it is important to stress that the
granting of citizenship logically involves a complex set of choices about
inclusion and exclusion, and is closely linked historically to the development
of national consciousness and the nation-state. In the contemporary context
this might even seem anachronistic given the "globalization" of so many
economic and social relations.
In a justly famous essay, T. H. Marshall (1950) suggested that the
content of citizenship has evolved over time. As he reconstructed the
process for Great Britain, initial concessions by authorities provided a
lever that could be wielded to demand others and these, in turn, led to
further claims for equal treatment. It began in the 18th century with the
struggle for equal rights and obligations before the law, expanded in the
19th century to cover equal formal participation in political life, especially
through universal suffrage, and in the 20th century shifted its attention to
equal opportunity to share in the country's material and cultural heritage,
especially through state provision of welfare. 11 Marshall's sequential
account is logically convincing, but it is not historically generalizable.
Both France and Germany took rather different routes to reach, more or
less, the same bundle of rights and obligations in the end.
It is also not clear from Marshall's analysis whether the process of
citizenship has terminated with the attainment of the contemporary wel-
fare state, or whether that merely places new levers in the hands of more
actors demanding further extensions of equal rights (and, to a lesser extent,
equal obligations). Ralf Dahrendorf, for one, sees a serious danger in
pushing its principles into new domains:

the extension of citizenship has reached in recent years, the apparently insur-
mountable walls of ascribed status. Men and women are not merely to be given
the suffrage and equal wages for equal work, but they are supposed to be treated
as equals in all respects; society is to be arranged in such a way that the
differences can be ignored. (1974, p. 683)
Philippe C. Schmitter 153

From his account it is not clear whether Dahrendorf regards such an

eventuality as a social impossibility or rejects it because of the state
intervention it would require.
But what if the next extension did not involve pushing those who are
already citizens, that is, adult individuals, into new domains of equal right
and obligation? What if the frontier involved specifying and developing
analogous properties for those who are not yet recognized as citizens,
namely, associations? What might that entail? How might such changes
improve (or diminish) the quality of existing Western democracies? An-
swers to these questions could provide the sort of postliberal alternative
to the present impasse that is so badly needed.
To a limited degree, existing associations have already acquired "quasi-cit-
izenship" rights. Actually, it would be more accurate to say that they have
taken certain liberties because, instead of commonly defined and equally
accessible rights, they have succeeded on a piecemeal basis-class by class,
sector by sector, profession by profession-in carving out and enjoying
specific privileges. Some have been granted autonomy from state interference
and immunity for their actions. Many have been formally recognized by the
state, sometimes as monopolistic organizations, and accorded regular access
to its deliberations with a presumptive right to be consulted on all issues in
their respective domains before binding decisions are taken (Vernehm-
lassungsverfahren is the untranslatable Swiss term for this). Occasionally, a
few have even received the equivalent of welfare entitlements: to financial
subsidies, goods in kind, state-mandated obligatory dues, off-loaded public
services, and so forth. 12 What is missing from this tale of gradual and uneven
accumulation of associational privilege is a general and systematic practice
of equalizing access to these rights and obligations for all organized inter-
ests-not to mention an explicit and public justification of them on grounds
of "organizational citizenship."
What, then, could be the principles upon which such a justification
might be based? Rogers Brubaker (1989) has recently laid out the six
"membership norms" that he claims define the ideal-typical conception of
citizenship. According to this model, membership should be:

(1) Unitary, that is, all holders of the status should have full rights and obligations.
(2) Sacred, that is, citizens must be willing to make sacrifices for the state or
community that grants them the status.
(3) National, that is, membership must be based on a community that is simul-
taneously political and cultural.
(4) Democratic, that is, citizens should be entitled to participate significantly in the
business of rule and access to citizenship should be open to all residents so that,
in the long run, residence in the community and citizenship in it will coincide.

(5) Unique, that is, each citizen should belong to one and only one political
(6) Consequential, that is, citizenship must entail important social and political
privileges that distinguish its holders from noncitizens.

These six normative criteria could be applied, mutatis mutandis, to any

proposal that aimed at making associations into citizens. Even then, the
potential reformer should tread cautiously because, as Brubaker himself
notes, "this model of (individual-s-Pf'S) citizenship is largely vestigial.
That it survives is due less to its own normative force than to the lack of
a coherent and persuasive alternative. It is, moreover, significantly out of
phase with the contemporary realities of state-membership" (1989, p. 5).
With this caveat in mind, let us then evaluate what might be involved.

Unitary: Here, the imperative is that the status of "associational citizen"

be general, as well as equal. In principle, no organization should be subject
to only some of its rights and obligations, and none should enjoy special
treatment. At present, the associational universe of most Western countries is
shot through with islands of exemption and privilege involving such things
as obligatory memberships, ear-marked subsidies, specific delegations of
public authority, unique fiscal treatments, and so forth. It may be asking too
much to abolish all these. They have emerged, often unobtrusively, over a
long period of time and some of them serve defensible public purposes, for
example, the licensing of "sensitive" professions through self-government by
obligatory guilds or orders. Moreover, existing associations would certainly
struggle to retain their privileges (and divest themselves of any special
obligations). Application of the unitary principle, however, would have the
salutary effect of, at the very least, compelling specific organizations to justify
their special treatment and even of making them apply for the organizational
equivalents of "resident alien" or "diplomatic immunity" status.

Sacred: This principle has been considerably attenuated for individuals.

Not only has citizenship been "secularized," at least in the West, by the
decline in militant nationalism and national antagonisms, but the obliga-
tions historically linked to it have declined. With the abolition of conscrip-
tion and militia armies in some countries, individuals are no longer
required to die for their nation or state. 13 The various symbolic allegiances
expected of citizens, for example, flag saluting and anthem singing, have
become increasingly ritualistic, 14 and one hears less frequently that classic
slogan of unreflective citizen allegiance: "My country, right or wrong!"
Sacredness for modem associations might hardly differ from that of
modem individuals: an obligation to obey the law, to pay taxes, to serve
Philippe C. Schmitter 155

others in time of emergency, to treat fellow citizens with respect and

tolerance, and to be prepared to make sacrifices in one's own interests for
the good of the society as a whole. That last item is, of course, the most
difficult because associations are specialized in the promotion/defense of
the self-regarding interests of a part of society. Nevertheless, it doesn't
seem excessive to ask that, as part of the process of becoming citizens,
associations in certain situations be required to "other-regard," to set aside
the satisfaction of their own interests for the good of the unit as a whole.
Defining those situations and deciding what to sacrifice will not be easy, but
then that is what "high politics" are all about-for individuals, as well as
organizations. Moreover, if the associational universe were structured along
citizenship lines, their very adversarial behavior might provide the best
possible definition of "the public interest" and guarantee that it would prevail.

National: Nowhere have the principles of citizenship diminished more that

on this item. It is no longer so preposterous to describe oneself as "a citizen
of the world" or, certainly, "a citizen of Europe." More to the point, the act
of acquiring citizenship used to imply the joining of a coherent, well-defined
community with its own homogeneous culture. In many countries, this in-
volved a lengthy and demanding "apprenticeship" in national laws and
mores. 15 First, foreigners were made into natives and then-and only then-
were they granted citizen status. Indeed, naturalized second-generation im-
migrants were often said to behave "more American than the Americans."
Nowadays, those becoming citizens can insist that they be allowed to retain
their culture of origin and not be thrown into the "melting pot." They may
even legitimately demand that the state of which they have become a citizen
subsidize and protect their subcultures from discrimination and attenuation.
In short, citizenship for individuals is no longer a guarantee of membership
in a unitary corresponding culture.
Can one ask more of associations? Must they be "national" in some
sense in order to be citizens? Certainly, the answer in the narrow sense
would be "no." For example, why should they be prohibited from allowing
foreign firms, practitioners, or workers to join their ranks? By being inside
the national system of production, these potential members inevitably
contribute to the formation of interests there, and it would seem self-de-
feating to force them to use external channels for the expression of these
interests. Even associations composed exclusively of foreigners resident
in a given territory should have a putative right to become citizens-if they
choose to do so. The national criteria, however, becomes more relevant
and defensible when it is applied to the relationship between associations
and public authorities. Beyond obeying relevant national laws as part of
their "sacred" imperative, should these new citizens be required to comply

with the informal cultural norms that surround the political process? Must
they conform to the "policy style," the "standard operating procedures," the
"rules of thumb" of the country they are joining? Except for a prohibition on
"extra-territoriality," that is, on insisting that they be treated according to the
norms of their homeland, there may be no reason to apply the national
principle to enforce conformity. The issue is likely to sort itself out much as
it does with individual citizens. They will test the flexibility of local norms
and, if these are ingrained and tenacious enough, the new collective citizens
will conform to them out of political prudence. If not, some "hybrid" or
"supra-national" synthesis will emerge to replace them.i"

Democratic: This is the crucial principle from the perspective of this

chapter. Indeed, its major premise is that reforms that would make interest
associations into secondary citizens would improve the practice of modem
democracy. In the traditional, individualist, conception of citizenship, the
core concern was participation-with acquiring the right to vote in contested
elections, to form and join associations, to speak out publicly and petition
authorities for redress, to consult and be consulted on matters of policy, and
even, in direct democracies, to be physically present and decide on all issues
binding on the community. A secondary concern was openness-keeping the
polity accessible to previously excluded actors, issues, and ideas.
This is not the place for a lengthy disquisition on the impoverishment
of individual participation in modern democracies-much has already
been written on the subject. 17 Indeed, one of the reasons why the reform
of associability is felt to be urgent is because it may help to compensate
for the declining significance of "primary" participation. For organiza-
tions to participate democratically in political life, they must first have the
necessary material and human resources to make their case effectively,
and these must be distributed on a basis that is not systematically skewed
by preexisting social and economic inequalities. As we shall see, this
requires that they be cut loose from exclusive dependence upon voluntary
contributions. They must also enjoy regular and assured access to public
authorities and this, too, must not be biased in favor of preferred clients
or sponsors. Short of major changes in the recruitment and socialization
patterns of civil servants and higher executives or a large-scale disman-
tling of existing capacities for collective action by privileged groups (both
dubiously democratic, even if preferred by a majority of the citizenry), the
only way of tackling this requisite is to diffuse the "art of association" as
widely as possible and to rely on a more balanced adversarial process to
pinpoint "the mobilization of bias."
Openness hinges on whether presently excluded or not yet existent
organizations can expect to get into the influence game without unreason-
Philippe C. Schmitter. 157

able effort. Setting too high a threshold on their acquiring citizenship

status and access to authorities would rig the game in favor of those
interests that are already "vested"; setting too Iowa threshold could bring
into play spurious interests that are not capable of representing the cate-
gories they lay claim to.
A second aspect of openness is perhaps better expressed by the German
term, Ojfentlichkeit. This implies that democracies arrive at their decisions
through a public process of deliberation in which citizens, individual and
organizational, restrict their actions and positions to those that are publicly
visible and defensible. They must be prevented from using their resources
to act collusively with authorities and to arrive at private arrangements
that undermine or circumvent public policies. Since "special interests" and
"lobbies" are frequently accused precisely of operating in this manner, the
expectation is that the acquisition of citizenship and the rights that go
along with it would be accompanied by obligations that would have the
effect of driving such behavior into the open.
Finally, participation and openness may not be enough when it comes
to ensuring the democratic bona fides of associational citizenship. We do
not insist that individuals be "internally democratic" in order to qualify
for citizenship-much as we may prefer that they be civic-minded, toler-
ant, and willing to compromise-but this may be a reasonable condition
to impose upon associations. This should not extend to dictating the sort
of political arrangements and institutions that ought to exist within each
organization, even less the range of norms or interests each is entitled to
promote/defend. There is simply too much variety in member interests for
this sort of detailed interference to be feasible, but a general requirement
that all associations conduct regular elections for executive posts, and that
they specify constitutionally the rights and obligations of their members
with respect to the associations' actions, does not seem excessive. Admit-
tedly, policing this would be difficult and one can easily imagine strategies
for getting around the competition and accountability; nevertheless, such
criteria might prove useful in resolving the always thorny problem of how
to deal democratically with the enemies of democracy-since it is highly
likely that some groups that are normatively opposed to democracy will
seek to take advantage of its processes.

Unique: This is a quality that presently seems most anachronistic at the

individual level and should not be imposed on organizations. Many per-
sons now have dual citizenship without this causing great confusion or
corruption. Once the tight link between citizenship and national culture
had been loosened, this came as an obvious adjustment to an increasingly
interdependent and physically mobile world. Individual firms and persons

should be allowed to join and contribute to as many associations as they

wish, in as many countries as they live or operate. Associations should not
only be allowed to operate in more than one political jurisdiction-beneath
or beyond the nation-state-but they might even be encouraged to do so.
Dual associational citizenship in the form of membership in broader,
supranational hierarchies would seem to be the wave of the future-espe-
cially in post-I 992 Europe.

Consequential: There is no doubt that the acquisition of citizenship by

individuals no longer carries with it the sizeable consequences it once did.
In contemporary Western democracies, resident noncitizens are protected
from most of the dangers of arbitrary and discriminatory treatment by the
general operation of the "state of law," and they are usually eligible for
the benefits of the "state of welfare"-once they have lived there for a
sufficient period and if they can demonstrate that they entered legally.
They can even exercise some participatory rights, such as forming associ-
ations and petitioning authorities, although with minor exceptions they are
denied the right to vote in elections. One indirect proof of this growing
inconsequentiality would be the increasing proportion of those who are
eligible to become citizens of the country in which they reside who do not
even bother to attempt to do SO.18
The consequentiality of extending citizenship to organizations is likely
to be more substantial, but diverse in its impact. Those "privileged"
associations already well funded and enjoying assured access would gain
relatively little. In fact, they might even stand to lose some of the advan-
tages they presently have under liberal, voluntaristic arrangements: free-
dom from responsibility for their actions; the option of withdrawing from
negotiations; the possibility of using inconfessable means and of making
secret deals; the absence of any obligation to control member behavior;
the satisfaction of articulating ideological purity; the need to be account-
able to general interests. Some of these they may be willing to "trade
away" for the status and greater security of becoming citizens. The major
impact, however, is likely to be felt by those categories of interest that can
presently not attract sufficient support to make their voice heard effec-
tively, that are restricted in their access for one reason or another or that
seek to represent "submerged," "unconventional," and "unpopular" causes.
Obviously, a great deal would depend on the specific content of the
reforms that would be coupled with the granting of organizational citizen-
ship, that is, the precise mix of rights and obligations that would go along
with it, but their overall effect should be designed to distribute the oppor-
tunity for permanent collective action as equally as possible across all
categories of interest, both those presently in existence and those that
Philippe C. Schmitter 159

might emerge in the future. If this can be accomplished, the consequence

would be substantial-perhaps as substantial as that of granting the electoral
franchise to workers and women, or the freedom of association to individuals.
If we are to make organizations and, more specifically, interest associ-
ations, into responsible and effective secondary citizens, deliberate re-
forms must be introduced. Their "natural behaviors" when competing in
the marketplace and/or attempting to influence public policy under the
prevailing norms of liberal democracy will not lead to this outcome. To
overcome their propensity for enhancing existing socioeconomic inequal-
ities through the systematic over- and underrepresentation of interests, to
reduce their temptation for engaging in opportunistic attempts to seek
momentary advantage, to encourage them to provide a stable and encom-
passing framework in which primary citizens, that is, individuals, can
deliberate about their interests, to enhance the probability that competing
versions of these interests may be expressed, and to ensure that all poten-
tial categories of citizens have more equal access to collective action and
representation, the rules of the game with regard to associability in modern
democracy should be changed.
This is not the place to explore specific proposals for reform in detail,
but three guidelines seem desirable:
(1) The need for organized collective action should be separated from the
ability or willingness to pay for it, that is, the voluntary right of association
must be coupled with an involuntary obligation to contribute to it. One could
call this the principle of No Representation Without Taxation.
(2) The support for associations should be subject to an open and
competitive process of selection by primary citizens, that is, the receipt of
funds should be made conditional upon the distribution of vouchers by
individuals. This might be termed the principle of Secondary Citizenship
Through Vouchers.
(3) The associations entitled to compete for voucher funds should agree to
conform to certain standards of practice in their internal procedures and
norms, that is, those eligible to receive vouchers should agree to respect such
practices as periodic elections for officers, respect for minority rights, public
inspection of financial records, and so forth. The label for this principle might
well be Democratic Accountability Through Semi-Public Status. 19


1. Robert Nisbet (1974) has stressed the historical connection in the West between the
rise of citizenship and the assertion of what he calls: "political individualization." He also
hints at a close link between the actual timing of concessions of citizenship in Europe and

the advent of war. The former makes the process seem almost functional-a necessary
adjustment to a prior and independent change in values; the latter depicts it more like a
calculated strategy, a "ransom" that elites paid to the masses to encourage them to fight or
compensate them afterward.
2. Both organizations and associations are forms of permanent collective action. Both
have a formal internal structure, a regular staff and a distinctive set of "standard operating
procedures," that is, they are bureaucratic to some degree. Associations (or, better, interest
associations) are a subspecies of the genus. They specialize in acting as intermediaries
between public authorities and differing categories of, for example, persons, firms, localities,
ethnies, or social groups. Organizations with a wide variety of primary tasks-such as
production, distribution, socialization, enlightenment, entertainment-may also intervene on
behalf in matters of public policy, but they do so on a more episodic and ancillary basis.
3. The earliest expression I have found of the awareness of the growing role of organi-
zations in political life is an obscure pamphlet in the Library of Congress by William Ellery
Channing (1830). If one were to trace the contemporary genealogy of this notion, I suspect
that its origins would be found-like so many other of our social scientific conceptions of
modernity-in interwar Germany. During the 1920s the "legal sociologists" or "sociological
jurists," Hugo Sinzheimer, Ernst Fraenkel, Franz Neumann, and Otto Kahn-Freund, devel-
oped the idea of a new social order they called "collective democracy"-in obvious affinity
to the then-prevalent notion of "organized capitalism." For a revival and discussion of the
concept, see Wolfgang Luthardt, (n.d.). Scandinavians should also be given a share of the
credit. Heckscher (1946) is an early expression of this awareness; Olsen (1983) is a more
recent one. See also Gustavson (1966) and Presthus (1962).
4. A conference was recently convened by James Fishkin on the topic of "Competing
Theories of Post-Liberal Democracy," University of Texas, Austin, TX, February 8-10, 1991.
Hopefully, this meeting will give rise to a volume to be edited by Fishkin, Claus Offe, and
5. For a brilliant new effort to reinterpret de Tocqueville in the light of contemporary
social science theory, see Hadari (1989).
6. The standard reference in the American political tradition is Calhoun ([1853)1943).
For a contemporary version, see Steiner (1974).
7. This charge of amorality seems to have disturbed many of the readers of an earlier
version of this chapter. They all had some organization in mind that "did good" and, therefore,
must have been capable of moral reasoning. The Red Cross, Amnesty International, the Sierra
Club, and other environmental groups were frequently mentioned. Leaving aside the fact that
for each of these do-good groups one could cite many historical examples of organizations
that committed egregiously immoral acts-even on the part of those such as churches that
are supposed to specialize in purveying morality-the real issue is whether it is realistic to
expect organizations to respect general norms when it is not in their members' interest and
when they may not be punished by the state or some interlocutor for doing so. Would they
continue to act according to internalized moral norms if their membership changed its norms
or if their resource basis were threatened severely? Certainly, the American Civil Liberties
Union has taken stands that were unpopular with wide segments of the American public, but
did these cause defections in their core membership or staff? If so, this demonstrates that
associational action may be conditioned by individual morality, but it doesn't prove that its
action per se is moral.
8. An analogous problem could be termed that of human mutability. The fact that
individuals may change their occupation, residence, social connections, and even nationality
in the course of their lifetime makes it irrational for them to "invest" in any given interest
identification because the collective action involved may only yield benefits to others holding
Philippe C. Schmitter 161

that role in the future. Roles that are virtually defined by their transitory nature (e.g., those
of student, apprentice, or conscript) would go unprotected. The answer, of course, is that
associations that outlive any of their immediate members ensure continuous coverage of such
interest, while inducing individuals to contribute for limited periods by providing them with
selective material goods or immediate status gratifications.
9. Actually, very little is known about the mortality rate of interest associations because,
at best, only "birth certificates" are handed out and almost never "death certificates." We do
know that their formation tends to come in waves and that some polities seem to have higher
rates of natality than others; see Schmitter and Brand (1979). What little research has been
conducted on the turnover in membership involves mainly American voluntary associations
and social movements. For some evidence, see Ziegler (1988, pp. 33-65).
10. This terrain is being explored under the label "generalized political exchange:' See
Marin (1990).
11. Or, as Marshall (1950, p. 11) put it, "the right to share to the full in the social heritage
and to live the life of a civilized being according to the standards prevailing in the society:'
12. This development may be similar to what Dahrendorf (1974, pp. 693-97) refers to as
"sectoral citizenship."
13. Although as Michael Walzer (1970, pp. 77-98) notes, citizens may still be compelled
"to live for the state" because national laws typically prohibit and punish suicide. To which
I would add that the current evolution of American legal doctrine and public policy seems to
be aimed at compelling its citizens "to procreate for the state" by outlawing abortion.
14. Although the recent controversy in the United States over the burning of the flag
suggests that the issue is far from over-at least in this most patriotic of advanced industrial
15. Anyone who doubts this or believes that it has completely disappeared should watch
the hilarious movie, Die Schweizermacher. which chronicles the steps that foreigners must
take to acquire Swiss citizenship.
16. Which may be what is emerging at the level of the European Community, although
the process has been hesitant and uneven. For a recent summary of the evolution of public
attitudes to the EC using data from the Eurobarometre surveys, see Worchester (1990).
17. For a recent work that criticizes liberal democratic practices and stresses the need for
enhancing individual participation, see Barber (1984). Jane Mansbridge (1983, 1991) has
placed emphasis on deliberation.
18. I know of no data on this subject, time-series or cross-sectional, and would appreciate
anyone's calling my attention to where I could find them. My impression, based on the places I have
lived in recent years, is that the proportion has tended to increase, but I would be the first to admit
that this is hardly a representative sample. Note that a rival explanation for this (unproven) fact
might be that more and more of the foreigners resident in advanced industrial societies believe that
they will eventually return (or be allowed to return) to their country of origin.
19. The more specific measures designed to enhance secondary citizenship through
changes in the system of associability are discussed in detail in Schmitter (1988). Critiques
to these proposals are discussed in Schmitter (1991).


Barber, Benjamin. 1984. Strong Democracy. Berkeley: University of California Press.

Black, Antony. 1984. Guilds and Civil Society in European Political Thought from the
Twelfth Century to the Present. London: Methuen.

Brubaker, Rogers. 1989. "Traditions of Nationhood and Politics of Citizenship." Social Science
Research Council (New York), States and Social Structures Newsletter 9(Winter):4-8.
Calhoun, John C. [1853]1943. A Disquisition on Government, edited by Richard K. Craller.
New York: P. Smith.
Channing, William Ellery. 1830. Remarks on the Disposition Which Now Prevails to Form
Associations, and to Accomplish All Objects by Organized Masses. London: Edward
Dahrendorf, Ralf. 1974. "Citizenship and Beyond: The Social Dynamics of an Idea." Social
Research 41(Winter):673-701.
Gustavson, Carl G. 1966. The Institutional Drive. Athens: Ohio University Press.
Hadari, Saguiv A. 1989. Theory in Practice. Tocqueville's New Theory ofPolitics. Stanford,
CA: Stanford University Press.
Heckscher, Gunnar. 1946. Staten och organisationera. Stockholm: Kooperative Forbundets
Hirsch, Fred. 1978. Social Limits to Growth. Cambridge, MA: Harvard University Press.
Hirschman, Albert O. 1970. Exit, Voice and Loyalty: Responses to Decline in Firms.
Organizations and States. Cambridge, MA: Harvard University Press.
- - - . 1982. Shifting Involvements: Private Interests and Public Action. Princeton, NJ:
Princeton University Press.
Lipset, S. M., et al. 1956. Union Democracy. Glencoe, IL: Free Press.
Luthardt, Wolfgang. n.d. " 'Collective Democracy' during the Weimar Republic. A Semi-
Corporative Conception for Societal Depoliticization?," unpublished MSS, Depart-
ment of Political Science, Free University of Berlin.
Mansbridge, Jane. 1983. Beyond Adversary Democracy. Chicago: University of Chicago Press.
- - - . 1991. "A Deliberative Theory of Interest Representation." Paper presented at the
Conference on Competing Theories of Post-Liberal Democracy, University of Texas,
Austin, February.
Marin, Bernd, ed. 1990. Generalized Political Exchange. Antagonistic Cooperation and
Integrated Policy Circuits. Boulder, CO: Campus/Westview.
Marshall, T. H. 1950. Citizenship and Social Class. Cambridge: Cambridge University Press.
Michels, Robert. 1962. Political Parties: A Sociological Study of the Oligarchical Tenden-
cies of Modem Democracy. New York: Collier Books.
Nisbet, Robert. 1974. "Citizenship: Two Traditions." Social Research 41(4):612-17.
Olsen, Johan P. 1983. Organized Democracy. Bergen, Oslo, Tromso: Universitetsforlaget,
Olson, Mancur. 1968. The Logic of Collective Action. New York: Schocken Books.
Presthus, Robert. 1962. The Organizational Society. New York: Knopf.
Salisbury, Robert. 1984. "Interest Representation: The Dominance of Institutions.' American
Political Science Review 78:64-76.
Schlozman, Kay Lehman and John T. Tierney. 1986. Organized Interests and American
Democracy. New York: Harper & Row.
Schlozman, Kay Lehman and Sidney Verba. 1979. Injury to Insult. Cambridge, MA: Harvard
University Press.
Schmitter, Philippe C. 1975. Corporatism and Public Policy in Authoritarian Portugal.
Beverly Hills, CA: Sage.
- - - . 1988. "Corporative Democracy: Oxymoronic? Just Plain Moronic? or a Promising
Way out of the Present Impasse?" Paper presented at the Colloquium on "Politische
Institutionen und Interessenvermittlung," Universitat Konstanz, April.
- - - . 1991. "Some Second Thoughts about Corporative Democracy: Oxymoronic or
Moronic, Promising or Problematic?" Paper presented at the Conference on "Compet-
ing Theories of Post-Liberal Democracy," University of Texas, Austin, February.
Philippe C. Schmitter 163

Schmitter, Philippe C. and Donald Brand. 1979. "Organizing Capitalists in the United States:
The Advantages and Disadvantages of Exceptionalism." Paper presented at the Amer-
ican Political Science Association Meetings, Chicago.
Steiner, Jurg. 1974. Amicable Agreement versus Majority Rule: Conflict Resolution in
Switzerland. Chapel Hill: University of North Carolina Press.
Tocqueville, Alexis de. [1835 and 1840] 1969. De la Democratie en Amerique [Democracy
in America]. 2 vols. Garden City, NY: Doubleday.
Truman, David B. 1962. The Governmental Process. New York: Knopf.
Walzer, Michael. 1970. Obligations. Essays on Disobedience. War, and Citizenship. Cam-
bridge, MA: Harvard University Press.
Worchester, Robert M. 1990. "European Attitudes to the European Community and to 1992."
International Journal of Public Opinion 2(3):227-48.
Ziegler, Harmon. 1988. Pluralism, Corporatism and Confucianism. Philadelphia: Temple
University Press.
Networks as Political Glue:
Explaining Public Policy-Making

David Knoke

Network concepts and principles offer a perceptive theoretical framework

with which to explain public policy phenomena. By gluing together sev-
eral levels of analysis-personal, organizational, systemic-the network
approach gives a comprehensive account of political activity and its
consequences that surpasses other more piecemeal explanations. Much of
this approach is based on theoretical and empirical work by Ed Laumann,
Franz Pappi, myself, and our colleagues over the past two decades (e.g.,
Laumann and Pappi 1976; Laumann and Marsden 1979; Knoke and Laumann
1982; Laumann, Knoke, and Kim 1985; Laumann and Knoke 1987; Knoke
1990c; Knoke forthcoming; Pappi and Knoke 1992; Knoke, Pappi, Broad-
bent, Kaufman, and Tsujinaka forthcoming). Here I provide a synthesis of
the conceptual components of the network approach to public policy-mak-
ing, illustrated with examples from recent empirical studies, and propose
an agenda for further development.


The central premise of the network perspective is its conceptualization

of social structure as an enduring pattern of relationships among social

AUTHOR'S NOTE: Support from the National Science Foundation for a series of projects
that led to the development ofthe conceptual model reported herein is gratefully acknowledged.
An earlier version of this chapter was presented at a thematic session "Changing Patterns
of Political Participation" at the American Sociological Association meetings in Washington,
D.C., August 1990. Comments on the session by Paul Burstein were especially appreciated.

David Knoke 165

actors (Laumann and Pappi 1976; Fararo and Skvoretz 1986). The basic
elements are:
(1) Social Actors: These units may be individuals, groups, formal organi-
zations, or other collectivities (gangs, classes, work teams, etc.). Only
selected dimensions of an actor are usually relevant for a particular
network. For example, the board of directors and the governmental affairs
office of a corporation, not its production and sales staffs, are the critical
participants for a study of campaign contributions and lobbying networks.
(2) Relationships: These connections result from interactions between
social actors, in which either some physical commodity or symbolic value
is exchanged, or a joint activity is undertaken. Thus elected officials and
their constituents exchange information, votes, money, and services; pub-
lic interest groups collaborate on a lobbying effort; scientific experts
provide evidence at congressional hearings; a labor union offers its meet-
ing facilities to a civil rights organization.
The pattern of linkages and nonlinkages between pairs of actors jointly
identify the basic positions within a social system and its global structure.
Opportunities and constraints on the flow of information and ideas, the
performance of tasks and the achievement of goals, the accumulation and
dissipation of resources-all depend much more on the network's struc-
ture than on the specific attributes of the actors embedded in the network.
Although both actors and relationships can be characterized according to
their various social attributes (such as actors' genders and ages, or their
relationships' intensities and durations), such classifications are largely
irrelevant to the network concept of social structure.
The operating part of any network is that combination of actors and their
relationships that jointly defines the distinct positions in a social structure
and the manner in which these positions interact with one another (White
et al. 1976; Burt 1976). Although actors who share similar attributes may
tend to behave similarly, we can understand this convergence only by
showing how they jointly occupy equivalent positions vis-a-vis their
relationships with actors in other network positions. Thus, if we seek to
explain how workers in a typographic shop decide for whom to vote in a
union election, we will do better by investigating how their friendship,
antagonism, and authority ties expose them to reinforcing or conflicting
political persuasion than to expect that their ethnicities, ages, or job tasks
alone can account for their choices (Upset et al. 1956). The network
perspective is committed to the idea that the dynamic connections, link-
ages, and bonds among actors in social positions capture a more accurate
image of the causal process in a social system than do covariations among
the actors' attributes. Nowhere is this claim more evident than in the study
of political power and public policy-making.


Political power is an aspect of the actual or potential interaction between

two or more social actors in which one exerts greater control over the
other's behavior (Emerson 1962, p. 32). Many typologies classify the
varieties of power (e.g., French and Raven 1959; Wrong 1979). But, in my
recent review of political networks (Knoke 1990c), I argued that an
essential dichotomy existed between influence and domination as funda-
mental analytic types of power relations within a network frame.
Influence requires that one actor intentionally transmit information to
another that alters the latter's actions from what would have occurred
without that information (Parsons 1963). For influence to occur, a com-
munication channel between potential influencer and influencee must
exist and the recipient must believe the transmitted information to be
trustworthy. Thus influence operates primarily by persuasion: by an ex-
change of information that changes an actor's perception of the connection
between a decision and its consequences. For example, a business lobbyist
gives a legislator data that purports to show how raising the minimum
wage will reduce employment of retail workers in her district. Depending
on how credible the legislator finds the lobbyist's information, her support
for the minimum wage bill may change or be reinforced. The lobbyist's
influence stems neither from coercion nor from the use of resources
extrinsic to the substantive policy issue (i.e., a campaign donation or
bribe), but from the manipulation of information that changes the legislator's
beliefs about the specific situation on which her vote is cast.
Domination, the second fundamental form of power, is a power relation-
ship in which one actor controls the behavior of another actor by offering
or withholding some benefit or harm, that is, a sanction that may be a
reward or a penalty. Again, both a connection between actors and a
credible belief in the dominator's capacity to deliver on the promise or the
threat are essential. An asymmetry exists: a positive sanction must be
exchanged to assure compliance with commands, but threats need not be
carried out as long as compliance is forthcoming (Oliver 1980; Laumann
and Knoke 1987, pp. 153-62). For example, a building contractors' asso-
ciation may secure a Congressperson's support for a construction bill by
promising a campaign donation, or by threatening to support the member's
election opponent. Successful domination (i.e., intimidation) in the first
instance requires that the association's money must actually be spent, but
in the latter case it simply stays in the association's bank account as a
resource for future use. In its most extreme form, domination may persist
only through continually escalating applications of coercion, negative
sanctions such as force, torture, and violence that may eventually lose their
David Knoke 167

effectiveness in gaining compliance, as Romania's Nikolae Ceaucescu

was only the latest tyrant to discover.
When influence and domination occur simultaneously within a relation-
ship between two actors, they take the form of authoritative power. Both
the exchange of persuasive information and the exchange of (usually
positive) resources characterize an authoritative transaction, as when a
bureaucrat allocates funds to his subordinates with instructions for imple-
menting a program. Authoritative commands carry strong expectations by
both the issuer and the recipient that obedience will be uncontested
(Wrong 1979, p. 35). The source of the order rather than its substantive
content assures compliance. An important instance of authority is Weber's
"legitimate power" (Weber 1947, p. 325). Subordinate participants strongly
believe that their superior's commands are appropriate and acceptable, to
the extent that latent sanctions seldom need to be invoked to assure their
compliance with orders. Rather, participants obey leaders' directions
because they believe it is "the right thing to do," regardless of the potential
rewards or penalties for compliance. Thus, President Bush's orders send-
ing troops to the Persian Gulf and into combat to free Kuwait from the
Iraqis was obeyed by the generals and soldiers because they implicitly
believed his command was authorized by the Constitution, the Congress,
and the United Nations. Of course, disobedience would have resulted in
negative sanctions against the offenders (courts-martial), but the fear of
punishment is seldom necessary to compel obedience in an authoritative
power relationship. Rather, assent rests primarily on the joint legitimacy
beliefs held by command-givers and command-takers, with the latent
potential for negative domination evoked only in rare circumstances.
The analytic influence and domination relations among political actors
are useful concepts in a network explanation of public policy-making. The
next section discusses how they can be used to construct a generic concep-
tual model.


The generic policy-making model proposed in this section is applicable

to any social system consisting of a plurality of social actors whose actions
are oriented toward a set of collectively binding decisions. Thus it should
apply equally to a college department making decisions about hiring,
curriculum, and administration, and to an international system of nation-
states arranging trade, military, and diplomatic ties. For many years,
Edward O. Laumann and I have been calling these collective decision
arenas policy domains, formally defined by "specifying a substantively

defined criterion of mutual relevance or common orientation among a set

of consequential actors concerned with formulating, advocating, and se-
lecting courses of action that are intended to resolve the delimited substan-
tive problems in question" (Knoke and Laumann 1982, p. 256). In addition
to this actor interest dimension, Burstein (1991) argued that each policy
domain develops a logically coherent substantive or functional basis for
framing its policies, and that its participants usually construct a common
culture about how society does and should work.
In setting the analytic boundaries around any policy domain, the pri-
mary questions are (1) whether an actor's interests and activities must be
taken into account by other participants (i.e., whether an actor can exert
influence or domination); and (2) whether the outcome of a decision about
a substantive problem will have a perceptible impact on any of the
participants? If the answers to both questions are "no," than the actors
and/or problems can safely be ignored as irrelevant to the domain. Of
course, the empirical boundaries of any policy domain may be difficult to
draw in practice. Hence, a certain fuzziness must be tolerated. As a
practical matter, policy researchers should draw generous domain bound-
aries that include marginal actors, rather than draw narrow boundaries that
risk excluding significant participants.
Assuming that a policy domain is identifiable, the central concepts
comprising the generic model are actors, relations, actions, and events as
listed in Table 9.1. Actors and their political relations have already been
discussed. An action is a behavior intended by one actor (or a combination
of actors working together) to change the power relationships in a domain,
with the aim of securing a more favorable outcome on some event of
interest to the actor(s). The major types of actions that actors or combina-
tions of actors might take are mobilization (acquiring control over re-
sources held by other actors to whom they are connected, to be used to
pursue policy-making objectives); publicity (expressing preferences to
audiences that include both public officials and others); and lobbying
(efforts to convince public authorities to make favorable decisions). In
each type of activity, a policy domain actor attempts to change the
attitudes and/or behaviors of other participants so that the probability
increases that the policy decision preferred by the actor is realized. Some
of these actions are aimed at the supporters or potential supporters of the
proponents, whereas others are targeted toward neutral observers and
against active opponents (more often to dissuade them from action than to
convert them to one's side). But the most important targets in any policy
domain are the public officials who possess the legal authority to make a
decision about an event that is binding on the domain as a whole.
David Knoke 169

Table 9.1 Components of a Generic Policy-Making Model

Analytic Elements Examples

Individual Persons Voters; employees; members; politicians; experts
Interest Groups Unions; trade groups; PIGs; corporations; bureaucracies
Peak Associations Federations; chambers; quasi-official advisory bodies
Government Institutions Legislatures; courts; executive departments; regulators
Influence Policy communications; ideas, data, strategies, advice
Domination Resource exchanges; funds, facilities, votes, coercion
Mobilization Social movements; grass-roots protests; donations
Publicity Mass media blitzes; targeted mailings
Lobbying Contacts with government officials; legal suits
Program Initiatives Agency formation; personnel reassignments
Allocations Cut a budget; redistribute expenditures
Rulings New regulatory standards; court judgments

An event is a concrete proposal for solving a domain problem or issue

to be decided by a policy-making body (Laumann and Knoke 1987, p. 17).
Thus an event occurs at a specific time and place and its outcome, whether
a success or a failure from the viewpoint of any of its advocates, marks a
definitive collective judgment either to change or to continue an existing
set of practices. In our tripartite federal government, policy events are
typically designated as judicial, executive, or legislative according to the
functional setting into which they fall. Table 9.1 classifies events accord-
ing to the kinds of actions that are taken to implement them. Program
initiatives result in some fundamentally new undertaking, either to be
conducted by an existing entity or by a specially created agency (perhaps
by splintering or agglomerating some existing organizations). A recent
graphic example of program initiation was the attempt to integrate federal
agencies fighting a drug war under the command of "Czar" William
Bennett. Allocations change the annual budget of an existing agency, and
may reflect a decision either to expand or contract some of its operations
while leaving an agency's mandate basically unaltered. The expected
cutbacks in Pentagon expenditures with the end of the Cold War illustrates
this type of policy action. Finally, rulings are substantial changes in the
"rules of the game" that govern interpretations and implementations of
existing policies. Rulings may expand the scope of a program's coverage,
exempt certain categories, or redefine the purposes to be served. The

Supreme Court's decisions on racial and gender discrimination cases

greatly modified the hiring and promotions procedures of businesses,
colleges, and government bureaucracies.
The concepts of policy actors, power relations, political actions, and
policy events suffice to construct a generic policy-making model for a
policy domain. The basic proposition of the network approach is that
policy domain actors attempt to use their relations with other actors to
bring sufficient influence and domination resources to bear on those public
authorities charged with deciding a policy event, with the intention of
increasing the probability that these authorities will produce an outcome
favorable to the actors' interests. Collaboration among actors who pool
their power resources is generally more effective in realizing their com-
mon interests than are efforts by actors attempting influence and domina-
tion on their own. Because a plurality of actors with strong interests in a
given policy event typically pursue their preferences on every policy
event, the configurations of allied and opposed actors may be quite fluid
and complex when examined over time. Further, because domain events
are seldom unique isolated occurrences but reflect numerous decision
points in a continuing policy stream, the interweaving of actors, relations,
and actions is best considered across multiple sequential policy events.
These contingent temporal patterns compel analytic attention to a large
array of domain collective events that reveal the stability and change in
social network structure as well as the impact of these event outcomes on
subsequent policy struggles.
An actors's interests in a series of domain events vary according to the
stakes it believes to be at risk in the outcome of each specific collective
decision. Whether a given actor attempts to affect the outcome of a specific
event depends in large measure on the degree to which that event engages its
interest. When perceived interest in an event is low, or when other more
intrinsically interesting events crowd the domain agenda, an actor is unlikely
to commit its scarce political resources to the fight. As Browne (1990) pointed
out, many organizations cultivate such specialized identities that only obses-
sively narrow and intensely directed facets of events carry any appeal. But if
an actor believes an event carries sufficient interest, or it is obliged to assist
other actors, or it can exchange its interests in that event for control over other
events of greater interest (Coleman 1990), then the actor is likely to collabo-
rate in collective action policy-making efforts.
In this conceptual model, the most significant social structural forma-
tions are jointly occupied positions, a subset of domain actors defined by
their interests in an event and their relationships to one another. Three
types of jointly occupied positions are likely to occur for many policy
events. Their identification involves increasingly stringent criteria, with
David Knoke 171

the result that the typical number of actors occupying each type of position
is successively smaller. In other words, as an event moves toward resolu-
tion, the more marginal actors fall by the wayside.
(1) An event public is the set of all domain actors that hold an interest
in an event, no matter how slight. This interest is an overtly expressed
concern, not a latent orientation to be imputed from an external analyst's
knowledge of social conditions (e.g., as in a Marxist allegation of the
working class's interest in socialism). In the case of actors consisting of
many persons, such as corporations or interest groups, not all the group's
members need express the interest. An interest held only by an organization's
leadership suffices to engage the collectivity's involvement in the event.
Event publics are heterogeneous audiences with regard to the policy
outcomes they prefer. Both proponents and opponents as well as neutral
observers who merely monitor the scene may all be involved in a specific
public. Thus a public is the broadest possible set of actors from which the
more intensely committed participants may be recruited. Note that mem-
bers of an event public are not required to maintain direct or indirect
communication ties to one another. Indeed, their primary modes of atten-
tion to the substance of the event may be via mass media channels.
(2) A collective actor consists of three or more formal organizations within
an event public that communicate directly or indirectly among themselves
about policy matters and that advocate the same outcome preference on that
event (Laumann and Marsden 1979). That is, the members of one collective
actor all support the "pro" outcome on that event, whereas the members of
another collective actor all prefer the "con" outcome for that decision. For a
subsequent event, other subsets of the event public constitute the distinct
collective actors. The members of collective actors for different events may
only partially overlap with one another. Indeed, the degree of coincidence
among collective actors across numerous events is important evidence about
the cleavage structure prevailing in a policy domain. The collective actor
concept still does not distinguish between those organizations that work
closely together to influence a concrete policy decision and those that are
either passive policy proponents or go their own way in trying to produce their
preferred outcome.
(3) The action set concept further restricts its scope to the members of
a collective actor who consciously coordinate their policy influence activ-
ities (Knoke and Pappi 1991). An action set's members work together to
influence the outcome of a specific policy event. Group cohesion is the
essential feature: All members of an action set prefer the same policy
outcome for the event; are directly or indirectly linked together in communi-
cation or other networks; and overtly coordinate their lobbying and other
policy influence activities (Knoke and Burleigh 1989). If communication

channels are weak and disjointed, more than one action set may coalesce
that involve different sets of organizations, each favoring the same policy
event outcome, as shown in Figure 9.1. More typically, a single action set
emerges on each of the two opposing sides in a contested event (one pro-
and one con-action set). Aldrich's definition of an action set is closely
related: "a group of organizations formed into temporary alliance for a
limited purpose," but he was not explicit about the alliance requirements
(Aldrich 1979, p. 280). An action set may have formalized agreements, an
internal division of labor, behavioral norms vis-a-vis other organizations,
and clearly defined principles for recruiting new members (Aldrich 1979,
p. 281). Thus action sets are short-term arrangements, most formed only
for single-event collaborations and disbanding after success or failure.
Because a collective actor and an action set are social formations unique
to a single event, they cannot reveal the global structures of a policy
domain. During any political era, many policy fights take place and each
event stimulates the formation of distinct collective actors and action sets
on opposing sides. An important question for a policy domain analyst is
the overarching opposition structure among these multiple events. For
example, organizations A and B may collaborate in the same pro-action
set for event W; belong to different con-action sets on event X; work in
opposing action sets on event Y; and sit out event Z entirely. Similar
diverse patterns apply to collective actors formed around each event.
When dozens or hundreds of other organizations' involvements in every
event must also be taken into account, measuring the global collective
actor and action set structures of the domain becomes critical. To uncover
these deep social structures, analysts can take advantage of the duality of
organization-by-event connections: some organizations participate in the
same events, and some events attract the same organizations. Mapping
these organizations-by-events co-occurrences reveals the entire network
of ties among opposing collective actors and action sets.
More formally, an opposition network is the pattern of overlapping mem-
berships among all collective actors or among all action sets that form around
a set of policy events during some period within a specific domain. The
structure of a domain's opposition network is a function of the degree to which
its collective actors' or action sets' members coincide. If virtually the same
jointly occupied positions reassemble for event after event, then the domain
network structure will tend toward bifurcated blocks, one block taking the pro
and the other block taking the con position. In a spatial representation, such
polarized opposition networks will coalesce around two distant positions.
But, if the collective actors or action sets tend to recruit different members
for each unique event, their infrequent overlaps will result in pro and con
groups that are scattered and interspersed throughout the domain's topo-
David Knoke 173

graphical space. Central, peripheral, connective, and other spatial positions

may be occupied by varied collective actors or action sets (see Laumann and
Knoke [1987, p. 131], Knoke and Burleigh [1989], and Knoke [1990c] for
suggestive analytic diagrams; and Heinz et a1. [1990], and Knoke and Pappi
[1991] for empirical patterns in concrete policy domains).
A policy domain's opposition network structure has important implica-
tions for the resolution of domain policy conflicts. Polarized structures
indicate bitter divisions whose great gulfs cannot be easily bridged, be-
cause few or no members exist in common that could serve as mediators
of disputes. More diffuse configurations suggest a system's latent capacity
to bargain and negotiate a compromise solution to an event conflict. The
actors who reappear in overlapping memberships suggest the potential
solution brokers. The few empirical analyses of opposition networks have
generally found patterns that imply linkages between proponents and oppo-
nents across events, rather than extremely polarized structures (Laumann,
Marsden, and Galaskiewicz 1977; Laumann and Marsden 1979). One
analysis of government officials' perceptions of conflict in four policy
domains (energy, health, agriculture, and labor) concluded that "the more
prominent and unified the peak associations in a given domain, the more
polarized its group structure and policy struggles become" (Salisbury et
a1. 1987, p. 1229). However, these analyses were conducted at such a high
level of aggregation that variation in group composition across concrete
policy events could not be examined.
Knoke and Pappi's (1991) comparison of dozens of legislative policy
events in the U.S. and German national labor policy domains found pairs
of opposing collective actors and several smaller action sets in the large
majority of events. Business associations and labor unions organized and
led most of these social formations in both nations. The opposition net-
work configurations for collective actors and action sets were rim struc-
tures with distinct hemispheres occupied by opposing business and labor
organizations. No collective actor or action set occupied the center of the
labor domain policy space, a finding similar to Heinz et aI.' s (1990)
discovery of "hollow cores" among the networks of elite representatives
in four American national policy domains. This absence of any overarch-
ing authority able to orchestrate mutually acceptable resolutions implies
that adversarial polarization predominates across numerous labor policy
conflicts. Organizational interests were the main factor driving action set
formation and organizational participation, with influence reputations and
resource holdings also important. Action set members were significantly
more likely to use a variety of actions in attempting to influence the
outcomes of events than were organizations working alone, especially in
the United States.


The handful of empirical studies using various elements of the concep-

tual policy-making model suggest that policy-making involves a progres-
sive winnowing of both network positions and policy alternatives as an
event moves over time toward a definitive resolution. The metaphor of a
funnel is apt: successively fewer mobilized positions remain to carryon
the policy fight as the outflow nears. Figure 9.1 depicts this funnel of
mobilization in terms of the relative proportions of issue publics, collec-
tive actors, and action sets that are activated as an event progresses from
initiation to collective decision. At the broad inflow end of the funnel are
numerous individual persons in their social statuses (such as employees,
consumers, professionals) and informal collectivities (such as classes,
groups, communities). Although many of these actors may have latent
structural interests or even consciously held concerns about a particular
policy event (e.g., automobile assembly workers' worries about import
policies), most of them lack the resources, including time, for effective
participation. Perhaps only a small minority comprise a meaningful issue
public for any particular event; the vast majority are simply disinterested
and hence unmobilizable.
Still near the funnel's wide end, organized interest groups are much
more likely to perceive that their interests are at stake in a policy event.
Many, however, will either have narrow policy mandates, charters that
proscribe overtly partisan activity, or insufficient political resources to
become involved in every policy controversy. Perhaps half or more of
them remain passive, but most of those with clear policy preferences are
likely to choose up sides as the events' alternatives become clearly de-
fined. Hence, membership in a collective actor is fairly common among
organized interest groups. However, relatively few will collaborate with
others as action set partners. The next segment of the funnel consists of
the "peak" interest associations of the domain. Some are mass membership
organizations such as the Sierra Club and the American Medical Associ-
ation, whereas others are organizations with institutional members. like
the AFL-CIO and the Edison Electric Institute. Peak organizations typi-
cally have large and wealthy memberships who give their professional
staffs broad latitude to fill up a full plate from the domain's policy menu.
Comparatively few in numbers, these highly visible entities often cannot
avoid being sucked into many policy fights. Their members and constitu-
ents expect them to enact their commitments to a core agenda, whereas
their partners from previous policy fights expect them to fulfill their
obligations to participate reciprocally when IOUs are called in. Hence, a
substantial proportion of a domain's peak association population belong
David Knoke 175



Interest Groups



Figure 9.1. The Funnel of Mobilization for Policy Domains

to action sets. Finally, the ultimate policy decision makers in the relevant
institutional arenas-such as the Congress or regulatory agencies-enter
the fray at the funnel's narrow end. Almost all these actors are mobilized,
as a collective decision is not possible without definitive action on their

part. Hence, the large majority of these institutional participants are mobilized
into one of the pro or con action sets where they collaborate with the key peak
associations and interest groups preferring the same policy outcome.
The funnel metaphor implies a temporal procession from earlier activities
that involve broad publics to later actions that are settled among a small
number of key decision makers. Many real policy events are unlikely to
conform to so tidy a sequence. Institutional actors, such as the President or
regulatory agency staff, are as likely to initiate policy proposals as to be
reactors to events raised by pressure groups. Peak associations may activate
their grass-roots members to lobby directly with their Congresspersons.
Passionate interest groups may bolt from their peak associations to pursue an
idiosyncratic agenda (e.g., the Chemical Manufacturers Association's sup-
port of a 1988 workplace safety bill opposed by the Chamber of Commerce).
Despite such chaotic thrusts and parries during the heat of policy battles, the
conceptual simplicity of the funnel is still helpful in highlighting a persistent
pattern: as a policy proposal moves ever closer toward concrete resolution,
the subset of domain actors with sufficient power to influence or dominate
the final decision becomes smaller rather than larger. Ordinary voters and
association members may express general preferences toward one side or
another of a proposal, but they rarely sit down with congressional committee
staff to design sections of a bill. Organized interest groups may prowl the
corridors of a regulatory agency, but when the rate structures are being
marked up only the lobbyists for peak associations are likely to get their phone
calls returned. The mobilization funnel captures the sense that policy-making
is an exclusionary process-locking out both actors and alternatives-rather
than inclusionary.
The eventual success or failure of a policy event fight depends on which
side of the controversy can bring enough influence and domination re-
sources to overcome the advantages on the other side. Such resources are
gathered from every stage of the funnel. Public opinion and election votes
come from persons and collectivities; publicity and grass-roots lobbying
are contributed by interest groups; peak associations pour professional
fund-raisers and lobbyists into the offices of bureaucrats and politicians;
within the institutional arena itself, lawmakers and regulators twist arms
and trade support for their pet projects.


The network approach to public policy-making can be applied to a

variety of traditional research questions. This section sketches its applica-
tion to three important situations, showing where it illuminates key issues
David Knoke 177

and where further concerted efforts to operationalize the model's elements

could yield substantial payoffs.

Voting Decisions. The first question is how individual voters' policy

interests affect their voting decisions. For several decades the Michigan
political science model dominated conceptualization and research on voting
behavior (Rusk 1982). Consequently survey research concentrated on mea-
suring voters' political party identifications, their preferences about the major
campaign issues, and their perceptions of how parties and candidates were
likely to deal with those policy issues of concern to the voter (e.g., Campbell
et a1. 1960). Generally ignored in this perspective were the political conse-
quences of the interpersonal social networks in which voters are embedded.
Despite the early demonstration by Columbia social scientists that such
primary group influences constrain voters' perceptions and decisions
(Lazarsfeld et a1. 1948), this promising avenue was bypassed for decades
(Sheingold 1973). Even political campaign advisors buy into the image of the
atomized voter, relying extensively on public opinion polls and focus groups
to provide clues on how to pitch their campaign appeals.
Academics and practitioners alike overlook the ways that a voter's personal
networks shape the cognitive schemes essential for interpreting and acting as
political beings in larger policy domains. A recent revival of the network
approach to political choice demonstrated that voters' egocentric networks of
important discussion partners have a significant impact on political actions
that cannot be reduced to personal social attributes (Knoke 1990a). In partic-
ular, the political partisanship of one's confidants and the frequency with
which one discusses political matters with them both affect such political
activities as turning out to vote, attending campaign rallies, working for
candidates, donating money, and trying to persuade others to vote.
Remaining to be investigated is how the substantive contents of such
micronetworks integrate voters into the public policy domains in which
key policymakers operate. We still know very little about how communi-
cation within a mass public and between the public and political elites
propagates or retards the spread of political information in all directions.
The old notion of a two-step flow-where elites inform local opinion-lead-
ers who in turn interpret the message in the language of their adherents
(Weimann 1982)-is surely too simplistic. Contemporary political dis-
course takes place through so many specialized channels, bridges so many
narrow audiences, and conveys so many cross-cutting messages that only
a network model that can separate these multiplex strands offers much
chance of faithfully dissecting reality. We need to learn whether a cohesive
position formed by intense emotional ties among voters tends to cut off its
members from new policy information, miring them in isolated pockets of

conservative resistance to political innovation. Conversely, do weak ties

really knit together fragmented positions into a genuine political commu-
nity whose members are accessible to influence by parties and politicians,
and, in turn, are capable of communicating the public's demands to these
policy elites? For example, Congress persons' most prevalent noncampaign
contacts with their districts occur in the casework troubleshooting they
perform on behalf of constituents aggrieved by federal bureaucracies (e.g.,
lost social security checks; delayed disaster aid; IRS bullying of small
businesses). Politicians believe this attention is vital to their reelection
prospects. Yet we have scant evidence that such individualized good
works ramify through the grass roots to generate socially constructed
loyalty, trust, and electoral support. Nor do we understand whether such
networks promote the mutual mobilization of voters and officials who
each seek to press their policy preferences on the other.
To restore the network component to its proper place in voting research,
political sociologists must insist on including in surveys more detailed
information about voters' policy discussions with their social intimates.
The specific contents of these communications must be measured and their
contributions to political persuasion gathered directly from all participants
in local networks. Bridging connections between voters and public policy
officials, both direct and indirect via party activists and interest groups,
must be recovered. Only when we carefully map the multiple mutual ties
within an electorate will we be in a position to determine whether the
promise of network analysis to elucidate the social foundations of voting
behavior has been kept.

Collective Action Organizations. The second question is how mass

membership organizations arrive at their collective decisions about which
public policy events to pursue. Peak associations and special-interest
groups are major types of policy actors in the funnel of mobilization. What
is less well recognized is that these associations are themselves significant
policy domains for their members. Grant McConnell drew parallels be-
tween the national state and the polities of private interest associations:

Certainly the state and private association are different, but the differences are
hard to define. Both are to some extent based on territorial lines.... Both state
and associate exercise some form of authority. Each makes rules or laws ... en-
force with penalties of varying degrees of severity, but some organizations'
penalties are more severe than many punishments of the state .... Of course not
all private associations exist in spheres of interest with such degrees of impor-
tance. Yet some do and where they do the problem of authority is substantially
the same as it is in the state. (McConnell 1966, p. 125)
David Knoke 179

A majority of American national associations take positions on some

national policy events that they communicate to a branch of the federal
government (Knoke 1990b, p. 208). How these organizations arrive at
their collective decisions and how they make these decisions binding on
their members is not well understood.
Analysis must begin with a recognition that policy events are likely to
provoke little internal dissension in associations that attract fairly homo-
geneous memberships. Given an organization's stated purposes, few per-
sons or organizations who are opposed to those interests are likely to join.
Thus an association's leaders usually can act with little fear that internal
opposition will erupt. For example, the Sierra Club routinely opposes
federal actions it believes will degrade the environment, such as offshore
oil leases. Prodevelopment people are unlikely to belong to the Sierra Club
in the first place, and thus to fight from inside the Club against its
environmental protection preferences. On occasion, however, an associa-
tion faces a controversial event that splits its members into opposing
camps that fight to control the policy positions taken by the organization.
Internal dis sensus is most likely to erupt within an organization that
controls the right of its members to practice a trade or profession, since
would-be practitioners have no alternative to joining. In a controversy, the
funnel of mobilization of association members into event publics, collec-
tive actors, and action sets within the organization should exactly parallel
the policy fights in the larger external policy domains between organiza-
tions. An association's elected leadership is the target of influence and
domination struggles by proponents and opponents trying to sway the
policy decision in the direction favored by each faction. The dominant
coalition's authority to run the association is challenged by insurgents
who, if they lose the fight for control, may ultimately exit to form a rival
organization. Factional disputes in social and religious movements are
notoriously replete with this scenario (Gamson 1975).
The network approach's contribution to understanding internal interest
group politics lies in its identification of communication as a central
mechanism for resolving internal conflicts. The exchanges of policy infor-
mation among leaders and members are heavily concentrated among
positions of organizational control and decision making (Katz and Kahn
1966, p. 223). But these patterns may vary with such factors as size,
bureaucratization, and external environments that generate pressures of
complexity and uncertainty on the processing and interpreting of informa-
tion. Consequently, the need for decentralized decision making may stim-
ulate the formation of autonomous positions in which oppositional inter-
ests can congregate, for example, in regional branches or occupational
specialties within large national organizations such as labor unions and

professional societies. One analysis of members of 35 national associa-

tions found that the volume of policy discussions was a strong predictor
of members' commitment to the organization and their participation in
internal and external activities (Knoke 1990b, pp. 163-86). Still unex-
amined is how communication structures channel association members
with interests in a public policy controversy into opposing coalitions.
To conduct such research would require a series of intensive case studies
of routine and controversial policy events in public interest groups. Mem-
bers would be asked to identify those officers and members with whom
they discussed these events, their preferred policy positions, and their
collaborative efforts to influence the association's collective decision.
Then the mobilization of specific event publics, collective actors, and
action sets inside the organization, and the emergence of an opposition
network across events could be uncovered and related to the ultimate
policy positions taken by the organization. The funnel of mobilization
within an organization differs from that of a larger policy domain by the
absence of formal interest groups and peak associations. But the general
process of successively winnowing the participants with influence over
the binding collective decision should be substantiated. An important
result of such research would be to explain the conditions under which a
policy fight leads to defection of significant numbers of supporters on the
losing side of the controversy.

National Policy Domains. The third question concerns some unexplored

aspects of national policy domain event decisions. I have already indicated
that much of the conceptual and empirical work on policy networks focuses
on this level of analysis. Nevertheless, the picture is far from complete.
Significant connections have not yet been drawn between elite policymakers
and mass event publics in both the mass electorate and the memberships of
interest associations. We have assumed that proximate policymakers in the
Congress, executive branch, and judiciary are most strongly influenced by
persuasion and sanctions from domain peak organizations. But our research
has really not considered whether or how these decision makers use policy
communications from constituents, either received directly or filtered through
various network connections. Nor have we included public opinion, mass
media, and social movement actions that impinge on policymakers. Conse-
quently, although the network policy-making model is elegant in its reliance
on a few basic concepts and relations, it suffers from failure to incorporate
the entire range of potential policy influence identified by policy analysts
from other traditions.
To consider only one glaring omission, we have paid no serious atten-
tion to expert circles as distinctive players in policy domains. An expert
David Knoke 181

circle consists of at least three communicating actors engaged in research

and evaluation relevant to a policy event. Unlike an event public, a circle's
members maintain direct or indirect communication ties that allow them
to exchange information on their common substantive concerns. Like an
event public, its members may not take an advocacy stance on the matter,
but they are capable of making informed assessments on the merits of a
policy proposal. Experts frequently are sources of research data, demon-
stration projects, performance evaluations, and consultations sought by
more partisan actors to bolster their contentions. Expert circles bear a close
resemblance to Shrum and Wuthnow's (1988) concept of a "technical
system" as a centrally administered network of actors oriented toward
solving technological problems. However, an expert circle is much broader
because it may include actors with primarily legal, financial, social, and
political as well as purely scientific and technical assessments of a policy
matter. Nor does it resemble Domhoffs (1987) concept of a "policy planning
network" that serves a national business power elite by formulating policies
to advance that class's interests. Rather than overt partisans of one stripe or
another, experts constitute a distinct social structure for developing and
transmitting ideas, arguments, and data that are used selectively by more
partisan advocacy groups. Their nonprofit status under 501 (c) (3) Internal
Revenue Service designation prevents many research organizations and think
tanks from taking official policy stands on partisan issues (Berry 1977, pp.
45-55). Some of their employees may act as policy advocates in event
controversies, but they speak as individuals rather than as representatives of
their employing organization. University professors and think tank employees
who advise the government about the scientific merits of public policies are
prime examples (e.g., Cronin and Thomas 1971; Nelkin 1975; Smith 1991).
A major contribution to policy-making research could be made by
investigating the social organization of the expert system and its articula-
tion with the advocacy and policy-making actors of a national domain.
How are expert reputations created? What conditions promote greater or
lesser communication integration among experts working in a domain?
How do advocacy organizations identify and secure advice from experts?
How can actors refute or neutralize experts' testimonies? Do experts lose
credibility when their judgments are challenged by other experts and found
wanting? Are experts co-opted by too-close ties to their advisees? At what
point do experts cross the line from disinterested commentators to policy
partisans, and to what consequences? And, does the involvement of ex-
perts have any discernible impact on the outcome and quality of collective
policy decisions? Answers to these questions are likely to differ across the
substantive domains where experts are engaged, from highly technical
fields such as nuclear energy to no-tech domains such as public housing.

Several other topics await future exploration with a network perspec-

tive: the micro-management of policy operations within the government
affairs offices of organizations; the politics of policy implementation
involving bureaucrats and claimant groups; the interplay of judicial and
regulatory bodies in the creation of new domains; the evolutionary trans-
formation of older domains under competitive pressures from interna-
tional economic and politico-military changes. There's enough here to
keep a generation of doctoral students in theses.


Public policy-making cannot be adequately understood without a thor-

ough attention to the structure of power relations that connect actors,
events, and decisions. The fundamental forms of political power-influ-
ence and domination-are inherently relational. Their explanation re-
quires explicitly relational concepts and research methodologies capable
of revealing the dynamic formation and reformation of coalitions across
extended sequences of policy-making activities. The conceptual frame-
work constructed for this purpose has already proven useful in illuminat-
ing several empirical analyses of national policy domains. The biggest
challenge ahead lies in integrating the network policy model across levels
of analysis ranging from the individual voter to social movements to
special interest groups to government policymakers. Exchanges of both
influence (persuasive information) and domination (inducements and sanc-
tions) must be analyzed within a dynamic perspective that allows research-
ers to trace their impact on alliance formation as policy events move
through the funnel from initiation to decision to implementation. As
network principles become routinely applied to policy-making research at
all levels of analysis, we shall arrive at a better assay of the glue that holds
the whole system together.


Aldrich, Howard E. 1979. Organizations and Environments. Englewood Cliffs, NJ: Prentice-
Berry, Jeffrey M. 1977. Lobbying for the People: The Political Behavior of Public Interest
Groups. Princeton, NJ: Princeton University Press.
Browne, William P. 1990. "Organized Interests and Their Issue Niches: A Search for
Pluralism in a Policy Domain." Journal of Politics 52:477-509.
Burstein, Paul. 1991. "Policy Domains: Organization, Culture, and Policy Outcomes."
Annual Review of Sociology 17:327-50.
David Knoke 183

Burt, Ronald S. 1976. "Positions in Networks." Social Forces 55:93-122.

Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. 1960. The
American Voter. New York: John Wiley.
Coleman, James S. 1990. Foundations ofSocial Theory. Cambridge, MA: Harvard University
Cronin, Thomas E. and Norman C. Thomas. 1971. "Federal Advisory Processes: Advice and
Discontent." Science 171:771-79.
Domhoff, G. William. 1987. "Where Do Government Experts Come From? The CEA and the
Policy-Planning Network." pp. 189-200 in Power Elites and Organizations, edited by
G. William Domhoff and Thomas R. Dye. Newbury Park, CA: Sage.
Emerson, Richard M. 1962. "Power-Dependence Relations." American Sociological Review
Fararo, Thomas J. and John Skvoretz. 1986. "E-state Structuralism: A Theoretical Method."
American Sociological Review 51:591-602.
French, John R. P., Jr. and Bertram Raven. 1959. "The Bases of Social Power." Pp. 150-67
in Studies in Social Power, edited by Darwin Cartwright. Ann Arbor, MI: Institute for
Social Research.
Gamson, William A. 1975. The Strategy ofSocial Protest. Homewood, IL: Dorsey Press.
Heinz, John P., Edward O. Laumann, Robert H. Salisbury, and Robert L. Nelson. 1990. "Inner
Circles or Hollow Cores? Elite Networks in National Policy Systems." Journal of
Politics 52:356-90.
Katz, Daniel and Robert L. Kahn. 1978. The Social Psychology of Organizations. Rev. ed.
New York: John Wiley.
Knoke, David. 19908. "Networks of Political Action: Toward Theory Construction." Social
Forces 68:1041-63.
- - - . 199Ob. Organizing for Collective Action: The Political Economies of Associations.
Hawthorne, NY: Aldine de Gruyter.
- - - . 199Oc. Political Networks: The Structural Perspective. New York: Cambridge
University Press.
- - - . Forthcoming. "Networks of Elite Structure and Decision Making." Sociological
Methods &: Research.
Knoke, David and Frank Burleigh. 1989. "Collective Action in National Policy Domains: Con-
straints, Cleavages, and Policy Outcomes." Research in Political Sociology 4:187-208.
Knoke, David and Edward O. Laumann. 1982. "The Social Organization of National Policy
Domains: An Exploration of Some Structural Hypotheses." Pp. 255-70 in Social
Structure and Network Analysis. edited by Peter V. Marsden and Nan Lin. Beverly
Hills, CA: Sage.
Knoke, David and Franz Urban Pappi. 1991. "Organizational Action Sets in the U.S. and
German Labor Policy Domains." American Sociological Review 56:509-23.
Knoke, David, Franz Urban Pappi, Jeffrey Broadbent, Naomi J. Kaufman, and Yutaka
Tsujinaka. 1992. "Issue Publics in the U.S., German, and Japanese Labor Policy
Domains" Research in Politics and Society 4:255-94.
Laumann, Edward O. and David Knoke. 1987. The Organizational State: A Perspective on
National Energy and Health Domains. Madison: University of Wisconsin Press.
Laumann, Edward 0., David Knoke, and Yong-Hak Kim. 1985. "An Organizational Ap-
proach to State Policy Formation: A Comparative Study of Energy and Health Do-
mains." American Sociological Review 50:1-19.
Laumann, Edward O. and Peter V. Marsden. 1979. "The Analysis of Oppositional Structures
in Political Elites: Identifying Collective Actors." American Sociological Review

Laumann, Edward 0., Peter V. Marsden, and Joseph Galaskiewicz. 1977. "Community
Influence Structures: Replication and Extension of a Network Approach." American
Journal of Sociology 31: 169-78.
Laumann, Edward O. and Franz Urban Pappi. 1976. Networks of Collective Action: A
Perspective on Community Influence Systems. New York: Academic Press.
Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet. 1948. The People's Choice: How
the Voter Makes Up His Mind in Presidential Campaigns. New York: Columbia
University Press.
Lipset, Seymour Martin, Martin Trow, and James S. Coleman. 1956. Union Democracy.
Garden City. NY: Doubleday.
McConnell, Grant. 1966. Private Power and American Democracy. New York: Vintage.
Nelkin, Dorothy. 1975. "The Political Impact of Technical Expertise." Social Studies of
Science 5:35-54.
Oliver, Pamela. 1980. "Rewards and Punishments as Selective Incentives for Collective
Action: Theoretical Investigations." American Journal of Sociology 85: 1356-75.
Pappi, Franz Urban and David Knoke. 1991. "Political Exchange in the German and Amer-
ican Labor Policy Domains." Pp. 179-208 in Policy Networks: Structural Analysis of
Public Policy, edited by Renate Mayntz and Bernd Marin. Baden-Baden: Nomus.
Parsons, Talcott. 1963. "On the Concept of Influence." Public Opinion Quarterly 27:37-62.
Rusk, Jerrold G. 1982. "The Michigan Election Studies: A Critical Evaluation." Micropolit-
ics 2:87-110.
Salisbury, Robert H., John P. Heinz, Edward O. Laumann, and Robert L. Nelson. 1987. "Who
Works with Whom? Interest Group Alliances and Opposition." American Political
Science Review 81:1217-34.
Sheingold, Carl A. 1973. "Social Networks and Voting: The Resurrection of a Research
Agenda." American Sociological Review 39:712-20.
Shrum, Wesley and Robert Wuthnow. 1988. "Reputational Status of Organizations in Tech-
nical Systems." American Journal of Sociology 93:882-912.
Smith, James A. 1991. The Idea Brokers: Think Tanks and the Rise of the New Policy Elite.
New York: Free Press.
Weber, Max, 1947. The Theory ofSocial and Economic Organization. New York: Free Press.
Weimann, Gabriel. 1982. "On the Importance of Marginality: One More Step into the
Two-step Flow of Communication." American Sociological Review 47:764-73.
White, Harrison C., Scott Boorman. and Ronald L. Breiger. 1976. "Social Structure from
Multiple Networks 1. Blockmodels of Roles and Positions." American Journal of
Sociology 81:730-80.
Wrong, Dennis, 1979. Power: Its Forms. Bases and Uses. Oxford: Basil Blackwell.

Financial Reorganization of
American Corporations in the 19808

Neil Fligstein
Linda Markowitz

One critical dimension missing from most theories of organizational change

is the role of culture in framing the possibility for strategic action. The
problem is that in order for actors to undertake new forms of action they
must decide what their interests are, develop a plan to operationalize those
interests, and have the power to enforce that view. Culture, defined
broadly as material practices, comes into play to provide actors with a
cognitive frame that offers solutions to the problem of strategic action
(Swidler 1986).
In the context of profit-making firms, the question can be put in a
different way: The issue is not that managers seek out profit, but how they
do so. It is to this issue that sociology and organizational theory can add
to economic views of how firms operate. Managerial actions are embed-
ded, constrained, and shaped by the cultural frames of actors and the
structure of interorganizational relations. The ways they will attempt to
make profit will, thus, reflect these frames and the relations to relevant
fields of action.
It is from this perspective that one can begin to view the financial
reorganization of the largest American corporations in the 1980s. This
reorganization has involved a number of related actions. Managers of
firms have divested productive product lines, bought others, repurchased

AUTHORS' NOTE: This chapter was prepared as a paper for a Thematic Session at the
Annual Meetings ofthe American Sociological Association held in Washington, D.C., August
1990. We would like to thank Jerry Davis for his comments on an earlier draft.


their stock, engaged in leveraged buyouts of their firms, and of course,

been the targets of takeovers as well. This chapter provides quantitative
models for which corporations were most likely to have engaged in
financial reorganization or have been the target of merger.
At the core of this explanation is the notion that these financial reorga-
nizations reflect the dominance of a cultural frame that views the essence
of the corporation in primarily financial terms. Such a perspective means
that actors will analyze their situations from the point of view of the
financial "facts" of the firm and undertake actions that reflect that evalu-
ation of the firm. There are two interesting questions to raise: Where did
this financial conception of the firm come from and which actors use it to
define and promote their interests?
There are three approaches that might shed some light on these issues.
Finance economics considers the conditions under which firms would
issue equity and debt and why the market for corporate control promotes
the overall efficiency of firms by forcing them to deploy their assets
properly (Fama and Jensen 1983a, 1983b; Fama 1980; for a review see
Lorie et al. 1985). Sociological theories of the merger movement suggest
two approaches. First, the extension of control over firms by banks and
institutional investors makes firms more vulnerable to financial manipu-
lation (Mintz and Schwartz 1985; Kotz 1978; Brancato and Gaughan
1988). Second, the dominance of finance CEOs in large firms, the spread
of the finance conception of the firm as indexed by competitors' behavior
and the product mix of the firm, and the lax antitrust policies and tax cuts
of the Reagan Administration have produced the conditions that have led
to the financial reorganization of firms (Fligstein 1987, 1990).
The results presented here show that the reorganization of financial
assets of the largest corporations can only be understood by considering
the finance conception of the firm. In firms where it dominates, managers
were more likely to engage in stock repurchases, mergers, divestitures, and
the purchase of stock of other firms. Where that conception did not
dominate, institutional investors were able to push managers toward those
actions. Thus actors who had the interests and power to enforce the finance
conception did so. Firms that were merger targets tended not to engage in
forms of financial reorganization. The presence of bankers on their boards
of directors made those firms more likely to be merger candidates. This
suggests that they were lenders who were brought on the board to protect
their loans. There was no evidence that the financial "facts" of the firm
promoted reorganization. This implies that without actors to interpret
financial information and act upon it, financial reorganization did not
Neil Fligstein and Linda Markowitz 187


Hirsch has argued that the 1980s merger movement legitimated the
hostile takeover as a "a normal event rather than a deviant innovation"
(1986, p. 821). His account of the process by which hostile takeovers
became commonplace and his discussion ofthe colorful language invented
to describe them forms the starting point for the discussion in this chapter.
The issue to be considered here can be more broadly put as the acceptance
of the finance conception of the corporation as being the normal way to
think about corporations (Fligstein 1990, ch. 7). This conception of the
corporation views the firm as a bundle of assets to be deployed or rede-
ployed depending on the short-run rates of return that can be earned. The
normative acceptance of hostile takeovers in the 1980s reflected the more
general triumph of this view of the corporation. It is useful to consider
what this view entails more carefully.
The American corporation in the early 1980s was under siege from two
exogenous forces: the high inflation of the 1970s and increased foreign
competition. The inflation of the 1970s produced a perverse effect for
large corporations. Their real assets (i.e., land, buildings, tools) were
increasing in value and at the same time, high interest rates meant that the
stock market produced lower returns than other forms of investment. This
left many large American firms undervalued in the stock market (Friedman
1985). Simultaneously, foreign competition, particularly with the Japan-
ese, heated up. This presented a "crisis" to the managers of large firms.
In the context of the response to these exogenous shocks, there are three
important facts to be explained. First, why have the responses to this crisis
been framed in primarily financial terms? Second, given the finance
discourse that dominates, how can one explain the timing of the current
merger movement? Finally, why did some managers embrace the finance
conception and some did not? For those managers who perceived the need
to raise short-term profits and stock prices, forms of financial reorganiza-
tion were attractive. We would expect managers of firms who engaged in
these tactics to avoid being targets of takeovers themselves. This is
because, if successful, they would be unattractive targets for takeovers.
For those managers who ignored the financial conception of the corpora-
tion, one would expect their firms to have been takeover targets. Thus we
assume that the finance perspective on the firm produced profits for those
who practiced it (for a review of some of the evidence, see Jensen and
Ruback 1983). For those who resisted its spread, it potentially made them
targets in the market for corporate control. The theories reviewed propose
quite different answers to these questions.


Finance economics has been concerned with two general questions:

what determines the capital structure of firms and how do financial
markets come to place value on those structures? The key imagery is that
firms can be conceived of in terms of their capital structures (Jensen 1983).
These structures consist of equity (the value of shareholders' stake in the
firm, i.e., the stock price multiplied by the number of shares), assets, and
debt (the value of what the firm owes its creditors, in particular, banks and
bondholders). The decision to hold some mix of equity and debt has been
the focus of continuous analysis (Modigliani and Miller 1958; Myers
1984; Ross 1977; Donaldson 1961; Marsh 1982; Miller 1977; Friedman
1985). Since both equity and debt are claims on the firm's cash flow and
assets, they are alternative ways for managers to fund firms' capital
structures. The issue is what should that capital structure look like from
the point of view of optimizing returns.
A related issue is the mode of governance for the corporation and its
relation to the mix of equity and debt (Jensen and Meckling 1976). Large
firms often have a separation of ownership from control. In this case, the
owners of the capital and the managers are often not the same persons and
this produces a principal-agent problem (Fama 1980; Jensen and Meckling
1976). The concern is the behavior of agents (i.e., managers) and how they
can be monitored to insure that they act to maximize the gains ofthe principals
(i.e., the owners). In the context of the capital structure of the firm, the
argument is that the sides might have different interests in the relative mix of
debt and equity and how the mix might effect dividends and stock prices.
Owners might prefer debt to equity issues as it will tend to maintain the
value of their holdings by not lowering stock prices or diluting earnings.
On the other hand, managers gain by issuing equity that diffuses potential
control of owners. Managers, however, also have an interest in keeping
the stock price up and dividends stable because they could lose control of
the corporation if the firm's performance deteriorates. Indeed, one of the
most potent arguments for why the principal-agent problem is not impor-
tant is that firms with ineffective managers will have low stock prices and
become merger targets (Fama and Jensen 1983a). There is evidence to
support this view: managers first pay for firm expansion through retained
earnings, then use short-term or long-term debt, and only in the last resort
do they issue equity (Myers 1984).
The important factors thought to determine the relative mix of the
capital structure are the bond and stock markets (Jensen and Meckling
1976; Fama and Jensen 1983a; Myers and Majluf 1984). The markets
operate to control the behavior of both managers and owners by properly
Neil Fligstein and Linda Markowitz 189

evaluating the relative value of the firm, its riskiness, and its future
prospects (Lorie et a1. 1985, especially, ch. 4, 5). To the degree that both
markets operate efficiently (i.e., change their evaluations quickly in re-
sponse to news about the firm's current and future prospects), managers
of firms will be forced to act to maximize assets or face losing control over
firms. The most important model of stock market evaluation, the capital
assets price model (CAPM), argues that the firm's price will reflect its
riskiness as an investment (Sharpe 1964).
The merger movement of the 1980s can be explained from this perspective
in the following way. The stock market declined significantly in the 1970s as
high inflation meant that investors sought out higher rates of return to protect
their money and left the uncertain stock market for high yields in the bond
market. The high inflation also meant that the assets of large corporations
(i.e., land, buildings, plants, product lines) were also increasing in value. The
value of corporations can be thought of as the net worth, the assets minus the
debt. Taken together, managers were in the ironic situation that their assets
(and net worth) were increasing in value at the same time their equity and
return on equity were decreasing (Fama and Schwert 1977; Fama 1981;
Taggart 1985; Hendershoot and Huang 1985).
From the point of view of finance economics, efficient managers real-
ized their predicament and engaged in actions to increase profitability and
equity whereas inefficient managers did not recognize their situation and
act quickly enough to remedy it. In the late 1970s and early 1980s,
financial operators began to buy undervalued corporations by paying a
premium to current stockholders and to sell the pieces of the firm at high
enough profits to make money. The decision to liquidate firms depended
on the intention of the buyer to operate purely to turn a quick profit or else
to attempt to run the firm more profitably. Thus the market for corporate
control over assets heated up and created the 1980s merger movement.
Efficient managers who wanted to maintain control over their firms had
to engage in actions to increase their equity and stock price and hence
bring their equity in line with their real net worth. They used a number of
tactics. First, many purchased shares of their own stocks either with
retained earnings or by borrowing funds. This had the effect of reducing
the amount of stock in circulation and driving up the share price. If the
money was borrowed, this increasing indebtedness thereby lowered net
worth, and taken together, these actions lowered the odds of takeover.
They could use debt to finance the buyout or purchase of stock of other
firms that could increase their indebtedness as well and thereby make them
less attractive candidates for mergers. Presumably, such buyouts would
operate to raise profits, increase assets, and not dilute equity. Finally, they
could divest divisions that were underperforming and use profits to repurchase

stock shares, distribute extra dividends to shareholders, or engage in

mergers or stock purchases of other firms.


Organizational and political sociologists have long been fascinated by the

issue of who controls the large corporations. Much of the original interest in
this field was sparked by the potential for the concentration of power by the
owners of the large corporations. Sociologists have generally attempted to
refute the Berle and Means hypothesis (1932) that managers were increas-
ingly coming to control large corporations. The central claim of political
sociologists was that the amount of managerial control of the largest corpo-
rations was greatly overstated (Zeitlin 1974; Burch 1972; Kotz 1978).
Scholars interested in this debate have recently turned their attention to
suggesting why and how ownership forms might produce different results.
Generally, most of these attempts use some form of resource dependency
theory as the mechanism of control (Pfeffer and Salancik 1978). The
argument is that organizations are dependent on their environments for
resources. Hence, the greatest leverage on a firm will potentially be from
organizations that have important resources a firm needs. For instance,
Burt (1983) argues that firms systematically seek to co-opt these depend-
encies, mainly by putting important suppliers or customers on boards of
directors, in order to secure their support.
Using organizational interdependence as a theoretical starting point, a
number of authors have proposed some advancements to the control
debate. Mintz and Schwartz (1985) have tried to show the centrality of
banks, insurance companies, and other institutional investors in control-
ling the large corporations by examining two forms of data: corporate
interlock data and selected cases of organizational intervention by banks.
Network methods reveal that banks and insurance companies are the
central links between firms (Mintz and Schwartz 1985; Mizruchi 1983;
Pennings 1980). It is also the case that when corporations are in crisis,
large debtholders, often banks or insurance companies, supervise financial
reorganization. Mintz and Schwartz argue that this produces financial
hegemony that implies that the dependence of firms on outside financing
opens them up to the potential of being controlled, or at least limited in
their actions, by banks.
Burt argues that the capture of financial investors or suppliers of boards
of directors allow firms to co-opt their important dependencies. This is an
alternative perspective to the one developed by Mintz and Schwartz
because it has managers actively controlling their environment and poten-
Neil Fligstein and Linda Markowitz 191

tially reducing the prospects for external control. Burt, however, is unable
to show the effects of such interdependencies on profits (1983).
The strongest statement of the potential of bank control comes from
Kotz (1978). He argues that management (and presumably family) con-
trolled firms would avoid debt in order to lessen financial control. Con-
versely, financial institutions would use their control over firms to in-
crease their business. Firms already under the control of large financial
institutions would be more likely to go into debt. Finance controlled firms
would be more likely to engage in mergers for similar reasons. Mizruchi
and Stearns (1988) have provided evidence that firm debt is correlated
with the presence of bankers on the board of directors.
Brancato and Gaughan (1988) argue that large institutional investors
have come increasingly to control the stock of large firms. They suggest
that this might have another unwelcome effect on firms. Because institu-
tional investors generally work with short profit horizons, finance con-
trolled firms (or any firm controlled by a large institutional investor) is
likely to be under considerable pressure to generate short-term profits.
Failure to do so would result in management turnover or the prospects of
having the firm sold off.
These literatures can be combined to suggest an account of the financial
reorganization of the large corporations in the 1980s. The finance control
literature would argue that banks and institutional investors have increased
their power over firms in the past 20 years (Kotz 1978; Mizruchi and
Stearns 1986; Mintz and Schwartz 1985). The basis of this control is direct
ownership of equity or large holdings of corporate bonds. Boards of
directors function to give banks monitoring power over firms. The in-
creases that have occurred in financial manipulations such as stock repur-
chases, higher levels of indebtedness, and mergers and divestments in the
1980s can be explained by the increase of bank and institutional investor
control. In the cross section, one would expect that bank and institutional
investor controlled firms would be more likely than manager or family
owned firms to undertake such actions. It is the rise of finance controlled
firms that accounts for the rise of financially oriented maneuvers.



The final perspective is presented by Fligstein (1985,1987,1989,1990).

This view is based on the notion that firms' strategic actions are shaped
by two primary forces: the behavior of their competitors and the actions
of the government to define what is competitive and anticompetitive

behavior between firms. Managers in firms have attempted to find ways

to control direct forms of competition since their inception in the late 19th
century. Their search for stability has been blunted by the invention and
implementation ofantitrust laws that have defined the limits of their ability
to control competition.
The key argument is that managers and owners in firms search for stable
patterns of interactions with their largest competitors. Once stable patterns
of interaction prove to be both legal and profitable, firms set up organiza-
tional fields that tend to produce and reproduce those patterns. The prin-
ciples that guide interaction in those fields can be termed a conception of
control. Conceptions of control are worldviews held by managers that
define what appropriate and inappropriate responses to situations are.
They operate to filter information and produce solutions to given crises of
firms. They are collectively held across large firms in established organi-
zational fields and operate to support the organizational field by establish-
ing a taken-for-granted quality to actions in that field. Managers across
firms develop expectations of one another's behavior and that increases
the reproducibility of a given set of rules. The managers who come to
control firms are selected on their basis of allegiance to a given conception
of control (Fligstein 1987).
Over the past 100 years there have been four conceptions of control that
have organized the worlds of the largest corporations: the direct, manufactur-
ing, sales and marketing, and finance conceptions (Fligstein 1990). Fligstein
(1990) has documented the conditions that have produced the rise and spread
ofthese various conceptions. For the purposes of this chapter, we will restrict
our attention to the rise of the finance conception of control.
The finance conception of control can be defined as the point of view that
conceives of the firm purely in financial terms. Firms are considered to be
bundles of assets, each earning varying rates of return. The basic impetus to
action is guided by the immediate profit prospects of any given set of assets.
All actions are undertaken to promote short-term financial gain and the basic
tactics include mergers, divestitures, stock repurchases, and the judicious use
of debt and equity. The finance perspective views the firm as not wedded to
a certain industry or product line. Instead, the finance conception of control
values assets as to their immediate profitability. One implication of this
finance conception of control is the attempt to hold assets in different
industries in order to smooth out the effects of business cycles on profits.
Indeed, one good indicator of whether or not a firm is operating with a
financial conception of control is a diversified product mix.
There were two conditions that produced the finance conception of
control. First, large firms in the postwar era were already fairly diversified
in their product lines. The problem of internally controlling a large number
Neil Fligstein and Linda Markowitz 193

of products opened an opportunity for executives who could claim to

evaluate the profit potential of each product line. Finance executives
reduced the information problem to the rate of return earned by product
lines and thereby made the large diversified corporation manageable.
Second, the federal government was strictly enforcing the antitrust laws
in the early postwar era and had passed an anti merger law that made it
difficult to merge with direct competitors or suppliers. This encouraged
product diversified mergers. The financial executives who could evaluate
products outside of a firm's expertise were invaluable in such efforts
(Fligstein 1990, ch. 6).
The most spectacular organizational examples of the new finance con-
ception came from firms outside the mainstream of American corporate
life. The men who pioneered the acquisitive conglomerate (Tex Thornton
at Textron, Jim Ling at L-T-V, and Harold Geneen at ITT) showed how
financial machinations involving debt could be used to produce rapid
growth with little investment of capital. All of the financial forms of
reorganization including mergers, divestitures, leveraged buyouts, the
accumulation of debt, and stock repurchasing were invented or perfected
in this period. The 1960s witnessed a large-scale merger movement whereby
many of the largest corporations substantially increased their size and
diversification. As a result of this success, finance executives increasingly
became CEOs of large corporations (Fligstein 1987).
By 1969, the finance conception of control had come to dominate the world
of the largest corporations. The merger movement of the 1960s was ended by
two phenomena. The Nixon Administration's Antitrust Division filed law-
suits against some of the largest acquisitive conglomerates and thereby put a
chill on the stock market. The decline of the stock market beginning in 1969
and the increase in interest rates made it less profitable to engage in merger
activity. The acquisitive conglomerates performed less well in the 1970s and
many of them divested substantial assets. Nonetheless, the finance conception
of control remained dominant in the corporate world.
From this perspective, the conditions were right for a continuation of
the 1960s finance-oriented merger movements. What was needed was a
spark to set the movement going. This was provided by the Reagan
Administration's antitrust officials who basically suspended the antitrust
laws in the 1980s and gave the green light to all forms of mergers, large
and small, vertical and horizontal. The Reagan Administration also sub-
stantially reduced corporate income taxes, thereby providing capital for
the merger movement. From this perspective, the 1980s market for corpo-
rate control was driven by the already existing finance conception of the
firm and the changes in the regulatory environment. The firms that would
be leaders in this movement would be those that were led by finance

executives or where diversification of products was an accepted fact. If

competitors were involved in these forms of reorganizations, then any
given firm would be more likely to engage in such actions itself. Firms
that resisted forms of financial reorganization become targets of those who
embraced financial reorganization.


All of the perspectives point to financial conceptions of the corporation

as pivotal to understanding what has happened in the past 10 years. Yet,
they point to radically different actors, interests, and power bases as
causally significant. Although the theories can be reconciled at some level,
there are distinct differences between the theories that suggest different
mechanisms by which the financial reorganization of large corporations
can be explained. It is possible, of course, that a number of these mecha-
nisms are in operation and their effects could have been complementary
(and indeed, the empirical results will show that this is the case).
It would be difficult to distinguish between these three explanations in
terms of their ability to explain the timing of the current merger wave
because all offer reasons as to why the late 1970s and early 1980s were
pivotal. But they can be distinguished as explanations of which firms were
likely to be reorganized or targets of mergers. To the degree that the
mechanisms they propose to explain which firms were involved actually
do so, one can offer more or less support for their general view of the
timing of that event. For instance, if firms that had more undervalued
assets were more likely to be merger candidates, then one can infer that
the general undervaluing of assets was important as a precipitating factor
for the merger movement.
The differences between the theories, however, are quite real and deeply
rooted. For finance economics, all managers have to pay attention to is
their stock price and its relation to debt, equity, earnings, and the value of
assets. Therefore, manager, bank, and family controlled firms would all
behave in terms of these underlying calculations. This means that the
mechanism that would predict action was centered in the underlying
financial fundamentals of firms. For firms whose stock price fell too low,
we would expect financial reorganization such as stock repurchases, di-
vestitures, or increased debt independent of who controlled the firm.
Similarly, for those firms that did not reorganize, we would expect it more
likely that they would be takeover targets.
The bank control thesis predicts that financial reorganization will be
more likely to occur in firms with bankers and large institutional investors
Neil Fligstein and Linda Markowitz 195

on their boards of directors as holders of both equity and debt. Mergers and
divestitures and increased indebtedness should follow from the presence of
bank and institutional investors. This view stresses the role of powerful actors
and the dependence of firms upon those actors. In this sense, the financial
perspectives of these actors and their base of power as holders of equity and
debt force managers toward forms of financial reorganization beneficial to
bankers and institutional investors interested in short-run gain.
The finance conception of control focuses on three mechanisms as leading
to increased likelihood of financial reorganization. First, the presence of a
finance-oriented CEO will increase the likelihood of financial reorganization
because the logical course of action for those executives will be to use those
tools to attempt to increase profitability. Second, in firms that do not have
finance-oriented executives as CEOs, we might still see a propensity to
engage in financial reorganization if other firms in their organizational field
are doing so. This is because the field may be dominated by the finance
conception of control. Hence, appropriate behavior will be suggested by
competitors, independent of the background of the CEO. Finally, the product
mix of a firm will signify the presence of a finance conception of control as
well. Firms that are diversified into related or unrelated products have
managers who are more likely to evaluate product lines by their relative
profitability. These managers are more likely to be operating with a finance
conception of control and should engage in forms of financial reorganization
as a result. It should be noted that this hypothesis counters the conventional
wisdom of the 1980s, which was that diversified firms performed poorly and
were generally targets of mergers.


The sample used to test the hypotheses include the 100 largest industrial
corporations in the United States in 1979. Data were coded for each year for
the period from 1979 to 1987. The beginning and end of any merger move-
ment cannot be dated precisely. The most recent merger wave arguable began
in 1981 and the selection of 1979 as the starting year allows us to obtain data
for years before the beginning of the real upturn in mergers. The list was
obtained from Fortune magazine (1980). Nine observations existed per case
for each company, unless that company was bought out. After a firm was
merged into another, no further information was collected.
The selection of the largest firms can be justified on a number of
grounds. The largest corporations are the most important business organi-
zations in American society. They have been the focus of most of the
theorizing and empirical work just discussed, partly because there is a

Table 10.1 List of the 100 Largest Firms in 1979 That Were Bought Out
by 1989

American Can Signal Co.

Cities Services Sperry Rand
Continental Oil General Foods
Getty Oil Continental Can
Gulf Oil Beatrice Foods
Kennecott Copper Superior Oil
Marathon Oil Federated Department Stores
RCA Safeway
Republic Steel Owens-Illinois Glass
Shell Oil Standard Oil-Ohio

great deal of data publicly available for these organizations. Further, the
actions of their managers operate as a role model for other firms. One could
argue that the largest firms are a restricted sample and therefore our
parameter estimates suffer from sample selection bias. This is only a
problem if one is asserting that the parameter estimates obtained here
apply to a larger sample of firms, say for instance, all publicly held
corporations. We make no such claims here. Indeed, the focus of this study
is the population of the largest firms and therefore we are not sampling,
but instead providing population parameters.
There were two dependent variables in the analysis: whether or not the firm
was merged and whether or not the firm engaged in the various forms of
financial reorganization. The variable indicating whether or not the firm was
merged in a given year was coded as a dummy variable with the code of I,
signifying that the firm was merged. Information about merger activity was
collected from the 1980 through 1988 editions of Moody's Manual. Out of
the 100 largest corporations coded in 1979, one-fifth had been bought out by
1989. A list of those companies can be found in Table 10.1.
We selected four forms of financial reorganization based on the theoretical
considerations discussed earlier: mergers, divestitures, acquiring one's own
stock, and acquiring stock of other firms. Information about acquiring one's
own stock was collected from the Wall Street Journal Index (for period
1980-1988). The other data were coded from Moody's Manual (for period
1980-1988). The four types of reorganization were coded in several ways: (1)
as dummy variables indicating whether or not the firm had evidence of taking
that form of reorganization in a given year, (2) as a simple scale in which we
summed, in a given year, whether the company engaged in any of the
reorganization activities, and (3) as a factor scale.
Neil Fligsteinand LindaMarkowitz 197

Table 10.2 Results of Factor Analysisof Forms of Financial Reorganization

Variables· Factor Weights

Mergers .56
Divestitures .40
Stock Purchases .49
Stock of Others .39
Eigenvalue 1.84
Percentage of Variance Explained 0.54

NOTE: a. Variablesdefined in text.

The factor scale was obtained using an oblique rotation in SPSSX. The
results are presented in Table 10.2. It is useful to consider them in some
detail. The four forms of financial reorganization do form one unique
factor with an eigenvalue greater than 1. The factor weights are relatively
similar implying that each of the forms of financial reorganization is
equally important for the underlying construct. Together, these results
give credence to our notion that the forms of financial reorganization
reflect a set of behaviors that are undertaken by managers with the goal of
increasing the financial value of the firm; that is, they reflect a financial
conception of the corporation. Managers were likely to undertake more
than one of the actions in agiven year and thereby were either dominated
in their behavior by a finance conception of action or not. Both the factor
scores and the simple additive scale were used in the analysis.
We created three independent variables to test the finance conception
of control hypothesis. The first is a dummy variable coding whether or not
the chief executive office (CEO) of the company has a finance background
(1 = finance CEO). The name of the CEO in 1979 was found using the
1980 edition of Moody's Manual. Other sources, such as Who's Who in
America and Who's Who in Business. were then used to decipher the
CEO's job background. If a CEO had held a number of jobs that were not
in finance functions, then that CEO was not coded as a finance CEO.
The second independent variable concerns the percentage of other firms
in an organizational field with finance CEOs. Our argument was that
although a company might not have a finance CEO, other firms in the field
might be dominated by the finance conception of control. If the company
believes the other firms with finance CEOs have a competitive edge, then
it is likely to engage in activities similar to its competition. To capture this
mimetic effect, in 1979, the percentage of finance CEOs among the 100
largest firms was calculated in each industry (measured at the two-digit
SIC code level). Each firm's CEO was excluded from that calculation in

order to make the measure orthogonal from the measure of the background
of the CEO of a given firm.
The last independent variable has to do with the product mix of the firm.
We argued that firms with related or unrelated product mixes were more
likely to be controlled by the finance conception of control than were firms
with production concentrated in one dominant line. A product dominant
strategy entails a firm that produces more than 70% of its revenues in a
single product line. A product-related strategy entails a firm that produces
many related products or market extensions (i.e., a chemical company
producing paints and explosives). However, no single line accounts for
more than 70% of the output. A product-unrelated strategy entails a firm
that involves itself in unrelated businesses for a substantial portion of its
revenue (for instance, L-T-V, at one point in time, produced guided
missiles, steel, and owned a rental car company). Information for this data
was collected from Rumelt (1974) and the 1980 edition of Moody's
Manual. Two dummy variables were coded to reflect this distinction. The
left-out category for the measure was product dominant. Each of the other
variables was coded 1 if the product mix was related or unrelated.
The second hypothesis being tested concerns how corporate activity
might be a function of who controls the company. There are three inde-
pendent variables used to test this hypothesis. The first measures how
owner activity might differ from manager activity in terms of engaging in
reorganization tactics. A dummy variable was created in which a company
controlled by an owner was coded as 1. We used Herman's (1981) coding
scheme as his data referred to the period closest to 1979.
The second and third independent variables concern the composition of
the board of directors. It has been argued that individuals on the board of
directors have a potential to control activities of any given company (Burt
1983). Thus, the second independent variable measures the percentage of
bankers on the board of directors for each given year, and the third
independent variable measures the percentage of nonbank financial insti-
tutions (such as insurance companies, pension funds, etc.) on the board of
directors for each given year. Information for these measures was col-
lected from Moody's Manual for the years from 1980 to 1988, Standard
and Poor's Register of Corporations, Directors and Executives for the
years from 1979 to 1986, and the 1985 and 1987 editions of The Corporate
The last hypothesis being tested concerns the finance economics argu-
ment. This view stresses the importance of maintaining a balance between
equity and debt so that a company is not in a position to be merged. The
prediction is that company actions will be a function of their financial
position. The variable measuring this is calculated as follows:
Neil Fligstein and Linda Markowitz 199

(Assets - Debts)/(# Of Outstanding Shares ee Price Of Shares)

The numerator of this measure contains the net worth of the firm,
whereas the denominator reflects the firm's equity. The argument is that
when this ratio is less than I, the equity is worth more than the net assets
and financial reorganization is unlikely. As the ratio grows greater than I,
the net worth is greater than the equity and the firm is more likely to be a
reorganized or a target of takeovers. The value of the debts and assets was
collected from Moody's Manual (1980 to 1988). The stock price and
number of shares was collected from Moody's Manual and The Daily Stock
Price Record for the years between 1979 and 1988. All measures refer to
conditions at the end of the year.
The analysis proceeds in the following way. First, we look for a causal
relationship between all of the independent variables and the first depen-
dent variable, reorganization activity. Then we use the reorganization
variable and the other independent variables to attempt to predict whether
or not the company has been the object of a merger. We argued that there
ought to be a negative relationship between having undertaken forms of
financial reorganization and being bought out.
The data are structured such that some of the independent variables are
constants throughout the 9 years investigated (i.e., finance presidents,
owner or manager controlled), and some are variable throughout the
investigation (i.e., composition of the board of directors, assets/stock
value). Because the goal of the analysis is to use the data yearly to predict
the dependent variables, the independent variables that changed yearly are
lagged by 1 year, thereby reducing the number of observations from 9 to
8 years.
Two methods were used to analyze the data. When the continuous
dependent variables were analyzed, we used an error components model
to specify the error structure. This is because we have multiple observa-
tions on the same units of analysis over time. When the dependent variable
was discrete, in the case of whether or not the firm was merged, a discrete
time, discrete data model was used (Allison 1981). All statistical tech-
niques were performed using LIMDEP.


Table 10.1 shows the names of those ofthe 100 largest companies that were
objects of a merger during the years from 1979 to 1987. This shows that the
merger movement greatly effected the population of the largest firms in the
American economy. A brief glance at the list shows that 7 out of 20 of those

Table 10.3 Means and Standard Deviations of Variables in the Analysis

(N= 753)

Variable Name" X S.D.

Merged 0.03
Reorg I 0.50 0.43
Reorg 2 1.03 0.89
% Fin pres 0.32 0.20
Finance pres 0.32 0.46
Related Strategy 0.55 0.49
Unrelated Strategy 0.24 0.43
Owner Control 0.25 0.47
% Bank Directors 0.03 0.05
% Institution Directors 0.04 0.05
Assets/Stocks 0.81 0.98

NOTE: a. Variablesdefined in text.

merged were oil companies. There were 22 oil companies on the list in
1979 and this percentage is higher than would be expected. Table 10.3
presents the means and standard deviations of the variables for the 8 years
we investigated (N =753). All firms with missing data were removed from
the analysis. The Reorgl variable is the reorganization variable computed by
factor analysis. The Reorg 2 variable is the additive scale that shows that each
company averaged one form of financial reorganization per year.
The variables concerning the financial conception of control hypothesis
reveal that 32% of the companies had finance presidents and in each field
there was an average of 32% of the firms with finance presidents in their
organizational fields. Most companies were likely to engage in some type
of product diversification: 55% of the firms had a product-related strategy
and 24% had a product-unrelated strategy, which shows that these tactics
that reflect the finance conception of control have become quite common.
Only 25% of the companies are controlled by the owners. The majority
(75%), thus, are controlled by managers. The board of directors are made
up of 3% bankers and 4% institutional investors. The rest of the board of
directors are company insiders or outsiders with no financial institution
attachment. The net worth/value ratio variable associated with finance
economics has a mean ofO.8!. This means that, on average, a company's
net worth is less than its stock value (number of shares multiplied by the
price of the stock). The average firm was thus not likely to be bought out.
As the ratio increases past 1, we find that assets become more valuable
than the stock. It is in these cases that finance economists predict a merger.
Neil Fligstein and Linda Markowitz 201

Table 10.4 Error Components Model Predicting Financial Reorganization

of Firm (N = 753)

Reorg 1 Reorg 2
Yariables" b SE(b) b SE(b)

% Fin Pres -0.12 0.20 -0.06 0.10

Finance Pres 0.06* 0.03 0.03* 0.01
Related Strategy 0.31 ** 0.08 0.15** 0.04
Unrelated Strategy 0.30** 0.10 0.14** 0.05
Owner Control -0.01 0.02 -0.01 0.01
% Bank Directors 0.57 0.61 0.25 0.29
% Institution Directors 1.66** 0.60 0.78** 0.29
Asset/Stock -0.03 0.03 -0.01 0.02
Constant 0.74 0.36
R2 0.30 0.32

NOTE: a. Variables defined in text.

* P < .05; ** P < .01.

Given that the standard deviation was quite large (0.98), there were likely
to be a great many firms that were merger candidates.
Table lOA is an attempt to model the two composite measures of the
reorganization variables whereas Table 10.5 models each of the reorgani-
zation activities separately. Because results from the factor analysis sup-
port the idea that the four reorganization activities are conceptually re-
lated, we will consider Tables lOA and 10.5 together.
Table lOA demonstrates that Reorgl (the factor measure) and Reorg2
(the additive measure) are predicted by the same variables. The first three
variables represent measures related to the finance conception ofcontrol. The
percentage of finance CEOs in an organizational field had no effect on the
reorganization variable. But, whether or not the CEO was a finance CEO and
if the company had a product-related or -unrelated type of product mix had a
significant, positive effect on firms engaging in the forms of financial reor-
ganization. This is strong evidence that firms that already were dominated by
the finance conception of control were likely to have managers who engaged
in financial reorganization of their firms.
Variables measuring the owner/managerlbanker factors show that only the
percentage of institutional investors on the board of directors had an effect
on the dependent variable. Neither the types of control over the firm nor the
percentage of bankers on the board of directors had a significant effect on
whether or not the company reorganized. In this case, institutional investors
were able to convince managers to undertaking financial reorganization.

Table 10.5 Logit Models Predicting Separate Forms of Financial


Stock Stock of
Mergers Divestitures Repurchases Others
Variables" b SErb) b SErb) b SErb) b SErb)

% Fin Pres 0.05 0.26 0.14 0.48 0.09 0.26 -0.25 0.81
Fin Pres 1.30* 0.63 0.54** 0.18 1.38* 0.62 0.32* 0.14
Related Strategy 0.34* 0.17 0.59** 0.21 0.29* 0.12 0.34* 0.16
Unrelated Strategy 0.33* 0.15 0.66** 0.25 0.29* 0.14 0.69 0041
Owner Control -0.03 0.08 0.05 0.05 -0.04 0.08 -0.03 0.10
% Bank Directors 0040 0.80 1.12 1.47 -0.27 0040 0.59 0048
% Institution Directors 1.64* 0.81 2.23** 0.74 1.16** 0.37 0.72 0.38
Asset/Stock -0.01 0.11 0.21* 0.10 0.02 0.08 0.08 0.11
Constant -1.83** 0.32 -1.24** 0.25 -1.71** 0.31 -2.51** 0040
,l(dO 10.79 (8) 21.0 (8) 9.15 (8) 11.80 (8)

NOTE: a. Variables defined in text.

• p < .05; •• P < .01.

The finance economic hypothesis had no support in this analysis. The

actual financial situation of the company in terms of its ratio of net to stock
worth had no effect on the likelihood of managers engaging in financial
reorganization. One interpretation of this is that the balance sheets of firms
were not sufficient conditions to promote the reorganization of firms.
Instead, such reorganization only took place if there were actors for whom
those actions made sense. The fact that firms who had institutional invest-
ors on their boards or who were already dominated by the finance concep-
tion of control were more likely to engage in forms of financial reorgani-
zation shows the importance of actors who have a perspective on the world
and the power to enforce it.
Table 10.5 treats each reorganization activity as a separate measure.
Generally, the results are almost identical to those from Table 10.4. The
differences from these results and the results from Table 10.4 are small
but worth mentioning. Whether or not the firm had a finance CEO was a
significant predictor of each of the four reorganization activities. How-
ever, whether or not the firm had an unrelated product strategy had no
significant effect on purchasing stock from other companies. The percent-
age of institutional investors on the board of directors also did not have a
significant effect on the stock purchase variable. In the factor score, this
variable was the weakest on the factor loading and therefore the fact that
the variables that explain it are less consistent than the variables that
explain the other variables is not totally surprising.
Neil Fligstein and Linda Markowitz 203

Table 10.6 Logit Models Predicting Whether or Not a Firm Was Bought
Out or Merged (N = 753)

Variables' b SErb) b SErb)

Reorg 1 -0.55** 0.10

Reorg 2 -0.26** 0.05
% Fin Pres -0.29 0.57 -0.38 -0.29
Finance Pres -0.43 0.76 -0.47 0.34
Related Strategy -0.38 0.69 -0.11 0.10
Unrelated Strategy 0.47 0.76 0.30 0.24
Owner Control -0.64 0.80 -0.04 0.04
% Bank Directors 4.36** 1.71 9.56** 3.79
% Institution Directors -0.41 0.25 0.48 0.58
Stock Assets -0.25 0.28 -0.24 0.42
Constant -2.35** 0.42 -2.15** 0.63
X2(dt) 12.5(8) 24.2(8)

NOTE: a. Variables defined in text,

• P < .05; •• P < .01.

There is also an effect of the net worth/stock ratio on divestitures. Firms

where assets were worth a great deal more than the stock were more likely
to divest divisions than firms where assets and stocks were more evenly
matched. This is consistent with the prediction that the finance economists
make. It suggests that the sale of assets is undertaken to increase the stock
price and make the firm less attractive to takeover.
Table 10.6 presents the results from the second part of our analysis where
a discrete time/discrete data model is used to predict a company being a
merger target. Only two variables had a significant effect on whether or not
a company was bought out. The first variable, as expected, was the reorgani-
zation variable. Reorganization had a negative effect on being merged. In
other words, companies engaging in financial tactics were less likely to be
targets of a merger. The second variable predicting company merger was the
percentage of bankers on the board of directors.
In firms operating with a finance conception of control or with institutional
investors convincing managers to accept the finance conception of control,
we see an orientation toward forms of financial reorganization as a way to
increase profits and not toward forcing the firm into merger with another firm.
On the other hand, bankers on the board of directors do not affect reorgani-
zation activity, yet they do affect whether or not a company is likely to be a
target of merger. One plausible interpretation is that bankers are brought
on to the board of directors as the firms are failing. In this context, in order
to secure their financial investments, they force the sale of the company.


The 1980s reflected an intensification of the finance conception of the

corporation. In firms where such a conception already dominated, those
with a finance CEO or with a diversified product mix, managers were most
likely to protect their positions by undertaking financial actions to pre-
serve their control over firms. Where the finance conception of control did
not dominate, we see the role of financial institutions as the agents of
change forcing managers to shift toward the finance conception of control.
Institutional investors got managers to concentrate on short-run financial
manipulations, presumably to increase stock prices and prevent mergers.
Banks, who probably held most corporate debt, often came on the scene
quite late in the process. Their presence was associated with selling off
the firm. Both banks and institutional investors caused non-finance-ori-
ented managers to pay attention to issues of financial reorganization.
The lack of effect of the basic financial facts of the firm in predicting either
reorganization or ultimate merger could be due to one of two factors. First, it
is possible that the measure we used was flawed and that other measures might
show such effects. Second, if the results are accepted, it shows that the
existence of rich assets relative to stock price may be a necessary but not a
sufficient condition for the reorganization of firms. From our perspective,
actors must come forward to perceive this situation and act.
These results suggest that the economic actions of the leaders of the largest
firms have increasingly been informed by a certain perspective on what the
firm is. This cultural construction has become a powerful tool of key actors
to make profits. We have shown that where it does not dominate, actors with
the finance conception can emerge on boards of directors to enforce such a
view. If top executives or important board members do not force realignment
with that perspective, then firms are liable to become the targets of executives
who do. These results support the view that economic actors are embedded
in a social world that shapes their interests, perceptions, and actions. This
social world provides the intellectual justification for a certain view of the
corporation and the power to act in terms of that view. The results presented
here reinforce the general sense that market relations are embedded in social
relations and that actions only make sense when understood from the context
of these social relations (Granovetter 1985).
Although we have provided no direct evidence for the importance of
public policy in the generation of the financial reorganization of large
firms, it is clear that one implication of our view is that some forces must
have produced the intensification of the finance orientation toward firms.
There are two plausible hypotheses in this regard. First, the Reagan
Administration's suspension of the antitrust laws and its tax policies in the
early 1980s certainly provided some impetus to finance-oriented managers
Neil Fligstein and Linda Markowitz 205

and institutional investors. Second, the 1970s did leave large corporations
asset rich and stock price poor. Additional work should attempt to adjudic-
ate between these competing explanations of the initial shocks that pro-
duced the financial reorganization of American corporations.
But whatever the cause, the finance conception of the large corporation
has triumphed. Even though mergers have once more slowed, the finance
conception of the corporation dominates discourse. The power and inter-
ests that have produced this view now control the world of the largest
firms. Unless public policy changes the rules to make these views of the
corporation less profitable or unless a recession discredits such views
entirely, the next 10 years are not likely to change this trend.


Allison, P. 1982. "Discrete Time Methods for the Analysis of Event Histories." Pp. 61-98 in
Sociological Methodology. edited by S. Leinhard. San Francisco: Jessey-Bass.
Berle, A. A. and G. C. Means. 1932. Modern Corporation and Private Property. New York:
Commerce Clearing House.
Brancato, C. K. and P. A. Gaughan. 1988. "The Growth of Institutional Investors in U.S.
Capital Markets." Institutional Investor Project. New York: Columbia University,
Columbia Center for Law and Economic Studies.
Burch, P. 1972. The Managerial Revolution Reassessed. Lexington, MA: Lexington Books.
Burt, R. 1983. Corporate Profits and Cooptation. New York: Academic Press.
Donaldson, G. 1961. Corporate Debt Capacity. Boston: Harvard Graduate School of Busi-
ness Administration.
Fama, E. 1980. "Agency Problems and the Theory of the Firm." Journal ofPolitical Economy
- - - . 1981. "Stock Returns, Real Activity, Inflation, and Money." American Economic
Review 91:545-65.
Fama, E. and M. C. Jensen. 1983a. "Separation of Ownership and Control." Journal of Law
and Economics 26:301-26.
- - - . 1983b."Agency Problemsand ResidualClaims." JournalofLaw and Economics26:327-50.
Fama, E. and G. W. Schwert. 1977. "Asset Returns and Inflation." Journal of Financial
Economics 5:114-46.
Fligstein, N. 1985. "The Spread of the Multidivisional Form." American Sociological Review
- - - . 1987. "The Intraorganizational Power Struggle: Rise of Finance Personnel to Top
Leadership in Large Corporations, 1919-1979." American Sociological Review 52:44-59.
- - - . 1989. "Bank Control, Owner Control, or Organizational Dynamics: Who Controls
the Large Modern Corporation?" Paper presented at the Annual Meetings of the
American Sociological Association, San Francisco.
---.1990. The Transformation of Corporate Control. Cambridge, MA: Harvard Univer-
sity Press.
Fortune Magazine. 1980. July issue.
Friedman, B. 1985. "The Substitutability of Debt and Equity Securities." In Corporate
Capital Structures in the United States, edited by B. Friedman. Chicago: University of
Chicago Press.

Granovetter, M. 1985. "Economic Action and Social Structure: The Problem of Embedded-
ness." American Journal of Sociology 91:481-510.
Hendershoot, P. H. and R. D. Huang. 1985. "Debt and Equity Yields, 1926-1980." In
Corporate Capital Structures in the United States, edited by B. Friedman. Chicago:
University of Chicago Press.
Herman, E. 1981. CorporateControl, CorporatePower. New York: Cambridge University Press.
Hirsch, P. 1986. "From Ambushes to Golden Parachutes: Corporate Takeovers as an Instance
of Cultural and Institutional Integration." American Journal of Sociology 91:800-837.
Jensen, M. C. 1983. "Organization Theory and Methodology." Accounting Review 58:319-39.
Jensen, M. C. and W. H. Meckling. 1976. "Theory of the Firm: Managerial Behavior, Agency
Costs, and Ownership Structure." Journal of Financial Economics 3:305-60.
Jensen, M. C. and R. S. Ruback. 1983. "The Market for Corporate Control: the Scientific
Evidence." Journal of Financial Economics 11:5-50.
Kotz, D. M. 1978. Bank Control of Corporations in the United States. Berkeley: University
of California Press.
Lorie, J. H., P. Dodd, and M. H. Kimpton. 1985. The Stock Market. Homewood, IL: Dow
Jones- Irwin.
Marsh, P. R. 1982. "The Choice Between Equity and Debt: An Empirical Study." Journal of
Finance 37:121-44.
Miller, M. 1977. "Debt and Taxes." Journal of Finance 32:261-75.
Mintz, B. and M. Schwartz. 1985. The Power Structure of American Business. Chicago:
University of Chicago Press.
Mizruchi, M. 1983. The American Corporate Network. 1904-1973. Beverly Hills, CA: Sage.
Mizruchi, M. and L. Stearns. 1988. "A Longitudinal Study of the Formation of Interlocking
Directorates." Administrative Science Quarterly 33: 194-210.
Modigliani, F. and M. Miller. 1958. "The Cost of Capital, Corporation Finance, and the
Theory of the Firm." American Economic Review 63:261-97.
Moody's Investor Services. 1980-1988. Manual of Industrials. New York: Moody's.
Myers, S. 1984. "The Capital Structure Puzzle." The Journal of Finance 39:575-92.
Myers, S. and N. S. Majluf. 1984. "Corporate Financing and Investment Decisions When Firms
Have Information That Investors Do Not." Journal of Financial Economics 13:187-221.
Pennings, J. 1980. Interlocking Directorates. San Francisco: Jessey-Bass.
Pfeffer,J. and H. Salancik. 1978. The ExternalControlofOrganizations. New York: Harper & Row.
Ross, S.A. 1977. "The Determination of Financial Structures: the Incentive-Signalling
Approach." Bell Journal of Economics 8:23-40.
Rumelt, R. 1974. Strategy, Structure, and Economic Performance. Boston: Harvard Graduate
School of Business Administration.
Sharpe, W. F. 1964. "Capital Asset Prices: A Theory of Market Equilibrium Under Condi-
tions of Risk." Journal of Finance 19:425-42.
Standard and Poor's Investor Services. 1976-1986. Register of Corporations, Directors. and
Executives. New York: Standard and Poor's.
---.1979-1988. The Daily Stock Price Record. New York: Standard and Poor's.
Swidler, A. 1986. "Culture in Action." American Sociological Review 51:273-87.
Taggart, R. A. 1985. "Secular Patterns in the Financing of U.S. Corporations." In Corporate
Capital Structures in the United States, edited by B. Friedman. Chicago: University of
Chicago Press.
Wade, W. (ed.). 1985,1987. The Corporate 1000. Washington, DC: Monitor Publishing.
Wall Street Journal Index. 1980-1988. New York: Dow Jones and Co.
Who's Who in America. 1980. Chicago: Marquis Publishing.
Who's Who in Business. 1980. Chicago: Marquis Publishing.
Zeitlin, M. 1974. "Corporate Ownership and Control: The Large Corporation and the Capitalist
Class." American Journal of Sociology 81:894-903.
The Conservative Revolution That Wasn't:
New Right Populism and the Preservation
of New Deal Liberalism

Anne Wortham

Ronald Reagan has moved the entire political debate sharply right. Conserva-
tives have now won the war of ideas on every foreign, domestic, and social
policy issue.
Benjamin Hart (1987)

Ronald Reagan should have repealed the New Deal and the Great Society and
pulled the United States out of the United Nations for openers. He didn't.
Joseph Sobran (1988)

Many thought the "Reagan Revolution" marked the sure turning of tides in
the early 1980s.... The great turning back turned out to be a symbolically
concealed gallop into statism.
Jack D. Douglas (1989)

In the 1950s historian Arthur Schlesinger, Jr., dismissed conservative

political activism as merely "the ethical afterglow of feudalism." In the
1970s he reacted to the resurgence of conservatism with the confession
that "no intellectual phenomenon has been more surprising in recent years
than the revival in the United States of conservatism as a respectable social
philosophy" (Rusher 1984, p. 307).
Max Lerner responded to Schlesinger's incredulity by saying: "Conser-
vatism gave plenty of notice, if the liberals had kept their eyes and ears
open." In a column titled "The Earthquake," Lerner explained what made

possible the "long-preparing" earthquake that was the 1980 election of

Ronald Reagan and the Republican control of the Senate. That political
upheaval-the greatest since Franklin Roosevelt's victory over Herbert
Hoover in 1932-was "too convulsive" to be explained as a protest vote
against Jimmy Carter and inflation. Rather, the support for Reagan "ex-
pressed a wide range of discontents," "some deep, pent-up social angers,"
and represented "a long-range retreat from the liberalism of the New Deal
welfare state." It was part of a conservative-led revolt by middle-middle
and lower-middle class achievers of the American dream who felt threat-
ened by the forces that seemed intent on taking this away from them." They
were reacting not only to economic distress but were also motivated by a
"reawakened religious impulse and the concern about the work ethic and
about the endangered value system" (Rusher 1984, pp. 307-8).
Max Lerner was correct in seeing the resurgence of conservatism as a
populist revolt; he was correct in his characterization of the social classes
involved (see Danzig 1962; Reider 1985). However, he was incorrect to
suggest that a class revolution was taking place. He was also incorrect to
interpret Reagan's election as representing America's retreat from New
Deal liberalism. Although the New Right activists contributed substan-
tially to the election of Ronald Reagan to two terms as President, at least
three factors made it impossible for their influence to result in a retreat
from the New Deal welfare state. First, the practical expression of the New
Right's populism requires the welfare state. Second, quite apart from the
institutional requirements of populism, Reagan himself endorsed the major
surviving programs of the New Deal, as candidate and as president (Niskanen
1988; Glazer 1988). Third, Reagan's governance was influenced less by
the New Right's agenda than by constraints of the federal bureaucracy and
Congressional politics, as well as tensions within conservatism and be-
tween conservatism and populism. If anything, the direction of influence
was in the reverse. That is, although rhetorically reinforcing conservative
hostility to government, Reagan also caused conservatives to come to
terms with the welfare state. It is arguable that his greatest impact on the
conservative movement was its gradual accommodation to the realities of
politics and government (Morganthau 1987a).
Although Reagan symbolized the beliefs and values of the New Right, the
movement's effect was greater as a countermovement to secular and religious
liberalism than as a positive influence on the Reagan Administration. A review
of policy decisions will show that liberal dominance over the social agenda was
not significantly diminished. New Right activism exposed the contradictions of
utilitarian liberalism but it has not hastened the decline of the liberal ethos (see
Gilbert 1983; Mishra 1984; Douglas 1989). To be sure, the New Right has
had some political impact, but primarily at the state and local levels.
AnneWortham 209

I shall discuss these issues by bringing the findings of several research-

ers and conclusions of policy analysts to bear on my argument that (1) the
Reagan Administration did more to serve the cause of the forces they
opposed than they did to serve their own as rhetorically stated, and (2) the
New Right failed to attain the power and influence it sought in large
measure because it could not overcome the demands of modernity.


The New Right is a network of activists, local single-issue and religious

organizations, and constituencies linked by protest demonstrations, elec-
toral campaigns, registration drives, petition drives, ballot initiatives,
lobbying activities, and conferences and newsletters that monitor and
analyze whether or not executive, legislative, judicial, corporate, and
media actions reflect its ideological tenets. But for the New Right's
creative use of modern media and communications technology, it repre-
sents only the latest version of three powerful cultural trends that recur in
American history and politics. (1) The ongoing populism of the southern
and western white lower middle classes; (2) the great one-issue moral
crusades of the American past such as the abolitionist and prohibition
movements; (3) the religious revivals that have swept across the land since
the middle of the 18th century (Phillips 1982, p. 49).
The roots of the secular single-issue groups are in the post-New Deal
conservative critique of statism, collectivism, and rationalism (Nash 1979).
The New Christian Right emerged during the 1970s in the context of the
broader New Right political movement. But even though the religious
Right made common cause with the secular New Right, its roots are not
in political conservatism but in Christian fundamentalism (Guth 1983).
What was new about the New Christian Right movement was its broad
sociomoral program that joined a range of single issues in an ideological
battle against secular humanism, the national scale of its constituency and
activities, and the large size of its membership (Liebman 1983).
Although the secular New Right and the New Christian Right are not
coextensive, there is sufficient overlap in their constituencies and activi-
ties to warrant the use of a single name to include both movements. Unless
otherwise specified, in this examination "New Right" refers to both the
secular and religious Right.
The key leaders of the New Right during most of the 1980s were:
Richard Viguerie, who founded direct-mail operations; Paul Weyrich of
the Committee for the Survival of a Free Congress; Howard Phillips of the
Conservative Caucus; the late John T. Dolan of the National Conservative

Political Action Committee; Jerry Falwell of Moral Majority; Jesse Helms,

founder of the Congressional Club; and Phyllis Schlafly of Eagle Forum.
The major themes of these groups are opposition to the Equal Rights
Amendment, to homosexual rights, abortion, no-fault divorce, sex educa-
tion, secular humanism, permissiveness, pornography, the use of drugs,
the Panama Canal Treaty, sanctions against South Africa, and, for the most
part, any kind of negotiation with the Soviet Union. They propose to
reverse these trends through parental control of education, including
selection of textbooks; voluntary prayer in public school; governmental
aid to "preserve the family"; tougher treatment of criminals; restriction or
elimination of trade with communist nations; aid to anti-Soviet movements
in Afghanistan, Nicaragua, Angola, and Mozambique (Pines 1982).
They champion the American flag, the free enterprise system, strong
national defense, a balanced budget, and limited federal power. Fueled by
antiestablishment, antibureaucratic, antiwelfare, antijudicial, antibusing, and
anti-affirmative action resentments, their overarching goals are to "set the
agenda of the debate," as Howard Phillips put it, to challenge the Left's
dominance over the political agenda, and to force the Left on the defensive.
New Right groups are distinguishable from ultraright groups such as the
John Birch Society and the Liberty Lobby, and keep them at arm's length.
Organizationally and tactically they are also removed from moderate Repub-
licans and old-guard free enterprise conservatives, whom they consider to be
too timid, too middle-of-the-road, too elitist, and too theoretical. Some writers
like Himmelstein (1983) claim that despite the organizational conflicts be-
tween the New Right and older conservative groups over tactics and turf, they
share the same ideology and goals. The New Right is certainly conservative,
but its populism and its desire to revolutionize political and social institutions
are not conservative. Despite some overlapping of Old Right and New Right
positions on such issues as taxation, military preparedness, and federal social
programs, the New Right represents a populist break with the Old Right and
is a radicalized conservatism (Crawford 1980; Lemann 1981; Phillips 1982;
Lipset and Raab 1978; Warren 1976).
The ideological distinction between the New Right and Old Right
becomes evident when their positions are identified in terms of disagree-
ments on the central assumptions of classical liberalism that provide the
basis and context for most political thought and public opinion in the
United States. To some degree all major ideologies in the United States
share the following basic tenets of classical liberalism: individualism,
instrumental view of the state, limited government, individual rights,
equality under the law, and representative government. Modern American
politics reflects two general areas of disagreement over the proper empha-
sis, interpretation, and application of these assumptions: (1) the extent to
Anne Wortham 211

which government should or should not intervene in economic affairs; and

(2) the extent to which government should or should not regulate individual
behavior in matters of morality and conscience (Maddox and Lilie 1984).
In their study of trends in public opinion on domestic issues since the
1950s, William S. Maddox and Stuart A. Lilie show that when American
attitudes toward these two issue dimensions are separated, four basic mass
belief systems are found to exist in the United States: liberalism, conser-
vatism, libertarianism, and populism. Modern liberalism is committed to
government intervention in economic affairs and the expansion of personal
freedoms. Libertarianism (or classical liberalism) supports expanded in-
dividual freedom but opposes government economic intervention. Conser-
vatism opposes both government economic intervention and expansion of
personal freedom. Populism opposes expansion of individual freedom but
supports government intervention in the economy. In this position, popu-
lism shares with liberalism support of government intervention in the econ-
omy, and it shares with conservatism opposition to expanded individual
freedom. The New Right's populism is reflected in its "hostility to estab-
lished interests, a desire to use government to enforce its own moral and
economic beliefs, and a greater distrust of classical liberal values than is
found among either liberals or conservatives" (Maddox and Lilie 1984).
As Maddox and Lilie's findings and statements by New Right leaders
indicate, the divisions between the economic conservatives of the Old
Right and the social and religious conservatives of the New Right, and
those between both conservative factions and the libertarians are not
merely over tactics, but concern ideology and policy (Weyrich 1982). A
significant example of the division between the Old Right and New Right
concerns the purpose of politics and governance (see Carey 1984). It is
expressed in the conflict between the social authoritarians (who include
some traditional Old Right conservatives and the social and religious New
Right) and the social libertarians (who include libertarians and some
libertarian-oriented Old Right conservatives). The social authoritarian
position is exemplified in a statement made in 1962 by Brent Bozell,
brother-in-law of William F. Buckley, Jr., and speech writer to Senators
Joseph McCarthy and Barry Goldwater. "The chief purpose of politics,"
wrote Bozell, was not the promotion of freedom but "to aid the quest for
virtue." He concluded that "the story of how the free society has come to
take priority over the good society is the story of the decline of the West"
(Nash 1979, pp. 176-77). Conservative journalist George Will agrees that
the highest possible purpose of government is "the improvement of the
soul of the nation" (Will 1983, 1989). In contrast, the social libertarians
believe that a just and moral society rests on personal and economic free-
dom (Hayek 1944; Carey 1984).

The debate between social authoritarians and social libertarians is sharpest

between traditional conservatives and libertarians, but their differences
are blurred when the Old Right rejects the particular form of social
authoritarianism espoused by New Right populists. A case in point is Barry
Goldwater's severe criticism of the New Right's use of political means to
obtain moral ends. In 1981 Goldwater, who thinks of himself as proudly
carrying the flag of the Old Right, declared that "the religious issues of [the
Moral Majority, prolife, and other New Right groups] have little or nothing
to do with conservative or liberal politics" (Miller 1981). Even though
Goldwater agrees with some of the positions taken by the New Right, such as
their opposition to abortion, he believes that the attempt to impose their moral
judgment on social issues risks compromising constitutional rights and splin-
tering the Republican party into "a wrecking crew of special interests as the
Democrats have done" (Goldwater 1988, p. 38). Sociologist Robert Nisbet, a
conservative also, protests the notion that the New Right's passion for moral
and military crusades has anything remotely to do with authentic American
conservatism. He condemns the Moral Majority's crusade to establish "a
moral-inquisitorial government well armed with constitutional amendments,
laws and decrees." (Nisbet 1986, p. 104).
Although libertarians can easily join Goldwater and Nisbet's condemnation
of the New Right's "moral-inquisitorial" politics, it is not a natural alliance.
The differences between all conservatives and the libertarians are deep and
longstanding (Carey 1984; Nash 1979). Their differences on the primacy of
individual rights was reflected in their clash over Ronald Reagan's nomina-
tion of Robert Bork to the U.S. Supreme Court in 1987. For most conserva-
tives the nomination was, in the words of Richard Viguerie, "the most exciting
news since the president's election" (Morganthau 1987b, p. 22) For libertar-
ians, who view Bork as a "moral skeptic," a defender of "populist majo-
ritarianism,' and a dilettante, his failure to be confirmed for the U.S. Supreme
Court was "by far the best political event of the Reagan years" (Richman
1987; see also Macedo 1987). Such differences and disputes are indicative of
the concern shared by all of the diverse groups that their distinctiveness does
not become blurred by their common, but hardly uniform, opposition to con-
temporary liberalism.


The secular New Right movement arose out of schisms within the
Republican Party that began with the right-wing opposition to President
Gerald Ford's selection in August 1974 of Nelson Rockefeller as his
interim vice president. The party's expectation that conservati yes would
Anne Wortham 213

support their old betrayer for the sake of "party unity" was, says Richard
Viguerie, "the last straw." Viguerie and his associates realized that if they
were to have any impact on the Republican party they would have to do
so outside the party (Viguerie 1982, pp. 27-28). Another "last straw" was
the Carter Administration's successful bipartisan campaign to ratify the
Panama Canal treaties that had been initiated by Presidents Nixon and
Ford. Although New Right conservatives were defeated by the adminis-
tration, they benefited by the addition of new names of potential supporters
acquired by the direct-mail campaign directed by Richard Viguerie and
others. By the election of 1980 Viguerie had reportedly acquired the names
of 4.5 million conservatives drawn from single-issue and multi-issue
groups. It was this mechanism, more than any other, that enabled the New
Right to operate apart from the major parties (Viguerie 1981).
The politicization of the religious Right was precipitated by a proposed
Internal Revenue ruling that required any private school to prove, on pain
of losing its tax-exempt status, that it was not founded to avoid racial
integration (Rusher 1984, p. 297; Skerry 1980). For fundamentalist leaders
this was government intimidation of private Christian schools. They were also
shocked by Supreme Court decisions that in 1963 outlawed prayer in the
public schools and in 1981 the posting of the Ten Commandments on the
classroom wall.
As Pines (1982, pp. 17-18) reports, other groups were "pushed to the
brink" at different times and in different ways, "most incrementally and
others by a single precipitating event or development." For some business
leaders, the brink was the congressional attempt to create a Consumer
Protection Agency. Parents and employers were shocked by the spectacle
of local public high schools graduating near-illiterates. Homemakers re-
sented the claim by the feminist movement that its radical agenda reflected
the views of all American women. Devout Christians and Jews were
intimidated and outraged by the "secular humanism" of those in control
of schools, courts, and other key institutions.
Feeling threatened and offended by out-of-control liberalism, by secu-
larism, permissiveness, and government intervention, these groups reacted
by mobilizing to recapture and affirm precious values, defend endangered
institutions, eliminate elitist rule by judges and bureaucrats, and restore
popular rule. Their activism was centered more around cultural defense
than around status defense as Upset and Raab (1978) have maintained
(Simpson 1983; Bruce 1988). They aimed to accomplish their goals by
building a coalition of social, religious, and fiscal conservatives that
would become the philosophical alternative to liberalism and Old Right
conservatism and the political alternative to Democrats and Republicans.
This meant affirming their loyalty to old-guard conservative tenets of free

enterprise, strong national defense, limited federal power, and balanced

budgets. However, although they joined the old guard in its opposition to
economic liberalism, they were far more upset with liberalism's moral and
cultural tone than with its effect on free-market economics (Phillips 1982).
Vanguard organizations like Moral Majority, Christian Voice, and Re-
ligious Roundtable took as their article of faith the warning of Proverbs
29:2, that "when the righteous are in authority, the people rejoice, but
when the wicked beareth rule, the people mourn" (Pines 1982, p. 295).
According to Paul Weyrich (1982), who sees a natural linkage between
political conservatism and religion, "The very essence of the New Right
is a morally based conservatism. As a matter of fact, our view is not based
in economics but in a religious view."
One of the lessons activist conservatives and New Right leaders learned
from the liberals and the Left was that in order to influence public policy
they needed more than just a large number of dissatisfied individuals. They
needed to mobilize those people into a social movement, and they needed
power resources-for control over resources can be translated into the
power to make decisions associated with those resources (McCarthy and
Zald 1977). While coming to terms with politics, says Viguerie (1982, p.
30), New Right groups learned what liberals had long known: "that
single-issue politics is only an aspect of coalition politics. We learned to
organize, lobby, and bargain just as our opponents had been doing for
many years." They began to exploit the media and establish their own think
tanks, public interest legal institutes, grass roots organizations, and lob-
bying efforts. On the premise that "politics is actively in relation to
power," and that "power is located in many places" (Weyrich 1982, p. 58),
they replicated the tactics of litigation, regulation, and legislation em-
ployed by liberal organizations and movements. And as they plunged into
political action, secular leaders like Howard Phillips gave pep talks to New
Right audiences around the country, reminding them: "Power is the ability
to set the agenda for the debate" (Pines 1982, p. 306).
In mixing pulpit and politics, leaders of the New Christian Right could
find examples in Martin Luther King, Jr.'s, leadership of the civil rights
movement, in William Sloane Coffin's opposition to the Vietnam war, and
in the use of Congressional voting indexes by such groups as the Friends
Committee on National Legislation. As they worked to translate moral
values into public policy, they took heart from a growing literature that
fostered the idea that religious freedom, as well as political and economic
freedom, is primarily a religious achievement. If mainline liberal denom-
inations could claim civil rights legislation as their achievement and play
a visible role in the anti-Vietnam movement, then the conservative clergy
Anne Wortham 215

could claim antiabortion legislation and the elimination of secular human-

ism from the schools as their achievement. "The idea that religion and
politics don't mix was invented by the Devil to keep Christians from
running their own country," Jerry Falwell told Rolling Stone (Bruce 1988,
p. 152).
Their mobilization efforts paid off. Phyllis Schlafly's anti-ERA cam-
paign succeeded in halting ratification of the equal rights amendment. In
1980 they helped the Republicans to capture the White House and the
Senate. A few New Right activists were appointed to very junior positions
in the Reagan Administration. Fundamentalist pastors were invited to
breakfast with the Administration officials and cabinet members. The
Administration declared a "year of the Bible." By the 1984 presidential
campaign, they had the clout to push the Republican Party in their direc-
tion. By and large the Republican platform consisted of the New Right
agenda. "It's just like I wanted it," said Jerry Falwell (Dillin 1984, p. 1).
Reagan's support of school prayer, antiabortion campaigns, and cre-
ation science helped to give the impression that the New Right's goals
were serious and not the ravings of religious fanatics (Bruce 1988, p. 154).
Republican moderates worried about the impact of Reagan's support of
the New Right agenda, and were distressed by the 1984 Republican
Convention's adoption of a platform that consisted primarily of the New
Right agenda (Dillin 1984). But they need not have been concerned. The
New Right's influence on the Republican party was ultimately only sym-
bolic. They failed to achieve leadership status either within the Republican
party, the Reagan Administration, or the wider conservative intellec-
tual/political movement (see Anderson 1988; Bruce 1988; King 1987). As
King (1987, p. 161) points out, although Reagan's endorsement of the New
Right agenda paid electoral dividends, the primary electoral attraction was
the Reagan Administration's economic arguments, not the causes of the
New Right. Their social conservatism provided useful explanations and
rationales for the consequences of liberal economic policies, but that was
a secondary role. The most influential of the factions that made up what
Jeane Kirkpatrick called the "Reagan Phenomenon" were the neoconservative
intellectuals and the libertarian and conservative supply-side economists.
The absorption of the New Right into the Republican party was a victory
in that it gave status to activists like Falwell and thereby legitimacy to
their positions. But the Republican party's acceptance of the New Right
cost it very little. As Bruce (1988, p. 162) points out, since the Republican
Party is not the majority party, the symbolic gains for the [New Right] of
becoming a faction within it are offset by the fact that it is the losing party.



The mobilization of New Right groups revealed their politics to be more

populist and activist than doctrinally conservative (Phillips 1982, p. 48;
Viguerie 1983). The constituency of the movement has largely a New Deal
economic past (Maddox and Lilie 1984). Their heroes are not Adam Smith or
Edmund Burke, but populists such as Andrew Jackson, William Jennings
Bryan, Franklin D. Roosevelt, and George C. Wallace. Ronald Reagan is also
more populist than conservative. Despite the fact that Reagan has been
preaching the gospel of political and economic conservatism for three de-
cades, says Robert Nisbet, "[his] deepest soul is not Republican-conservative
but New Deal-Second World War Democrat. Thus his well noted preference
for citing FDR and Kennedy as noble precedents for his actions rather than
Coolidge, Hoover, or even Eisenhower" (1988, p. 104).
The populist strand in New Right conservatism is most salient when its
concerns are articulated as integral elements of the American Dream that has
shaped the American national character. In that context, American populism
is an expression of faith in the common sense of the ordinary person; a
preference for action over reflection that is antitheoretical, antiexpert, and
anti-intellectual; stress on practicality and immediacy, the ethic of the mar-
ketplace, and "street sense" (Bums et al. 1985, pp. 563-64). Reverence for
the common person explains in part the ambivalence that the New Right
shares with most Americans toward power, politics, and governmental au-
thority. That ambivalence consists of a strong identification with the United
States as a national community juxtaposed to the negative view of "govern-
ment" and "politics" (Bellah et al. 1985, p. 250). New Right populism is also
animated by the exceptionalism that sees Americans as a select people singled
out for a special destiny to, in the words of Woodrow Wilson, "show the way
to mankind in every part of the world to justice and freedom and liberty"
(Bums et al. 1985, p. 564; see also Handy 1976). Whether one accepts this
righteous sense of mission or not, there is no denying its persistence in the
discourse that governs what Henry James spoke of as "the complex fate of
being an American" (Donoghue 1988, p. 230).
As a conservative advocate with roots in the New Deal, Ronald Reagan
was the embodiment of these elements of the American Dream. "He under-
stood Middle American folklore and myth very well," writes Laurence I.
Barrett (1989).

When he declared that "government is not the solution to our problem; govern-
ment is the problem," he appealed to his countrymen's primordial suspicion of
authority. When he talked of God's plan for American freedom, he revived the
Anne Wortham 217

nation's self-image as uniquely blessed. When he inveighed against tax rates, he

playedon Everyman'sresentment againstthe burdens of the commonweal. (p. 15)

Reflecting his own paradoxical sentiments of distrust in American politi-

cal institutions and pride in the nation, Barry Goldwater (1988, p. 391)
praises Reagan for carrying the banner of populism-albeit, a conservative
populism-as "something that no one in the nation has accomplished since
Teddy Roosevelt."
Myths and dreams like the American Dream help to hold a society
together, to maintain its sense of integration and consensus (Durkheim
1960). Moreover, they can be more powerful than logic, and shrewd
political leaders attain and maintain power by appealing to them (Burns
et al. 1985, p. 562). Whether or not Reagan consciously recognized the
relation of the American Dream and power, the populist idealism and
moralism that the dream entails were certainly central to his politics and
the interests of the New Right. Of course these qualities of national
identity (as distinct from ideological populism) are not unique to conser-
vatism, but cut across the different mass belief systems in America (see,
e.g., Boyte and Riessman 1986). As a master of what Barrett (1989) calls
"genial symbolism," Reagan's credibility was enhanced by his belief in
his own rhetoric, and in himself. Even Reagan's harshest critics acknowl-
edged the integrative function of his leadership. So, too, have more
dispassionate observers like political scientist Hugh Heclo (1986) who
writes that, "The ideas embodied in the political movement Reagan sym-
bolized will have helped organize a previous incoherence in our public
conversation about the meaning of American national society" (p. 60).
Reagan not only put the ideas of the American Dream to work in the
service of the cohesion, commitment, and coherence that are essential to
a dynamic political culture; he also restored the public morale that had
been flagging during the 1960s and 1970s. And as his populist rhetoric
reassured the citizenry, he lent legitimacy to New Right conservatism as
George McGovern ("The Prairie Populist") did for New Left liberalism.
However, the ambivalence toward authority and the American exceptional-
ism he shared with the movement and millions of other Americans could
not be translated into many social gains for the New Right or political
triumphs (other than election victories) for his Administration.


When the New Right movement is viewed in terms of the paradoxes of its
populism rather than merely as the continuation of Old Right conservatism

of the 1950s and 1960s, several unintended benefits to New Deal liberal-
ism become apparent. One was Reagan's contribution to increased public
confidence in government, in which historian Robert S. McElvaine (1986,
p. A35) observes a dialectical irony. Reagan's antigovernment rhetoric
(though not most of his programs) and his geniality as head of government
restored the people's faith in government, and to that extent served con-
temporary liberalism's belief that "affirmative" government can better
people's lives. The effect, says McElvaine, was that Reagan's seeming
success (prior to the Iran-Contra scandal) damaged the ultimate conserva-
tive objective of instilling in people a skeptical view of state power.
Reagan also benefited liberalism by doing very little to change the basic
structure of the welfare state. As Nathan Glazer (1988, pp. 56-57) observes,

The administration made no head-on assault on the idea ... that government had
the responsibility to maintain a safety net of basic social programs to provide for
the income needs of the destitute, and it announced its complete commitment to
social security to provide for the aged and disabled, unemployment insurance for
the unemployed, health insurance for the aged and destitute.

Far from being revolutionary, the Reagan administration's mixture of

conservative populism and New Deal liberalism was put in the service of
what William Niskanen (1988), former member of Reagan's Council of
Economic Advisers in 1981-1985, described as "a rather cautious evolu-
tion of policies supported by a broad bipartisan consensus beginning in the
late 1970s." For example, high on the Reagan Administration's agenda
was the policy of federalism, which was to be the mechanism for "disman-
tling" the welfare state. But the Administration's proposals to return to the
states such social programs as Aid to Families with Dependent Children
(Welfare) and food stamps remained only proposals (Glazer 1988, p. 118).
The contradictory impulses of welfare state conservatism were evident
also in the Administration's positions on education. During the 1980
presidential campaign Reagan called for the abolition of the Department
of Education, deregulation, and parental choice in education. Those goals
reflected a desire for decentralization and the reduction of federal control
of education. Yet Reagan's support of a school prayer amendment showed
a willingness to use federal legislation to bring about a particular educa-
tional result (Boaz 1988; Finn 1984; Uzzell 1985). The paradox in this
instance was not, as was the case generally, that reducing the state's
regulatory role actually required increased action by the state. Rather, the
contradiction was between noninterventionism and a specific interven-
tionist goal.
Anne Wortham 219

Another irony observed by McElvaine (1986) was that liberal pessi-

mism prepared the context in which Reagan conservatives could bring to
public life the optimism that could "produce a climate in which liberalism
could once again germinate" (p. A35). However, Reagan's optimism was
limited by his antibureaucratic outlook and his disdain for the "social
engineering" approach to social problems. Yet, it was precisely in the
limited nature of his optimism that liberals could find their own goals
served, says McElvaine. The dawn of Reagan's "morning in America"
could signal a new era of liberalism; for if "America is back," as the
Reaganites insisted, McElvaine quite rightly reasons that liberals could
ask: "If anything is possible, why tolerate hunger and poverty and the
threat of nuclear war?"
Although the Reagan administration attempted to redefine how govern-
ment should respond to social problems, it lacked the political courage to
implement the measures that were necessary to accomplish its ideological
convictions (Glazer 1988). Its lack of courage and paradoxical ideology
explain in part the uneven success of Reagan's economic policy. As William
Niskanen (1988) points out, the economic recovery was sustained longer than
usual, but average economic growth in the 1980s was about the same as in
the 1970s. The reduced growth of federal spending was offset by increased
absolute government spending, and the tripling of the national debt. The
reduction in individual and corporate tax rates was financed by the deficit or
by increasing the taxes on new investment. Some deregulation was offset by
a net increase in trade restraints. The rapid reduction in inflation was offset
by inflationary foreign aid policies. Niskanen argues that in the end, despite
the achievements of Reaganomics, "there was no Reagan revolution."
Many conservatives criticized the Reagan administration's failure to
diminish the force of New Deal liberalism. But other observers argue, as
Hugh Heclo (1986) does, that

it is not that Reaganism embraced all the prevailing ideas about the social
functions of government, but that by tolerating, marginally adjusting or simply
failing to reverse so many policies and attitudes, the Reagan experience unwit-
tingly served more or less to preserve the base of political argument and domestic
policy for future development. (p. 60)

Like Dwight Eisenhower, notes Heclo, Ronald Reagan was a "consolidator

of the American welfare state" (p. 60). His restoration of trust in govern-
ment energized both the "cynics on the right," who perceive government
to be too liberal on most social issues, and the "cynics on the left," who
want more social change (see Miller 1974; Citrin 1974).

The Administration's default on its commitment to decentralization and

noninterventionism was compounded by the New Right's approach to the
conservative critique of the welfare state as a threat to the family. The
welfare state's threat to the family is recognized by critics and supporters
alike. Conservatives and neoconservative critics call for a privatized
approach to family policy, wherein self-help and community volunteerism
are viewed as more conducive to self-responsible behavior and individual
freedom than the irresponsible rationality of the centralized welfare state
(Berger and Neuhaus 1977; Glazer 1988). This approach was consistent
with Ronald Reagan's celebration of the personal self-help ethos he
believed could still be found in families, communities, workplaces, and
voluntary groups. However, New Right critics used antiwelfare policy to
pursue a moral agenda that required the very blurring of the distinction
between private morality and public policy that had drawn them into
politics in the first place (Wuthnow 1983; Simpson 1983; Liebman 1983).
Their stated aim was to revive the traditional nuclear family and ensure
that responsibility for such personal issues as sexual and social relation-
ships were returned to the private family from the public sphere (David
1986). Yet, as self-appointed guardians of personal moral purity, they went
on the offensive by transforming to their own interests the tenet of their
liberal opposition that "the personal is political." As a result their activism
had the effect of expanding the boundaries of the pri vate domain of family
issues and shifting the emphasis of morality from self-responsibility to
social responsibility.
The impression that they wish to restore the separation of the private
and public realms is simply false. As Robert Nisbet (1986, p. 104) points
out, their fatuous use of the family to justify their tireless crusades to ban
abortion, to impose "voluntary" prayers in the public schools, and so on,
is no less statist than liberal family policies. They want to use state power
to substitute their own "pro-family" programs designed to strengthen the
traditional family and parental authority for what they believe are the
thoughtless actions and mistaken policies designed by "antifamily" forces
to provide safety nets and support. In doing so they switch the choice of
"free family autonomy or state intervention into family life?" with the
question of "which kind of state intervention?" The pressure of the com-
peting liberal and New Right pro-family forces certainly did not make it
politically expedient for the Reagan administration either to rescind liberal
family policies or to withhold its prestige and authority from New Right
legislati ve initiatives.
The Reagan Administration's failure to dismantle or reduce the welfare
state was a function not only of contradictions in conservatism and the
statist tendencies in New Right populism, but also a function of the crisis
AnneWortham 221

of the democratic-welfare-capitalist society (see Gilbert 1983). As Ramesh

Mishra points out in his interpretation of Lester Thurow's thesis of the
crisis of the "zero-sum society," the United States has reached a point
where further economic growth requires major structural adjustments in
which some groups stand to lose while others stand to gain. "But post-war
developments in the political economy of the welfare state have made it
increasingly difficult if not virtually impossible to impose the costs of
change on any particular group in society, at least not of the magnitude
necessary for economic growth to be feasible" (Mishra 1984, p. 146). The
impasse is between democracy and welfare capitalism. The problems of
the welfare state can be solved, says Thurow, but the solutions require that
someone bear the burden of large economic losses. However, no one wants
to volunteer for this role, and the democratic process is incapable of
forcing anyone to take it on (Thurow 1980, p. 11).
Although Thurow's solution-structural change in production, transfer
payments, progressive taxation, and a federal guaranteed job program-is
open to question, his illustration of "the complexity of the problem"
presents sociological and political realities of governance under welfare
capitalism that the Reagan Administration failed to appreciate. Whether
the solution to the crisis of the welfare state is free enterprise and privatiza-
tion, as Reaganomics proposed, or social democratic solutions proposed
by Thurow and others, the structural changes required cannot be brought
about if no one wants to bear the costs of change. The Reagan Adminis-
tration was unable to overcome the ability of special interests to use the
democratic system to resist the "dismantling" of the welfare state. Despite
the problems that plague the welfare state, its legitimacy remains intact among
the majority of citizens in capitalist democracies (Mishra 1984, p. 167).


Conservative and liberal analysts and commentators have almost uni-

versally agreed that the 1980 election of Ronald Reagan was a watershed,
that it was evidence "that America moved collectively to the right, defi-
nitely ending liberalism's long reign in the political sphere" (Rusher
1984). This unwarranted interpretation of American political culture has
been reinforced by a series of seemingly liberal-turned-conservative de-
velopments-for example, Governor Mario Cuomo's advocacy of nation-
as-family communitarianism, Betty Friedan's defense of the traditional
family, feminist advocacy of antipornography laws, Democratic support
of economic growth and cuts in federal spending, yuppies voting Republican.
Academic and journalistic debates on various facets of the theoretical,

empirical, and rhetorical "crisis of liberalism" were also seen as evidence

of the rightward shift of public opinion (see, e.g., Deutsch and Soffer
1987). However, all the handwringing over the legitimacy of liberalism,
the defensiveness of liberal politicians, and the public's rejection of left
liberalism do not translate into gains for conservatism in general or for the
New Right's specific agenda. As Seymour Martin Lipset (1988, p. A35)
reports, "Surveys indicate that majorities of the same electorate that voted
for President Reagan oppose most of his programs."
The lack of evidence for a general and significant shift to the right is
also supported by analyses of opinion poll data by Ferguson and Rogers
(1986) and Maddox and Lilie (1984). Ferguson and Rogers found that
opinion poll data from the 1960s to the 1980s show an increase in support
for social programs and government regulation of corporate profits (Bruce
1988, p. 166). Maddox and Lilie found that significant evidence since the
emergence of "social issues" (e.g., the right to abortion, legalization of
marijuana, women's rights, new protections for the rights of defendants,
and prayer in school) suggests that "significant minorities of the public
have supported the pure liberties position in every case over the last two
decades," and that "the small changes in public opinion over the decades
are in the direction of increasing support for individual choice" (Maddox
and Lilie 1984, p. 53). The significance of these findings is that neither
the New Right's successful fund-raising campaigns and their influence on
elections nor the media attention given to their activities were reflected in
the public's consistent support of personal freedom (Maddox and Lilie
1984, p. 54). Although these studies do not suggest that Americans have
become more liberal, neither do they provide reason to suppose that they
have become more conservative.
Indeed, one reason the New Right was unable to change public opinion
is that Americans are deeply divided over these issues. This is evident on
the economic front, where "for each well-publicized spending or taxing
cutback that is approved by voters, there is usually another being defeated
somewhere else in the same election" (Maddox and Lilie 1984, p. 56). This
has been no less true of the "social issues." For the most part New Right
successes on issues such as textbook and library censorship, prohibition
of pornography, and creationism in some areas of the country were offset
by defeats in other areas (Guth 1983). Successes at the state and local level,
such as statutes allowing school prayer, were overruled by Supreme Court
decisions. The reluctance of the justices to rule in favor of New Right
positions on sociomoral issues is evidenced by their failure to repeal Roe
v. Wade that legalizes abortion although ruling for restrictions on abortion
in Webster v. Reproduction Health Services. By 1989, in disregard of the
New Right's opposition to homosexual rights, 11 states and 80 municipal-
Anne Wortham 223

ities had laws prohibiting discrimination against homosexuals in such

areas as housing and employment. Throughout the 1980s, the media and
the entertainment industries increased their efforts to promote a sympa-
thetic approach to homosexuality in motion picture and television dramas.


As the crisis of the welfare state, the persistence of liberal policies, and the
contradictions of conservative populism deprived the Reagan administration
of an economic revolution, so the forces of modernity and competing defini-
tions of the meaning of the American experience deprived the secular and
religious Right of its "cultural counter-revolution," to use Jonathan Reider's
(1985) term. In his meticulously argued analysis of the emergence and decline
of the New Christian Right (NCR), Steve Bruce (1988) points to several
factors that undermined the ability of the NCR to achieve its goals. Many of
his observations are applicable to the secular New Right, especially insofar
as secular spokesmen and groups share the moral agenda of the NCR. The
following discussion relies heavily on the main points of Bruce's argument.
One of the NCR's strategies for resisting the consequences of modernity
was to attack secular humanism. Most of the things the NCR did not like
or felt threatened by shared the features of "secularity" and "humanism."
But the New Right cannot succeed in its goal of defeating secular human-
ism because most of what it means by secular humanism actually amounts
to consequences of modernity and necessary elements of a modern demo-
cratic society that "will not go away."
As Bruce points out, it is conceivable that a democratic society might reject
feminism, socialism, government controls on business activity; and dramatic
medical interventions connected with birth and death-all of which the New
Right designates as secular humanist practices. But it is not possible for
modern democracy to satisfy the New Right by prohibiting certain other
practices it characterizes as secular humanism: the exclusion of religion from
the public sphere; the extension of moral choice; and tolerance of alternative
sexual life-styles. For these practices are not the result of actions by secular
humanists as such. Rather, they are the consequences of the key components
of modem identity that Peter Berger calls the "pluralization of life-worlds"
and its "secularizing effect" (Berger et al. 1973, pp. 76-81).
Neither can democracy satisfy the New Right's demand that primacy be
accorded to prescientific ideas (such as creationism) over modern scien-
tific and technical knowledge. There is no reason why modern industrial
societies cannot permit the survival of premodern ideas in certain limited
areas of social life, says Bruce. But these ideas can only contest matters

of scientific interest by challenging the very knowledge base of modern

society (Bruce 1988, p. 188). And when that occurs, as when the New
Right demands that public schools teach "creation science" along with the
theory of evolution, scientists of the highest credentials reassert the bound-
aries of science, as 72 Nobel laureates did when the dispute over teaching
evolution reached the Supreme Court.
The Supreme Court rulings against the censorship of textbooks and
library books reflect similar aspects of modernity that obstruct the efforts
of New Right groups to eliminate "secular humanism" from the schools.
The manifest function of those rulings was the protection of First Amend-
ment rights. However, like the decision on "creation sciences," they had
the unintended effect of legitimating the authority of professionals whose
opinions carry more weight than lay opinions "even on matters such as
education, or the civil rights of the unborn or the terminally ill, where
technical considerations are obviously informed by moral judgements"
(Bruce 1988, p. 188). The subordination of the moral to the technical and
lay opinion to professional authority thwarts the efforts of populists "to
establish the principle that arguments such as that over the origins of the
species should be settled by votes rather than by the consensus of accred-
ited experts" (Bruce 1988, p. 189).
The New Right attempted to counter these consequences of modernity by
representing themselves as persecuted by secular humanists and demanding
that same principle of church-state separation that justifies the banning of
prayer in the public schools, the posting of the Ten Commandments in
classrooms, and the teaching of creationism should also justify outlawing
"godless" textbooks and prohibiting the hiring of homosexual school teach-
ers. By constructing secular humanism as a religion, fundamentalists could
claim that the failure to outlaw secular humanism in the public sphere is a
breach in universalism and discriminates against Protestantism. However, as
Bruce points out, their claim to a disadvantaged status cannot be remedied.
For it is precisely the principles of universalism in the public sphere and
particularism in the private realm that have produced the changes that offend
the religious Right (Bruce 1988, p. 190).
The NCR's failure, in its 1980s form, to re-Christianize America and
prevent further displacement of the values of its supporters was symbolized
by Jerry Falwell's renaming of Moral Majority, Inc., to Liberty Federation on
January 3, 1986. The new name would permit the organization to become
more openly political and, as Falwell put it, permit Protestants to "coalesce
with fellow Americans with whom they have theological disagreements for
the purpose of effecting moral and social change" (Bruce 1988, p. 183). In
terminating Moral Majority, says Bruce, and "by accepting the need to
separate religious values and socio-moral positions so that alliances can be
AnneWortham 225

formed with advocates of competing religious values," Falwell conceded

crucial ground to modern pluralism (1988, p. 190).
The final symbolic act of accommodation to modernity came when
Falwell announced on November 4, 1987 that he was resigning from the
presidency of Liberty Federation to rededicate himself to his ministry.
Thus the man who had taken credit for "breaking the psychological barrier
that religion and politics don't mix," and had thereby abandoned the
separatism that was fundamentalism's tradition, was now leaving politics
and thereby submitting to the religious pluralism that modernity demands.
The religious Right's accommodation to the demands of modernity re-
flects the inherent contradiction within Christian Protestantism between
its theocratic ambitions "to rule religiously" and religious individualism
and democracy. The unintended consequence of this is that Protestantism
is deprived of the position of setting the moral standards for the entire
nation and must remain "the partial world-view of a self-selecting minor-
ity of saints" (Bruce 1988, p. 174).


Ronald Reagan began his presidency by telling Americans, in the words

of Thomas Paine, "We have it in our power to begin the world over again."
Reagan ended his presidency believing, as he put it, that "A revolution of
ideas became a revolution of governance on Jan. 20, 1981." To be sure,
Reagan left the presidency, the federal government, and the nation in very
different states than they were when he came to office. But those changes
can hardly be called a revolution. As Reagan's claims indicate, social
change depends on political will and a change in prevailing ideas. To be
sure, the Reagan Administration gave high visibility to free-market ideas,
but it also encouraged the moral climate of populist Puritanism. Its rhetoric
alternated between quasi-libertarianism and conservative populism, but
the administration's policies were influenced by the right-wing social
democrats known as "neoconservatives," whose aim is to scale-down and
refine the welfare state, not to dismantle it.
Most of what the Reagan administration intended politically and eco-
nomically was not achieved. Most of what the New Right attempted
socially and morally did not succeed. The secular New Right declared it
would "set the agenda of the debate" and force the Left on the defensive.
The religious Right aimed to re-Christianize America and destroy all
vestiges of secular humanism in social institutions. The seeds of their
failure to achieve these goals were inherent in the paradoxical relationship
of Protestantism and democracy, the paradoxes of American populism,

and the demands of modernity. Instead of forcing the Left on the defensive,
they reinvigorated democratic populism (see Boyte and Riessman 1986).
Instead of imposing their morality on the country, they had to accommo-
date the demands of pluralism.
Ronald Reagan made Americans feel good about America, which was
an important achievement. But he also reinforced the hegemony of New
Deal pragmatism in American politics. His leadership was, in the final
analysis, one of continuity not revolution. Similarly, the New Right move-
ment was one of adjustment to the ongoing revolution in tolerance in
sexual behavior and race relations and the increased centralization that
accompanies it.


Anderson, Martin. 1988. Revolution. San Diego, CA: Harcourt Brace Jovanovich.
Barrett, Laurence. 1989. "Going Home a Winner." Time (January 23):14-18.
Bellah, Robert N., et al. 1985. Habits of the Heart: Individualism and Commitment in
American Life. Berkeley: University of California Press.
Berger, Peter L., Brigitte Berger, and Hansfried Kellner. 1973. The Homeless Mind: Mod-
ernization and Consciousness. New York: Vintage.
Berger, Peter L. and Richard J. Neuhaus. 1977. To Empower People: The Role of Mediating
Structures in Public Policy. Washington, DC: American Enterprise Institute.
Boaz, David. 1988. "Educational Schizophrenia." Pp. 291-303 in Assessing the Reagan
Years, edited by David Boaz. Washington, DC: Cato Institute.
Boyte, Harry C. and Frank Riessman, eds. 1986. The New Populism: The Politics of Empower-
ment. Philadelphia: Temple University Press.
Bruce, Steve. 1988. The Rise and Fall of the New Christian Right: Conservative Protestant
Politics in America 1978-1988. New York: Oxford University Press.
Burns, James MacGregor, J. W. Peltason, and Thomas E. Cronin. 1985. Government by the
People. 12th ed. Englewood Cliffs, NJ: Prentice-Hall.
Carey, George W., ed. 1984. Freedom and Virtue: The Conservative/Libertarian Debate.
Lanham, MD: University Press of America.
Crawford, Alan. 1980. Thunder on the Right. New York: Pantheon, 1980.
Citrin, Jack. 1974. "The Political Relevance of Trust in Government." American Political
Science Review 68(September):973-88.
Danzig, Daniel. 1962. "The Radical Right and the Rise of the Fundamentalist Minority."
Commentary (April):27-29.
David, Miriam. 1986. "Moral and Maternal: The Family in the Right." Pp. 136-67 in The
Ideology of-the New Right, edited by Ruth Levitas. Cambridge: Polity Press.
Deutsch, Kenneth L. and Walter Soffer, eds. 1987. The Crisis of Liberal Democracy: A
Straussian Perspective. Albany: SUNY Press.
Dillin, John. 1984. "New Right's 20-Year Rise to Power." The Christian Science Monitor,
August 27, pp. 1,30.
Donoghue, Denis. 1988. "The True Sentiments of America." Pp. 229-48 in America in
Theory, edited by Leslie Berlowitz, Denis Donoghue, and Louis Menand. New York:
Oxford University Press.
Anne Wortham 227

Douglas, Jack D. 1989. The Myth of the Welfare State. New Brunswick, NJ: Transaction
Durkheim, Emile. 1960. The Division of Labor in Society. Glencoe, IL: Free Press.
Ferguson, T. and J. Rogers. 1986. "The Myth of America's Tum to the Right." Atlantic
Monthly (May):43-53.
Finn, Chester E., Jr. 1984. "Our Schizophrenic Educational System." Wall Street Journal.
October 23, p. 28.
Gilbert, Neil. 1983. Capitalism and the Welfare State: Dilemmas ofSocial Benevolence. New
Haven, CT: Yale University Press.
Glazer, Nathan. 1988. The Limits of Social Policy. Cambridge, MA: Harvard University
Goldwater, Barry M. 1988. Goldwater. Garden City, NY: Doubleday.
Guth, James L. 1983. "The New Christian Right." Pp. 31-45 in The New Christian Right:
Mobilization and Legitimation. edited by R. C. Liebman and R. Wuthnow. Hawthorne,
NY: Aldine de Gruyter.
Handy, Robert T. 1976. ''The American Messianic Consciousness: The Concept of the
Chosen People and Manifest Destiny." Review and Expositor 73(Winter):857-66.
Hart, Benjamin, ed. 1987. The Third Generation: Young Conservative Leaders Look to the
Future. Washington, DC: Regnery Gateway.
Hayek, Friedrich A. 1944. The Road to Serfdom. Chicago: University of Chicago Press.
Heclo, Hugh. 1986. "Reaganism and the Search for a Public Philosophy." Pp. 31-63 in
Perspectives on the Reagan Years. edited by John L. Palmer. Washington, DC: Urban
Himmelstein, Jerome L. 1983. "The New Right." Pp. 13-30 in The New Christian Right:
Mobilization and Legitimation. edited by R. C. Liebman and R. Wuthnow. Hawthorne,
NY: Aldine de Gruyter.
King, Desmond S. 1987. The New Right: Politics. Markets and Citizenship. Homewood, IL:
Dorsey Press.
Lemann, Nicholas. 1981. "The Evolution of the Conservative Mind." The Washington
Monthly 13(March):34-41.
Liebman, Robert C. 1983. "Mobilizing the Moral Majority." Pp. 49-73 in The New Christian
Right: Mobilization and Legitimation. edited by R. C. Liebman and R. Wuthnow.
Hawthorne, NY: Aldine de Gruyter.
Lipset, Seymour Martin. 1988. "Americans Sneer at Liberalism. Why?" The New York Times.
October 28, p. A35.
Lipset, Seymour Martin and Earl Raab. 1978. The Politics of Unreason. Chicago: University
of Chicago Press.
Macedo, Stephen. 1987. The New Right v. The Constitution. Washington, DC: Cato Institute.
Maddox, William S. and Stuart A. Lilie. 1984. Beyond Liberal and Conservative: Reassess-
ing the Political Spectrum. Washington, DC: Cato Institute.
McCarthy, John D. and Mayer N. Zald. 1977. "Resource Mobilization and Social Move-
ments: A Partial Theory." American Journal of Sociology 82(May): 1212-39.
McElvaine, Robert S. 1986. "Why the Debacle Shouldn't Hearten Liberals." The New York
Times. December 9, p. A35.
Miller, Arthur H. 1974. "Political Issues and Trust in Government." American Political
Science Review 68(September):95 1-72.
Miller, Judith. 1981. "Goldwater Vows to Fight Tactics of 'New Right.' " The New York
Times. September 16, p. 1.
Mishra, Ramesh. 1984. The Welfare State in Crisis: Social Thought and Social Change. New
York: S1. Martin's.

Morganthau,Tom. 1987. "Reagan: Eyes Right." Newsweek, March 30, pp. 20-23.
Morganthau et al. 1987. "Trying to Leave a Conservative Legacy." Newsweek, July 13, p.
Nash, George H. 1979. The Conservative Intellectual in America Since 1945. New York:
Basic Books.
Nisbet, Robert. 1986. Conservatism: Dream and Reality. Minneapolis: University of Minne-
sota Press.
Niskanen, William A. 1988. Reaganomics: An Insider's Account of the Policies and the
People. New York: Oxford University Press.
Phillips, Kevin P. 1982. Post-Conservative America: People, Politics and Ideology in a Time
of Crisis. New York: Random House.
Pines, Burton Yale. 1982. Back to Basics: The Traditional Movement That Is Sweeping
Grass-Root America. New York: William Morrow.
Reider, Jonathan. 1985. Canarsie: The Jews and Italians of Brooklyn Against Liberalism.
Cambridge, MA: Harvard University Press.
Richman, Sheldon. 1987. "The Majority vs. the Majoritarian: Robert Bork on Trial." Liberty
Rusher, William A. 1984. The Rise of the Right. New York: William Morrow.
Simpson, John H. 1983. "Moral Issues and Status Politics." Pp. 187-205 in The New Christian
Right: MobiliZlltion and Legitimation, edited by R. C. Liebman and R. Wuthnow.
Hawthorne, NY: Aldine de Gruyter.
Skerry, Peter. 1980. "Christian Schools versus the 1.R.S:' The Public Interest 61(Fall): 18-41.
Sobran, Joseph. 1988. "My Kind of Guy:' National Review (August 5):38.
Thurow, Lester C. 1980. The Zero-Sum Society. New York: Basic Books.
Uzzell, Lawrence A. 1985. "Contradictions of Centralized Education:' Cato Institute Policy
Analysis 53(May 30): 11.
Viguerie, Richard A. 1981. The New Right: We're Ready to Lead. Falls Church, VA: Caroline
---.1982. "Ends and Means." Pp. 26-35 in The New Right Papers, edited by Robert W.
Whitaker. New York: St. Martin's.
- - - . 1983. The Establishment vs. the People: Is a New Populist Revolt on the Way?
Chicago: Regnery Gateway.
Warren, Donald I. 1976. The Radical Center: Middle Americans and the Problems of
Alienation. Notre Dame, IN: University of Notre Dame Press.
Weyrich, Paul. 1982. "Blue Collar or Blue Blood?: The New Right Compared with the Old
Right." pp. 48-62 in The New Right Papers. edited by Robert W. Whitaker. New York:
St. Martin's.
Will, George F. 1983. Statescraft as Soulcraft. New York: Simon & Schuster.
---.1989. "How Reagan Changed America:' Newsweek (January 9):13-17.
Wuthnow, Robert. 1983. "Political Rebirth of American Evangelism." Pp. 167-85 in New
Christian Right: Mobilization and Legitimation, edited by R. C. Liebman and R.
Wuthnow. Hawthorne, NY: Aldine de Gruyter.
Part III

The Public Agenda: Addressing

High Priority Social Problems
How Families Manage Risk and Opportunity
in Dangerous Neighborhoods

Frank F. Furstenberg, Jr.

with the assistance of
Alisa Belzer, Colleen Davis, Judith A. Levine,
Kristine Morrow, and Mary Washington

In recent years, scholarly debate has centered on the size and sources of
the underclass. This effort to chart trends in persistent poverty has ob-
scured an observation undisputed both by those who believe and those who
doubt that the underclass is growing. Poor people, even those living in
entrenched poverty. are not all alike. Indeed, differences among the un-
derclass may be as conspicuous and consequential as any commonalities.
Yet the conditions creating dissimilarity among the disadvantaged have
received comparatively little attention.
My interest in these conditions grew out of a 20-year longitudinal study
of teenage mothers and their offspring in Baltimore. The early years of
this investigation were devoted to understanding how early childbearing
shaped the life course of young mothers and their children (Furstenberg
1976). But as the study evolved, I became more curious about the different
patterns of managing premature parenthood than in the contrast between
early and later childbearers. In fact, the diverse outcomes of teenage
parenthood proved to be one of the most surprising results of the research

AUTHOR'S NOTE: This work has been supported by the MacArthur Foundation Research
Program on Successful Adolescent Development Among Youth in High-Risk Settings. Helpful
comments on the research design have been provided by the members of that Committee. I
would also like to thank Jane Ballin, who contributed valuable insights as the project
unfolded. Some ofher interview data has been incorporated in this chapter.


(Furstenberg et al. 1987). What was true for the mothers seems to be no
less true for their offspring. To be sure, early parenthood seemed to have
real and lasting consequences for the children's prospects of educational
success and for becoming teen parents, themselves (Furstenberg et al.
1990). But many of the offspring of teen parents, even a number of those
whose mothers had remained on welfare throughout the study, were faring
reasonably well in early adulthood.
Intrigued by these findings, my collaborator Brooke Gunn and I spent
some time talking to some of the youth in the Baltimore study in hopes of
discovering conditions that promoted a successful response to disadvan-
taged circumstances. We were impressed in these conversations by how
many youth, who appeared to be doing well, had experienced close calls
that nearly resulted in serious trouble with the law, school dropout, or
teenage parenthood. They were not so very different from those with less
successful profiles. One conspicuous difference between the early "win-
ners" and "losers" was the significant role that family members played in
their children's lives. Parents (mothers especially) and sometimes other
close relatives (grandmothers, siblings, aunts, or uncles) often represented
the margin of difference between the successes and failures.
Consider Jerome, a 20-year-old male, who had managed to graduate from
high school and find a job in a seafood processing plant. During Jerome's
junior year of high school, he was suspended more than a dozen times.

I would get suspended like on a Wednesday. My mother would get me back there
by Thursday. That Friday I was suspended again but, no matter what, she was always
there. She would take days from work.... My mother was really what made me. I
got my diploma. I was proud for her. It was like she went to school over again.

It is not surprising that parental effectiveness is so key an ingredient in the

successful adaptation to poverty. Almost everyone believes that parental
competence matters for children's well-being. Countless studies, focusing
largely on middle-class children, have explored the dimensions of parenting
that seem to promote social and psychological development (Maccoby 1984;
Dornbush 1989). Furthermore, the few studies that have examined the deter-
minants of success among children growing up in poverty confirm what we
learned from the youth in the Baltimore study (Williams and Kornblum 1985;
Jeffers 1967; Clark 1983). And yet for the past couple of years, I have puzzled
about Jerome's story, which seems both so transparent and perplexing at the
same time. Jerome's mother was no less eager to see him complete high school
than many of the parents of children in the Baltimore study who dropped out.
Why are some parents able to marshal their efforts and others not? Jerome's
mother was unusually persistent and self-sacrificing. But other parents, no
Frank F. Furstenberg, Jr. 233

less persistent or self-sacrificing, were unable to achieve their aims. What

conditions enable parents living in poverty to help their children escape
disadvantage in later life?
I make no claims to answer these questions in this chapter. But thinking
about how to answer them has raised a whole set of issues that are not
ordinarily included in the standard approach to studying parenting processes.
Developmentalists are accustomed to regarding child rearing as a dyadic, or
at most, an interpersonal activity that takes place within families. Although
not wrong, this perspective is incomplete. It ignores the fact that styles of
parenting are an adaptation to conditions outside the household, specifically
by the social organization of the surrounding community. 1 I shall try to show
that the context in which parenting occurs influences both the parents' style
of management and their success in implementing their goals.
Social historians remind us that in past centuries the community's
involvement in the socialization and regulation of children was more
manifest (see, e.g., Demos 1986; Hareven 1984). Parents, as representa-
tives of the larger social order, shared child-rearing responsibilities with
other agents of society-members of their community, their church, and
their household. As the family became a more private and specialized
institution, individual parents increasingly became solo practitioners (Shorter
1975; Lasch 1977; Zelizer 1985). We continue to recognize that parents
benefit from community support (Garbarino and Sherman 1980; Reiss
1981; Bronfenbrenner et al. 1984). Nonetheless, parents have assumed an
ever-expanding share of the moral and economic burden for their children's
social development and success as adults (Ianni 1989).
The growing individualization of the parental role and its cultural
separation from community responsibility does not mean that parents
never rely on other agencies for help. Families, perhaps especially those
in the middle class, subcontract a lot of their parental duties to day care,
schools, summer camps, and private tutors (Medrich et al. 1982; Zigler
and Weiss 1985). It follows that a large portion of parenting involves
oversight of these external resources. How parents perceive, organize, and
manage the world outside the household has not been much studied by
sociologists or developmentalists interested in the socialization process.
With some conspicuous exceptions, most researchers rarely take adequate
account of the direct and indirect ways that parents manage (or fail to
manage) their children's involvement in the world outside the household. 2
These family oversight strategies may be no less consequential for children's
development than the more direct parenting practices observed inside the
home (Pratt 1976; Steinberg forthcoming; Medrich et al. 1982). This chapter
records some preliminary observations on the process of family management
in poor families. The focus on poverty families is not inadvertent. The

scarcity of community resources puts a high premium on parents who

know how to minister sources of social capital to their children. Jerome's
mother, who doggedly took on the indifferent school system, is a case in
point. After each of his transgressions, she badgered the school into
readmitting Jerome, both discouraging his attempts to drop out and re-
minding the school of their obligations to him. How well parents oversee
the external world depends in no small measure on parents' psychological
qualities. But I will also try to show that parents' management strategies
reflect their location in the surrounding community and the receptivity of
local institutions to parental initiative. This less visible feature of family
life-the reliance of parents on support from the larger community-has
intriguing implications for social policy. I shall return to this point in the
conclusion of this chapter.


This study is part of an ongoing research program on Successful Adolescent

Development Among Youth in High-Risk Settings sponsored by the MacAr-
thur Foundation. An effort to understand better the interconnections between
neighborhood and family, we launched a study in Philadelphia to see how
families manage opportunities and dangers in disadvantaged neighborhoods.
The research process is best described as "focused fieldwork," extended
open-ended interviews with parents and their teenage children residing in five
distinct inner-city neighborhoods. Beginning in the summer of 1989 and
lasting for nearly a year, several families in each neighborhood were visited
once a week or so. Family members were questioned about their views of the
community, particularly how they perceived resources, risks, and opportuni-
ties in their immediate locality. Whenever possible, taped interviews were
obtained. Whether or not conversations were recorded, the field-workers kept
a detailed log of their encounters with family members.
Frequent contact with the participating families provided some of the
benefits of ethnography; over time the field-workers came to be trusted
confidants of the parents and in some cases the youths as well. As the
involvement with the family deepened, it also broadened to include con-
versations with husbands and partners (when they were in the home),
extended kin, and neighbors. Over many months, each field-worker had
an opportunity to hear about and sometimes observe family members'
encounters with formal institutions and informal networks in the five
different communities.
Originally, I planned to contrast white and African-American families
living in areas of high and medium concentrations of poverty. When the
Frank F. Furstenberg, Jr. 235

opportunity arose (because one of the field-workers was fluent in Span-

ish), a Puerto Rican community was added as well. Within each commu-
nity I had hoped to study six families, half of whom were living in
relatively organized blocks and half on blocks that were less socially
organized. This design was quickly compromised by the formidable prob-
lems of making contact with families who were willing to be seen on a
regular basis and the enormous time it took to become intimately ac-
quainted with those who consented to participate. The study was scaled
back so that each field-worker became attached to only three families.
Ultimately we were forced to rely on what is best described as a catch-as-
catch-can sample, that is the families were obtained either by direct solicita-
tion or organization sponsorship. I make no claims that the resulting partici-
pants are representative of the communities in which they reside.
Nonetheless, we were able to see distinct differences across the communities
while recognizing that families in each area were quite varied. Some were
beset by problems and others were functioning remarkably well, especially
considering the enormous economic and social stresses that they faced. But
whether they coped well or poorly, there was some tendency to adopt
distinctive styles of managing the external world in different neighborhoods.
The field-workers routinely recorded instances of family management
both within and outside the home as reported or exhibited by parents and
their children. My findings accordingly are based on field notes supplied
by the interviewers and the discussions that occurred throughout the year
in weekly seminars where field-workers shared their experiences. It
should be clear that this chapter reflects the results of a team of researchers
rather than my work alone.
With the one possible exception, all the neighborhoods have recognized
names and at least vague geographical boundaries. Nonetheless, enormous
block-to-block variation exists within each neighborhood. Moreover, in
all but one community, family members had considerable difficulty agree-
ing upon the precise borders of their neighborhood. For example, when
asked to draw a map of their neighborhood, families in each neighborhood
generally displayed little consensus, as did individual family members. To
make matters even more complicated, the concept of a neighborhood is
much like the concept of a family; it changes according to use. For some
purposes, the neighborhood is the block on which the family resides or the
blocks immediately surrounding the family's residence; for others, it
encompasses a wider physical area that included shopping, schools, and
community facilities. Even so, the results of the field study persuaded us
that the neighborhood, despite its phenomenological elusiveness and its
geographical imprecision, operates as a significant constraint on parents'
approaches to and strategies of family management.


The data drawn from field notes were organized to illustrate the connec-
tion between neighborhood characteristics and family functioning. Ac-
cordingly, I am leaving out a great deal of relevant information about other
important determinants of the parents' style of managing the external
world. Within each neighborhood tremendous variation occurred in par-
ental competence-the term employed to recognize differences in parents'
child-rearing skills. Certain parents were more successful than others in
commanding respect, obtaining cooperation, transmitting and implement-
ing their goals. Often, these same parents had a greater capacity to cope
with high levels of stress, were more imaginative and persistent in finding
solutions to day-to-day problems, and were generally more positive and
easygoing when facing troubles. Many displayed a pronounced willing-
ness to sacrifice their own needs for the betterment of their children. Not
surprisingly, these more efficacious parents also tended to be more re-
sourceful in regulating their children's behavior outside the home and in
dealing with the formal and informal institutions in which their children
participated. They generally knew more about what happened to their
children when they were not in direct contact with them because of their
greater attentiveness and superior communication skills.
Field-workers had no difficulty discerning differences in parental com-
petence or noticing the effect of parental skill on the adolescents in the
study. Even allowing for differences among children within families, more
adept parents generally had more conforming children who were less
prone to problem behavior. But as the study progressed, we learned that
certain features of the social environment conditioned parents' managerial
skills. These same elements of neighborhood context enhanced or inter-
fered with parents' ability to implement goals for their adolescents: school
completion and the attainment of economic security, staying out of trouble
with the law, the postponement of childbearing, and avoidance of exces-
sive use of drugs and alcohol. The analytic task that I have set in this
chapter is to show the special ways that neighborhood context interacts
with different levels of parental competence to create different styles of
family management with, perhaps, varying outcomes for adolescents.
As a way of distilling the evidence, the discussion will initially feature two
families living in quite different circumstances. I have deliberately altered the
identities of the families to protect their anonymity, and sometimes merged
information from other families as well. The cases, then, can be regarded as
synthetic portraits of families we studied in different neighborhoods.
If they ever happened to meet, Lillie Gales and Meg Moran would
probably not recognize how much they actually have in common. Both are
Frank F. Furstenberg, Jr. 237

single parents with three children, the oldest of whom is a young teenager.
Both were receiving welfare at the beginning of the study although each
was making periodic efforts to get off public assistance. Both are con-
cerned parents who fervently want their children to have a better life than
they have been able to enjoy. Both worry about the problems that their
children are having in school, but their daily concerns are consumed as
well by the threat of physical violence in their neighborhoods, the prob-
lems of drugs, crime, and teenage childbearing. In fact, they are unlikely
ever to meet for Lillie, an African American, resides in a public housing
project in the heart of one of the poorest sections of North Philadelphia,
whereas Meg lives on one of the less desirable blocks of a small white
working-class enclave in South Philadelphia.



Lillie once lived in an impoverished section of South Philadelphia, only a

couple of miles from Meg's neighborhood, but moved to the Projects across
town several years ago to set up an independent household. She felt too tied
to her family who, according to her, were overly involved in her life. The
price of her independence has been high. Lillie now lives in an area that offers
practically no resources for parents or teenage youth. Funding for commu-
nity-based services has been severely cut. Neighborhood organizations have
withered in the process. Among those that remain, few offer social programs
for children, especially teens. Within walking distance of the Projects are
several public schools, a library, and a small shopping strip.
The Projects are dangerous. Drug-related murders on the premises or in
the immediate neighborhood have become almost routine. A guardhouse
being installed at the entrance of one of the buildings in the Project was
torn down before being completed. Community members believe this was
done by drug dealers who lounge at the entranceway to Lillie's building
in search of customers. Lillie and her children have learned to ignore the
entreaties of dealers on the streets peddling their wares. Lillie's involve-
ment with her building neighbors is minimal.

I don't know too many people here. I sticks to the neighbors. It's not too bad but it's
not too good neither. People don't know you-so they feel you out too. My kids-I
basically don't let them go down and play. When I take them out, we all go together.

Lillie does a certain amount of chaperoning of her children. She is able

to keep an eye on her younger son's performance in school because she

recently got a job as a parent helper. Meeting the other parents at school
did little to shake her views that other community people do not have much
to offer. She complains that they spend too much time in idle gossip and
expresses disdain for their values. Lillie believes her neighbors lack any
interest in bettering the community.
We found a lot of confirmation of Lillie's attitudes from other infor-
mants in the study. Leah, a long-standing resident of the Projects, echoes
many of the same sentiments about her neighbors and her neighborhood.
Leah agrees with a member of her household who declares in a conversa-
tion, "you can live in the projects, but you ain't got to be like the project."
When Leah was asked to draw a map defining the area where she travels,
she responded: "I don't go nowhere around here. I don't know nobody. 1
don't want to know nobody." The field-worker reports that, in fact, Leah's
acquaintance with her neighbors and neighborhood is much greater than
she admits to, but her pretense of ignorance is an accurate expression of
her feelings about the impoverished social world in and around the Pro-
jects. Among the residents there is a high level of superficial familiarity,
but investment in strong and lasting relations rarely occurs. A sign of the
relatively low level of social integration in the Projects is that residents
typically do not know the names of their neighbors, even some of those
that they had dealings with on an occasional basis. Leah's attitude, like
Lillie's, was to manage to survive until you can find a better place to live.
Says Leah, "Now I'm not going to make no mansion out of this place but
1 would want to live nice and comfortable until you leave .... 1 mean it's
still your home." She is not alone in her desire to move out, an objective
that undermines a sense of community.
The low level of cohesion in the Projects does not prevent neighbors
from entering into exchanges. Leah, for example, took in a young mother
and her child for a time during the study until the woman violated a
household prohibition on drug use. Lillie is characteristically more cir-
cumspect than many of her neighbors.

When some people be friendly, it's for something that you have that they want.
If they come and they knock on my door-and it's within reason-I give it to
them. Then after three times, I tell them, "Look. I don't mean no harm to you,
but I really feel I'm being taken advantage here."

The social networks that were formed around cooperation often seem
precarious. As Lillie's comment indicates, ties are often too shallow to
withstand the pressures of the constant barrage of claims and requests.
Resourceful parents generally segregate their children from the sur-
rounding community. Lillie attempts to teach her children to resist the
Frank F. Furstenberg, Jr. 239

local culture, encouraging them to feel different from (and better than)
their neighbors. Other parents, less confident of their ability to internalize
their values, maintain a tighter rein of control than Lillie. These parents
are extremely wary of entrusting their children to the care of others,
especially others in their immediate neighborhood. Indeed, the most adap-
tive pattern, quite unlike the collective pattern of parenting described by
Stack (1974), is a highly individualistic style offamily management where
parents devote enormous personal time to monitoring, supervising, and
controlling their children's behavior.
An employed neighbor of Lillie's insists that her teenage daughter call
her at work as soon as she comes home from school. And she checks in
hourly by phone until she comes home. Across the city, in another danger-
ous neighborhood in West Philadelphia, a parent of an l l-year-old boy
confesses, "Well, I know that I'm very protective of my son. I have to give
him more space so he can grow." Parents like these are reluctant to give
up direct control. They feel that they must insulate their children from the
neighborhood either by keeping them home or chaperoning them when
they go out. As the parent above suggests, this becomes more difficult and
less appropriate when children reach adolescence.
There are some safe niches within very dangerous communities. The
field-worker in North Park, a Puerto Rican neighborhood not far from the
Projects, volunteered at a local settlement house that was organized by
some of the parents in the neighborhood to combat the growing drug
problem in the area. But organizations, like this one, were unusual and
generally used only by the most active members of the community. Most
parents seemed either unaware of these protected enclaves for their chil-
dren or were unwilling to relinquish the responsibility they felt for directly
supervising them.
Maria, a resident of North Park, expresses misgivings about community
institutions. She criticizes many of her neighbors for being far too permis-
sive with their children and inattentive to the ever-present dangers of drugs
and drug-related violence. Maria insists that she know the whereabouts of
her children at all times and she is wary about entrusting them even to
community organizations. She recalls the time when her son had gone on
an after school trip: "They were supposed to leave at 12 and return by 3,
so supposedly he would be with the woman who takes care of him when I
got home from work." Her son wasn't home at 5, the time he usually
arrived home from the woman's house. At 6, when he was still not there,
Maria went to get him. Not finding him, she began "losing my mind."

It was already 6:30, almost 7:00 and my son still hadn't appeared. I couldn't even

Eventually her son returned by bus, the plans of the after school program
having been changed. "Why didn't you call me?" Maria asked the bus
driver. He replied, "I took all the kids home and no mother was like this."
Maria answered "I can't tell you what other mothers are like, but I'm not
like that. 1 don't like that my kids are walking around like that." Maria
now refuses to permit her children to participate in the program. Her means
of protecting her children from the dangers of the neighborhood now
resembles what has been called a "lock up" strategy. "My world is here,"
Maria remarks about her household, "sitting." And she goes on to say, "I
cannot permit things outside of me to damage me. If you don't have the
money to move to a better place, then you have to fight from the inside."
This tactic is costly, for it helps to sustain the social paranoia that prevents
many parents from banding together in anomie communities like the area
where Lillie, Leah, or Maria reside. The distrust of neighbors is understand-
able. Drug dealers hang around at the entrance to the Projects, forcing Lillie
to use the back door. Competent parents see themselves as a beleaguered
minority in their communities. A favorite stratagem for keeping their children
in line is negative comparisons-pointing out improper behavior of their
neighbors as a means of reinforcing their own values. This technique was used
in all of the communities that we studied, but the power of bad example is
omnipresent in the anomie areas. Ultimately, the message provided to the
children is that they are different from other families in their community.
Lillie's I5-year-old daughter, Davenna, comments:

Well, they were always pointing out bad examples of, you know, every neighbor-
hood. There's good things and there's bad things. And they showed me that that's
not the wayto go. And they might thinkthat I don't listen. But I do. I really do.

Davenna is so convinced of her dissimilarity from her neighbors, that she does
not tell her closest friends at school where she lives and refuses to bring
friends home to the Projects. Lillie understands, "I would like to say well
yeah, bring'em up. 1would like to meet them. But 1don't for the fact that the
way she feels about it. So I don't push the issue." Lillie finds herself in a bind.
Her strategy of encouraging Davenna to disengage from the community
lowers Lillie's ability to monitor Davenna's whereabouts outside the home.
The self-segregation of families in anomie areas does not entirely
prevent parents from banding together. However wary of their neighbors,
families find others like themselves. The best friend of Maria's son
happens to be the child of Cecilia, another women in our study. Cecilia is
considerably less skilled as a parent but she understands very well that
Maria is a good influence on her son. She encourages him to hang around
with his pal, Roberto. So parents employ an informal referral system to
Frank F. Furstenberg, Jr. 241

vet their children's friends, usually relying on their own direct knowledge
of the families. One unusually strict parent simply refused to pass on phone
messages from undesirable callers.
Relatives, when they are available, supply the safest companions or, at
least, they can become the connecting link to the other desirable friends.
Again, here is an account from Lillie's daughter, Davenna, about her
friends in her old neighborhood:

Davenna: There were a couple of people that were good examples, and we all
hung together. We were a crew.
Interviewer: How did you find each other?
Davenna: Well, my cousin. She was one of my buddy-buddies and she's going
to college.

In fact, Davenna does not have close relatives living in or near the Projects.
She goes to school outside the neighborhood. When she wants compan-
ionship, she often crosses the city to see her old friends and relatives. She
finds protection by maintaining links to another community that offers
greater resources. In the Baltimore study, we learned that families sent
their children to live with relatives outside of the city during the summer
to keep them out of trouble.
Limited social networks and scarce services in anomie neighborhoods
compel many families to seek sources of support outside their community.
There are simply too few functioning institutions within the immediate
neighborhood. Lillie explains why she goes back to the health clinic she
was using in South Philadelphia.

Now, see the clinic my kids go to, it's all bright and lively. Even if you is sick,
you don't be looking like I'm going to die. But that place (referring to the clinic
near the Projects) looked like the slaughterhouse.

The unavailability or unreliability of resources extends as well to infor-

mal systems of support as indicated above. Many families in socially
disorganized neighborhoods either had few kin living close by or, for one
reason or another, had weak ties with their relatives. Lillie has no family
living close by. Maria's extended kin were all back in Puerto Rico. Ida,
one of the parents in North Park, could not rely on her kin because virtually
all were involved in drug trade or were addicts themselves. She was forced
to cut herself off from her brothers and her brother's children as a means
of protecting her offspring.
The weak institutional supports in anomie neighborhoods do not present
an insurmountable barrier to the most resourceful parents. As I have

reported, they find ways to channel their children into the outside world,
that is, to establish and retain links with agencies beyond their immediate
neighborhood. These parents assiduously seek out after school programs,
summer camps, and the like that place their children in safe settings that
also provide opportunities for mobility. The ability to sponsor their chil-
dren may be limited, but resourceful parents are often unusually adept at
locating and cultivating people who sponsor their children. Channeling is
a political skill, but it is enhanced by a well-developed social network.
Relatively few of the parents in our study could lay claim to such a
network. Therefore, parents typically looked to the schools, the churches,
or community agencies to provide sponsorship. Occasionally, their expec-
tations were met when an interested teacher, guidance counselor, minister,
or coach discovered their child. Maria's son was selected in a city-wide
contest to participate in a youth forum for improving the schools. Several
of the children in the study were or would be attending magnet schools
that offered academically enriched programs for college-bound youth or
those with high technical skills.
Ida's youngest son, a child with some academic talent, was offered a
scholarship at an exclusive private school in the center of the city. She
never told the boy about the possibility, feeling that transportation would
be prohibitively expensive. In fact, her apprehensions about sending the
child outside the neighborhood were as much psychological as economic.
One ofthe most restrictive parents in the study, Ida was unprepared to see
her child so far removed from scrutiny. The task of avoiding dangers while
promoting opportunities involves a different sort of balancing act than the
one facing Lillie. Parents, like Ida, who attempt to isolate their children
from threats in their immediate environment run the risk of prohibiting
them from taking steps to escape poverty.
Most youth in our study were not among the lucky few to be discovered
by an adult sponsor. The number of "talent scouts" in neighborhoods like
the Projects or North Park is too small to provide a steady source of
referrals. And parents are frequently mistrustful of the institutions outside
of their community that might place their children in contact with adult
sponsors. Those parents unwilling to seek opportunities outside the neigh-
borhood usually place little confidence in local institutions as well. Leah
talks about her educational ambitions for her children but tacitly supports
their disaffection from school. Leah tells her children not "to hookey
school." But, in fact, what she really means is that they should not cut
school without telling her. She freely offers to cover for them when they
don't feel like attending. When her daughter was suspended from school,
Leah did nothing to get her reinstated. "That school is a trip ... it's not
like it used to be [when] they made you learn." Leah admits, "I don't know
Frank F. Furstenberg, Jr. 243

the system, but I am a concerned parent." Not surprisingly, Leah's children

are faltering in the school system.
There are striking contrasts among parents like Lillie and Leah or Maria
and Cecilia who display quite different styles of managing disadvantage.
Our study provides no clues to whether parents like Lillie or Maria are
fairly common or extremely rare. We can say that ordinary parents in
socially disorganized neighborhoods with limited resources have poor
prospects of implementing their goals. It requires exceptional parents, like
Lillie or Maria, to cultivate the sparse opportunities within their commu-
nity and to search out opportunities beyond the confines of their local area.
These individuals can be best described as "supermotivated parents," a
category that implies that ordinary parental motivation is insufficient to
create opportunities for children from disadvantaged communities even if
it suffices to protect them from danger. Otherwise competent parents lack
the extraordinary initiative, know-how, or social connections to channel
their children to opportunities in more remote areas. And sometimes, as
in Ida's case, their emphasis on protectiveness may actually limit their
children's chances of mobility by discouraging exploration in the wider
community or merely by confining youth to local institutions that can not
adequately develop their talents.
Whether they are highly skilled parents like Lillie and Maria or ade-
quately skilled parents like Ida or Leah, residents of anomie communities
like the Projects or North Park are inclined to adopt an individualistic style
of family management in negotiating the world outside the household.
Occasionally, they look for help in raising their children from formal
institutions like schools or settlement houses or for informal support from
their kin or friends. But for the most part, they are compelled to isolate
themselves and their children from the surrounding community. They
typically do not feel a part of their neighborhood institutions. They distrust
the schools, regard local services suspiciously, and, to the extent that they
use supportive services at all, take their business outside the community.
The family system is largely disconnected from the community, and
parents are left to manage on their own.


From all outward appearances, Meg Moran and her neighbors in Garri-
son Heights seem to share many of the same problems as families living
in the Projects or North Park. The Heights is inhabited mostly by white
working-class families, but many residents, like Meg, are quite poor. On
all sides, the community is surrounded by low-income black sections of

South Philadelphia. Meg and her immediate neighbors feel an intense

loyalty to their neighborhood. The sharp boundaries of the community
create a sharp sense of isolation from the rest of the city. Though only a
mile or two from Garrison Heights, most residents only rarely go to Center
City or to West Philadelphia, where the University of Pennsylvania is
located. These middle-class bastions seem worlds apart.
Like Meg or her sister Georgia who lives two blocks away, most
residents of Garrison Heights have grown up there. Indeed, many home-
owners live in houses first purchased by their parents or even grandparents
earlier in this century. Even relatively prosperous residents are discour-
aged from moving to more affluent sections of Philadelphia by loyalty to
the community. Those Heighters who do leave for better schools or bigger
houses complain about the absence of community attachment in suburban
areas. Says one informant:

A part of me always wishes I was somewhere else-particularly when I see what

my brother has. But who's to say that your better neighborhoods have all these
activities for the kids. My nephew wants to play basketball, and my sister-in-law
has to drive 25 minutes each way.

Those who stay sometimes complain about the tremendous burdens im-
posed by membership in the community-the high obligations placed on
residents, that everyone minds everyone else's business, the endless fam-
ily commitments, and the constant fear that the community will lose its
precious sense of sovereignty. Heighters boast of extensive kinship net-
works and lifelong friendships. All of the families that we observed in the
Heights had at least one grandparent living within walking distance and
all had other siblings nearby. One family in our study claimed to have more
than 70 relatives living in the immediate neighborhood that occupies only
about a square mile. "Well my kids," one informant reports "get a lot of
positive attention from their grandparents who live around the corner;
from my family-my parents and sister and brother-who usually see
them once a week." One of her husband's sisters is constantly buying toys
and clothing for her children.

More than I ever would.... And then we have to bail her out when she charges
so damned much that she can't pay the bills. But still, she's good to my children;
they know she cares a lot about them.

Better off residents frequently use surplus income to subsidize the rent
of relatives or to buy a home for a child, parent, or sibling. Meg's brother
helps pay for her housing. Her mother lives across the street and cares for
Frank F. Furstenberg,Jr. 245

her children almost as many hours a week as she does. Meg constantly
complains that her mother undermines her authority as a parent. But, as a
highly stressed single parent, it is difficult to see how Meg would function
without her mother's help.
Garrison Heights is one of a handful of urban villages in Philadelphia
that have survived successive waves of urban deterioration and gentrifica-
tion. It is easy to overlook the darker side of this community. But the
Heights, and other neighborhoods like it, are held together by a fierce
racism that helps contain the almost constant internal frictions and politi-
cal infighting within the community. Heighters are brought up from early
childhood to defend their territory from the encroachment of blacks, who
reside nearby. An informant, no longer living in the neighborhood, recalls:

the determiningfactor in the way people think about race relations is that people
think they're surrounded ... I remember having that consciousness at 6 or
7 ... I wasn't really filled with fear about that. That just was the way it was.
There was fear when ... violent incidents started to happen. You were kind of
obligated to beat them up. You had this sense that if you didn't beat them up,
they would be in all the time. We're talking about the mentality of 8- and 9-year
olds. This is how you proved yourself, too.

During the day nearby black residents shop at local stores, for the poorer
areas provide no commercial facilities. At night, they are virtually barred
from the area. Throughout the study, dozens of racial incidents took place.
And news of confrontations travels quickly throughout the Heights. The
morning after one such occasion, a 4-year old was able to recount in some
detail the violent events of the night before to the field-worker.
Heighters complain that outsiders are responsible for the serious drug
and crime problem in the community. However, in their descriptions of
the problems facing the youth of the community, Heighter parents describe
the dangers of drug use, delinquency, school dropout and teen parenthood
in terms that would be familiar to the residents of the Projects or North
Park. Complaining about the laxity of many families, Meg states "There's
a lot of that around here. Parents just don't care. And a lot of drugs and
alcohol. A couple of girls I know got their kids taken away from them."
The similarity ends here. For parents like Meg, Garrison Heights offers
a remarkable amount of support to adolescents. The well-developed insti-
tutional resources and the extensive social networks create an ample
supply of adults who assume responsibility for the children in the commu-
nity. Jane O'Brien moved back into the Heights some years ago to assume
an administrative position in the Catholic junior high school where her
daughter attended school. Within a few years, she was also teaching a craft

course for kids in the park (for her own profit). Her brother-in-law is well
connected in the city and has helped her arrange the funding for several
teen programs. Jane now employs several teenagers herself as counselors
in her summer program. She serves on the advisory committee of the
antidrug program and is active in the neighborhood association. Like many
other adults in Garrison Heights, Jane combines her work, community, and
family roles to mutual benefit. She is well connected in the school system,
dispenses summer employment, and often is in a position to exchange
favors with other parents.
Residents of the Heights complain that when they were children, adults
used to assume a lot more responsibility for children in the community.
This may well be true, but by contrast to the other neighborhoods that we
observed and, indeed to most middle-class communities, adults are ex-
tremely visible to children. Locals operate the community businesses, staff
the community institutions, and are in evidence on the street corners,
parks, and house steps. Meg may complain about the disintegration of the
family in the Heights, but in the same breath she says of her neighborhood,
"Somebody's always watching over the kids. Not just all my family here.
But just people I know on the block, in the neighborhood."
Heighters are vigilant about controlling their community. People expect
to know others in the community. The field-worker who worked in the
Heights was only permitted to approach families after she worked for a
drug prevention program in the community. During this volunteer experi-
ence, she was closely watched by parents to make sure that she would not
exert a negative influence on the youth of the community.

I was supervising the neighborhood clean-up on my own when all the neighbors
turned out on the street corners to watch (they were also watching, I think, to
help as well); these people don't really hesitate to discipline each other's children
and, I think, when an outsider is involved, all neighborhood kids represent the
whole neighborhood, so everyone assumes a stake in their behavior.

The field-worker first met Meg's 13-year-old son, Kevin, when she was
asked to help supervise a community-based work program for younger
teens in Garrison. Children as young as I I were provided with summer
employment to clean up the neighborhood-a program arranged through
the auspices of Jane O'Brien and other members of the community council.
This program is an example of the political largess received by Garrison
Heights from an influential City Councilman who represents the district
and began his political career there. The community has other powerful
allies. Heighters are good Catholics who benefit from church supported
schools, social services, and recreational programs that keep kids in the
Frank F. Furstenberg, Jr. 247

community busy. Kate, another resident of the Heights, describes the

week's activity for her two adolescents:

Monday nights has been basketball practice, Tuesday night practice, Wednesday
nights there is usually a game, and Thursday nights, there's peer counseling.
And, they are not supposed to be out on school nights. Well, they are never in
on school nights. But, at least they're not hanging out on the corner.

In fact, hanging out in the Heights is a major form of recreation for adults
and youth. But it, too, is a quasi-supervised activity as Heighter youth
generally frequent public places that are open to community scrutiny.
The field-worker describes meeting Meg one evening in the park when
she encounters a string band procession.

The street life here, the unbelievable density of people on the streets in the
evenings is strikingly reminiscent of a time gone by. The streets really become
the focal point of life in the summer, not just for kids, but for adults .... When
the mummers passed, adults didn't participate directly with the mummers ...
until they could grab a kid to do it with. Donations, too, were made the same
way ... "Here, Billy, go give the man a dollar."

Children are socialized to community values while being supervised by

parents, relatives, friends, and neighbors. Gates separate the park from
nearby housing projects. The field-worker notices that black children are
playing within sight. "But they are unsupervised, there are no teams,
swings, basketball courts, jungle gyms or drinking fountains, coaches, and
waves of welcome or unwelcome adult attention."
The rich supply of formal and informal community assistance has been
essential to Meg, who has had severe personal problems throughout most
of her own life. Meg is not an especially skillful parent, and her son has
experienced both serious behavioral and academic problems for the past
several years. It would be difficult for Meg to ignore Kevin's periodic
transgressions at school even if she were inclined to do so. Constant
reminders come in the form of notes from the teachers who complain of
Kevin's bad attitude. Meg has been called to the school on numerous
occasions. And a distant relative who is on the school staff provides her
with informal reports on Kevin's problems with his teachers. When Kevin
committed a minor act of vandalism at the school, Meg first heard about
it through her priest, who had been contacted by the school principal for
advice as to the best way of dealing with the situation.
Kevin's baseball coach, a longtime resident of Garrison Heights, is the
brother-in-law of Meg's sister. Parents show up at team practices and are

expected to accompany the team on trips outside the neighborhood. This is

but another way that parents come to know their children's friends--or, for
that matter, the youths who are not friends of their children. When Kevin
disappeared from home one evening, Meg was able to summon a couple of
his friends on the street to tell him that he better get back quickly or he was
going to be in serious trouble. Within 10 minutes Kevin returned home.
The close lateral links within generations allow parents like Meg to
receive invisible or barely visible forms of assistance. Kevin spends time
with his cousins, who are less troubled than he. They and their families
provide a kind of sheltered environment outside his often conflict-ridden
household. During a particularly difficult week at home, the field-worker
observed that Kevin ate dinner at his grandparents every night.
Parenting is a shared activity in Garrison Heights. Parents are, of course,
primarily responsible for their own children. And negligent parents-and
there are some in Garrison Heights-are severely chastised by their neigh-
bors. But parents look to friends and neighbors to enforce prevailing
community standards. They can delegate parental prerogatives because
they are confident that most members of the community share their views
of how Heighter children should be raised.
The high degree of perceived consensus in the community reflects the fact
that almost all residents of the community are known to one another. Most
are linked by kinship, friendship, or at least affiliated through one or another
community organization. They can reasonably expect that other adults would
act as they might if an adolescent gets out ofline. Accordingly, Heighter youth
cannot expect anonymity from the streets. It also means that children, like
Kevin, who do not have the benefit of a skilled parent have many other sources
of adult support. Here again is the field-worker's description of attending
baseball practice with Meg and her daughter Sally.

Formal and informal coaches almost outnumber the kids. T-ball for children 6
and under has 40 kids, uniforms, regular games and five grown-men coaches.
Outside the batting cage, benches are filled with family and friends of family,
all of whom know each other, look after each other's kids and so on.

In contrast to the Projects and even the Puerto Rican community of

North Park, males played a significant supervisory role in the Heights.
They served as block captains, as coaches, and were active in political
organizations. When present in the home, and most Heighter families were
couple headed, males exercised considerable authority even if sometimes
rules they set were often implemented by women. Adults in the community
recalled with a mixture of nostalgia and resentment the beatings that they
had received from their parents when they transgressed family rules.
Frank F. Furstenberg, Jr. 249

The rich supply of resources and dense network of adults that come into
contact with Heighter children serves as an effective system of sponsor-
ship as well as a source of social control. There is an abundant supply of
talent scouts in the form of priests, teachers, coaches, and the like who are
on the lookout for worthy candidates for academic scholarships and job
opportunities. Tommy Mitchel's father was a well-known figure in the
community-one of the "low-life" parents who come in for so much
community scorn. But Tommy, it seems, enjoys a different reputation. One
community leader comments, "He's the nicest kid you'd want to meet and
he's smart, too. He won the St. Matthews scholarship to Catholic High."
Tommy was given a leadership role on the youth council and received an
award (by a local dry-cleaning establishment) for good sportsmanship.
Just as important as the formal referral system is the informal system of
trading favors among parents for one another's children. Holly, one of the
residents of Garrison Heights, explained how her nephew received the
benefits of local sponsorship. He wanted to attend a local engineering
school but had failed to meet the deadline for application. The sister of a
close friend worked in the admission's department at a local university.
When appealed to for assistance, she arranged a special interview that
opened the way for the young man's entrance to the program.
Heighter parents participate in a system that promotes shared parental
responsibility through delegation of control and sponsorship to both for-
mal agencies and informal networks. The availability of resources, the
relatively high degree of normative consensus, and strong social bonds
forged by kinship and friendship all contribute to a close connection
between local institutions and the family. Parents feel responsible to the
schools and schools are responsible to parents. Youth cannot easily escape
the scrutiny of neighborhood, schools, and families when these agencies
are so interconnected. The high degree of observability keeps youth in
check. The task of parents inside the home is reinforced by the support
rendered by other parents outside the home. It is for this reason that
parenting in Garrison Heights must be viewed as a collective activity, a
style that distinctly contrasts to the individualistic mode of family man-
agement forced upon parents in more anomie neighborhoods.


The Projects and Garrison Heights represent contrasting types of com-

munities that give rise to divergent strategies of family management.
Obviously, most communities are neither as anomie as the Projects or
North Park, nor as socially cohesive as the Heights. Two of the neighborhoods

in the study fall more toward the middle of the continuum if neighborhoods
were arrayed from the most to the least organized and endowed with
resources. Both these communities are in decline, demoralized by a steady
attrition of neighborhood services and community spirit.
Kerryville is a changing neighborhood, historically inhabited mostly by
lower income whites but with a growing population of African Americans
and Puerto Ricans. It lies in the heart of a former industrial area of the city
that has been steadily deteriorating for some decades. In contrast to
Garrison Heights, many more of the residents of Kerryville are unem-
ployed, living in single-parent families, and on public assistance.
Maplewood, situated in West Philadelphia, is solidly African American.
Earlier this century, it was the first way station for Jews and Italians out
of poverty neighborhoods. Then, it became a haven for working- and
middle-class blacks. Like Kerryville, Maplewood in the past two decades
has witnessed an exodus of better off families. A core of employed
homeowners are trapped by their shrinking housing assets or by long-
standing obligations to older relatives.
In these two neighborhoods, Kerryville and Maplewood, we can see the
vestiges of social organization, recalling an era when schools, churches,
and local groups functioned as effectively as they still do today in Garrison
Heights. Residents of both communities bemoan the decline of local
institutions and local customs.

It's not like when I grew up, I remember all the neighbors would be sitting on
the steps. And if! did anything, they wouldall know I did it. "Ahh, you shouldn't
have done that. Your mom's right. You should be in for a week."

This same parent later comments, "We used to have a clean block commit-
tee, but that kind of fell apart.... They used to clean the school yard up
and everything. But they got tired of it."
The whites of Kerryville attribute the change to the incursion of blacks
and Puerto Ricans. Nancy, who has spent much of her life in poverty but
has recently found stable employment, is bitterly critical of her Latino

I didn't have anything. I'm just beginning to have anything. But my house was
always my home. I didn't trash it.... Why are their houses trashed? Cause they
put 15 people in there so, of course, it's trashed.

Maplewood residents attribute the decline of their community to loss of

the middle class, rising rates of unemployment, and the withdrawal of
government funding for programs and services. They recall the 1960s,
FrankF. Furstenberg, Jr. 251

when visions of the Great Society invigorated their neighborhood organi-

zations. An informant who moved to Maplewood in the 1960s recalls that
"a lot was going on" in the community then. Another resident describes
the feuding in the wake of the withdrawal of antipoverty funding and
laments that "nothing has picked up since then."
Receding resources in both communities is part of a downward spiral
whereby residents lose confidence in formal institutions, prompting many
families to flee the area. Consequently, a primary strategy of family
management in transition communities is geographical mobility. Says one
parent about the growing problems in the neighborhood, "I think this
neighborhood needs to be developed. The children and parents don't have
much direction. They get involved with drugs and violence. I'm distressed
by it. I'm not real comfortable here. That's why I'm considering moving."
Families, left behind, bitterly lament their failure to get out when they
could. Dave, the husband in one of the Kerryville families that we ob-
served, tells about his brother who lives a few blocks away in a particularly
bad section of the neighborhood:

Theyhaddrug housesright acrossthe street fromthem.They hadshootingsright

aroundthe corner.... Carsare gettingdestroyed left andright. I meanright there
that would have told me, "Yo,let's pack up and go." I can't afford to live up in
the Northeast, but he could have afforded it 10 years ago.

Dave remarks that his sister-in-law has been trying to persuade them to
move, but he says "It's not that bad around here." His wife, Millie,
disagrees "It's just as bad [referring to her sister-in-law' s old block]. You
just don't get out. You don't see it."
Families who remain behind, unwilling or unable to leave, bemoan the
breakdown of social ties and the loss of community consensus. Several of
the parents that we became acquainted with in Kerryville continued to
invest in local institutions. Their chronic complaint was that their efforts
at rebuilding the local institutions were thwarted by the demoralization
and apathy of their neighbors. Nancy, an activist in the community,
constantly complains about her neighbors' unwillingness to participate.
Field-workers in both neighborhoods report a pervasive sense that the
community is no longer "we" but "they." Kerryville parents feel that they
have been invaded by outsiders and deserted by insiders. Many now refuse
to allow their children to participate in recreational programs that attract
Puerto Ricans or African Americans, complaining that they will be ex-
posed to drugs or threatened by violence. Residents feel suspicious of their
neighbors, displaying the same sort of social paranoia that we observed in
the Projects. Nancy observes that every time her kids go out, she gets

hysterical. "I don't think they're safe anywhere. If I could, I would keep
them chained to the porch."
Maplewood families also perceive a sharp decline in active community
involvement. And they, too, experience the evaporation of social trust
described by their counterparts in Kerryville. Says one informant: "It's not
as easy as when I was a kid ... I've lived in West Philly for 22 years. We
used to be more open with each other." And then she added, "If you try to
talk to people, they think you are trying to get into their business."
More competent parents shift their attentions away from organizational
investment to channeling strategies of locating resources outside their
neighborhood. Liz Jones lives on one of the more stable blocks in Maple-
wood. She sends her daughter to a magnet school outside the area. Through
her work, Liz was able to find her a summer job in the suburbs, requiring
her daughter to spend more than an hour commuting each way. According
to Liz and her husband, they must guide their children to where the
resources are-out of their local community.
The disengagement of these competently managed families from neigh-
borhood organizations further hastens the decline of these local institu-
tions. The lack of community involvement further justifies the withdrawal
of outside resources, fueling a spiral of demoralization among residents in
both of these communities. Kerryville parents are especially critical ofthe
deterioration of order in the schools. They feel that their authority has
disappeared and that schools now refuse to support stricter parents.

I swear, I just feel like pulling them out of school. And I don't care if they ever
get their degrees. I just don't want them in school. I can't take it any more.

The decline of formal institutions also has the effect of weakening informal
social networks that sustain and are sustained by public arenas that bring
parents together. The social world of transitional communities contracts as
neighbors become strangers. Low participation at school meetings and neigh-
borhood improvement associations reduce the networks. Parents complain
that their neighbors are not really interested in cooperating with each other.
Here is Millie's account of the participation of parents in the community
athletic league: "Halfthe parents don't even show up to the games. They send
the kids and you're a baby-sitter." Then she adds, "It's everything. It's school.
You send your kid just to get him out of your hair, you know." And returning
to the subject of the games, she protests,

You have about three parents to take about 23 kids to a game. And some of the
parents had the nerve to come and then leave before the game was over; So you
had to take their kids home.
Frank F. Furstenberg. Jr. 253

This prompts the field-worker to comment, "Millie feels that she is fight-
ing almost alone to make the neighborhood a better place and she gets little
Parents who despair of the possibility that they can help to restore
schools, churches, or community services may nonetheless create informal
arrangements that provide collective support. One Kerryville resident, for
example, described how she had created a "rat line," a telephone network
to inform parents whose children were suspected of drug use. She reported
that parents were sometimes at first put out when they received a phone
call that their child was probably using drugs from a neighbor whom they
didn't even know. Most of the parents, she claimed, were grateful for the
interest of other neighbors. The rat line was an attempt to reinstate the
informal social control that had once resided in the community.
General doubts were voiced by many informants about the child-rearing
commitment and competence of their neighbors. How could they entrust their
children to parents who seemed incapable of managing their own children?
In fact, many of the participants in these communities were experiencing
moderate to severe problems with their children. Millie, whose oldest daugh-
ter has had drug problems in the past and is now a teen mother, condemns
other parents who, she feels, are too permissive. Parents in Kerryville and
Maplewood commonly complain about the subversive influence of their
children's peers who are poorly supervised by their families.
We see in transitional neighborhoods a shift from collective to individ-
ualistic strategies of family management. Fewer parents are willing to
delegate authority to formal institutions that have lost their credibility and
command of external resources. Informal networks become attenuated as
kin and close friends move out and are replaced by new residents who are
regarded as outsiders. The perception of normative consensus in the
community diminishes accordingly. As parents come to believe that their
standards for child rearing are not shared by others, they begin to distrust
other families nearby.
These changes at the community level help to bring about the disengage-
ment of the family from the community. The family becomes less and less
embedded in formal and informal structures that reinforce parental stan-
dards. Other adults in the community are no longer relied upon to supervise
and sponsor children. And systems of formal and informal control are
attenuated. As the family becomes separated from the community, parents
resort to more individualistic strategies of managing the external world.
They are disinclined to delegate authority to other parents or, reciprocally,
to assume responsibility for other children. Some parents in transitional
neighborhoods, like the women who started the rat line, try to develop
more cooperative arrangements but they are wary about relying on formal

institutions. Competent parents continue to look for opportunities to create

safe niches such as after school programs, church groups, or youth clubs.
But, as resources wither, they are forced to channel their children to
opportunities outside the neighborhood. They become more selective in
whom they are willing to entrust their children, relying exclusively on
close neighbors, good friends, and, of course, kin. And, a growing number
of parents begin to display practices more characteristic of anomie areas-
chaperonage and confinement.


The dynamic of neighborhood and family change occurring in transi-

tional neighborhoods recapitulates to an uncanny extent the historical
process of family withdrawal from the community. At one time, social
historians tell us, the family was far less differentiated from the surround-
ing community and from political, religious, and social institutions. For
the past century and more the family has come to stand alone as a more
independent agency, even as many of its functions were partially or wholly
assumed by other social institutions. One by-product of this change is that
parents have been left to assume a greater obligation for supporting their
own children and a lower commitment to investing in children other than
their own. Ideally, parenting has come to be viewed as a highly privatized
enterprise. Regardless of where they resided, virtually all parents that we
observed subscribed to the prevailing notion that good parents make good
children. In our culture, parents feel personally and almost exclusively
responsible for the way their children turn out. Yet at the same time,
parents recognize, sometimes only dimly, that their parental effectiveness
is conditional, at least in part, on the surrounding social milieu. Most
realized that their efficacy as parents depended to some extent on the
support that they received from schools, neighbors, and kin.
The results of our exploratory fieldwork seem to bear out the perception of
parents that conditions outside the family interact with parental competence.
Where parents live affects how parents manage their children-their means
of shielding their children from dangers and exposing them to opportunities.
By extension, the interplay between neighborhood and parenting process
probably affects their children's success in averting serious problems and
finding pathways out of poverty. The direct and indirect effects of neighbor-
hood on adolescent development remain to be explored in an ongoing longi-
tudinal study that builds upon this exploratory investigation.
Briefly, I have tried to show that a configuration of organization features
of neighborhoods-the availability of resources, the extensiveness of
Frank F. Furstenberg, Jr. 255

social networks, perceived normative consensus on parenting, and social

trust-all directly affect the support parents receive, either by reinforcing
or attenuating the link between family and surrounding formal and infor-
mal institutions. The connectedness or embeddedness of the family in its
immediate context shapes the strategies of parenting. The outlines of this
thesis are summarized in Figure 12.1. In neighborhoods like Garrison
Heights parenting is augmented and enforced by formal and informal
institutions. In the Projects or North Park it is largely left up to individual
parents to marshal whatever support and resources they can. To succeed,
I have argued that parents must be supermotivated, that is, exceptionally
adept at working the system and unusually diligent in monitoring their
offspring. Most parents are impelled to make desperate and often unsuc-
cessful attempts to exercise tight control by close monitoring or even
confinement. These strategies necessarily put higher priority on avoiding
the omnipresent dangers than cultivating scarce opportunities. I have also
tried to suggest how parents in transitional neighborhoods adapt to declin-
ing resources and growing anomie.
The dynamic described in transitional neighborhoods bears a certain re-
semblance to William Julius Wilson's (1987) description of the transforma-
tion of socially organized communities into so-called underclass areas. Our
observations support Wilson's description of community disintegration when
stable families flee distressed areas leaving behind a contingent of demoral-
ized parents to contend with a growing number of problem-ridden families. 4
Upon closer inspection, the highly varied nature of the population becomes
more evident. To the extent that they can, competent parents look for ways
of leaving or of managing as best they can if they are unable to escape. Even
in the most distressed neighborhoods, many parents are struggling to find a
way of protecting their children from the extreme dangers of everyday life.
Others, who may not be as capable of protecting their children, would function
reasonably well if they had the good fortune to reside in better organized areas
with more adequate resources.
I fully recognize that the empirical evidence presented here at best provides
slender support for the thesis that I have advanced. Along the way, I have
made no attempt to caution about drawing general conclusions since it seems
too obvious given the exploratory nature of the study. A more systematic
study of the interconnection of neighborhood and family management tech-
niques is presently underway in Philadelphia. This research should provide
more systematic data on the connection between the social organization of
neighborhoods, family functioning, and adolescent development.
If the observations reported here were to be confirmed by sounder
evidence, the findings have important policy implications for those who
advocate programs to strengthen the family or improve the situation of

Neighborhood Types Cohesive Transitional Anomic

Neighborhood Characteristics
rI Resources I Rich Declining Impoverished
ISocial Networks Extensive Limited Restricted
Perceived Perceived
HNeighborhood Norms I High Medium Low
Consensus Consensus
Social Trust I High Selective Low

Family/Neighborhood Linkages
I------- InstitutionaUFamily I High Partial Low
Connectedness I Embeddedness Embeddedness Embeddedness
~ Observability of
Youth by Adults
I High Scrutiny Partial Scrutiny Low Scrutiny

of Adults
I High Enforcement
Large Role for Men
Partial Enforcement
Some Role
Low Enforcement
Low Role

Family Strategies
Consequences for Collective Mixed Individual
Family Functioning Strategies Strategies Strategies
Family Strategies Delegation Mobility Confinement
Sponsorship Chaperonage
Community Participation Channeling
Community Building
Consequences for Greater Promotion---------- More Restricted
Adolescents of Opportunities Opportuni ties

Figure 12.1 Organizing Features of Neighborhoods

children. Much of our social policy emanates from our cultural ideal of
parenthood, an ideal that I have characterized as highly autonomous. We
sometimes act as if parenting is carried on exclusively in the face-to-face
interactions that take place within families. Many of our social programs are
designed to equip parents with better resources-material, informational, and
psychological-to manage the task of child rearing. In our ideology and our
services, we locate the full burden of caretaking on parents. And when their
children do not succeed, we assign them the full measure of blame.
Throughout this chapter, I have tried to show how parents are supported
or undermined by the immediate community beyond the household. Of
course, both competent and incompetent parents can be found in all kinds
Frank: F. Furstenberg, Jr. 257

of areas, but I contend that ordinary parents are likely to have more success
when they reside in communities where the burden of raising children is
seen as a collective responsibility and where strong institutions sustain the
efforts of parents. If this idea is correct, then if we are committed to
strengthening the family, we must give more attention to rebuilding local
institutions-schools, churches, neighborhood centers, and recreational
services-that support families. Doing so offers the best hope of creating
and nurturing informal social networks within neighborhoods that are
important to family functioning. The calamitous decline in resources for
schools, housing, recreational programs, and social services that occurred
over the past decade and a half has contributed importantly to the problems
that are faced by poor urban families. Rebuilding local community insti-
tutions may be a potent way of supporting beleaguered poor parents and
ensuring a better future for their children.


1. This point has been nicely developed by a small group of researchers who have had a
wider purview on parenting. See, for example, the excellent review essay by Steinberg
(forthcoming); the relationship of the family to the community is also the subject of a recent
special issue of Marriage and Family Review edited by Unger and Sussman (1990).
2. A notable exception is the seminal ethnographic research by British anthropologist
Elizabeth Bott (1957). Important connections between child rearing and neighborhood
patterns have also been observed by a number of field-workers such as Gans (1962) and Stack
(1974). This literature has been well summarized in a recent paper by Jarrett (1990).
3. This work also benefited from frequent exchanges with other members of the MacAr-
thur Committee on Successful Adolescence and members of the SSRC Committee on
Communities, Families, and Individuals, as well. I am especially indebted to the contribu-
tions ofthe Philadelphia Project: Tom Cook, Jaque Eccles, Glen Elder, and Arnold Saneroff.
4. For an interesting discussion of the conceptual and empirical challenges of testing
Wilson's hypothesis, see Tienda (1990).


Bott, E. 1957. Family and Social Network. London: Tavistock.

Bronfenbrenner, U., P. Moen, and J. Garbarino. 1984. "Child, Family, and Community. Pp.
283-328 in The Family, .. edited by R. D. Parke. Chicago: University of Chicago Press.
Clark, R. 1983. Family Life and School achievement: Why Poor Black Children Succeed and
Fail. Chicago: University of Chicago Press.
Demos, J. 1986. Past, Present, and Personal. New York: Oxford University Press.
Dornbush, S. M. 1989. "The Sociology of Adolescence." Annual Review of Sociology

Furstenberg, F. F., Jr. 1976. Unplanned Parenthood: The Social Consequences of Teenage
Childbearing. New York: Free Press.
Furstenberg, F. F., Jr., J. Brooks-Gunn, and S. P. Morgan. 1987. Adolescent Mothers in Later
Life. New York: Cambridge University Press.
Furstenberg, F. F., Jr., J. A. Levine, and J. Brooks-Gunn, 1990. ''The Daughters of Teenage
Mothers: Patterns of Early Childbearing in Two Generations." Family Planning Per-
spectives 22(2):54-61.
Gans, H. 1962. The Urban Villagers. New York: Free Press.
Garbarino, J. and D. Sherman. 1980. "High-Risk Neighborhoods and High-Risk Families:
The Human Ecology of Child Maltreatment." Child Development 51:188-98.
Hareven, T. K. 1984. ''Themes in the Historical Development of the Family." pp. 137-78 in
The Family, edited by R. D. Parke. Chicago: University of Chicago Press.
Ianni, F. A. J. 1989. "Providing a Structure for Adolescent Development." Phi Delta Kappan
Jarrett, R. L. 1990. "A Comparative Examination of Socialization Patterns Among Low-In-
come African-Americans, Chicanos, Puerto Ricans, and Whites: A Review of the
Ethnographic Literature," Paper prepared for the Social Science Research Council.
Jeffers, C. 1967. Living Poor: A Participant Observer Study 0/ Priorities and Choices. Ann
Arbor, MI: Ann Arbor Publishers.
Lasch, C. 1977. Haven in a Heartless World. New York: Basic Books.
Maccoby, E. E. 1984. "Middle Childhood in the Context of Family." Pp. 184-239 in
Development During Middle Childhood: The Years from Six to Twelve, edited by W.
A. Collins. Washington, DC: National Academy Press.
Medrich, E. A., J. A. Roizen, V. Rubin, and S. Buckley. 1982. The Serious Business of
Growing Up. Berkeley: University of California Press.
Pratt, L. 1976. Family Structure and Effective Health Behavior: The Energized Family.
Boston: Houghton Mifflin.
Reiss, D. 1981. The Family's Construction of Reality. Cambridge, MA: Harvard University
Shorter, E. 1975. The Making of the Modern Family. New York: Basic Books.
Stack, C. B. 1974. All Our Kin. New York: Harper & Row.
Steinberg, L. Forthcoming. "Communities of Families and Education." In Education and the
American Family: A Research Synthesis. edited by W. Weston. New York: New York
University Press.
Tienda, M. 1990. "Poor People and Poor Places: Deciphering Neighborhood Effects on
Poverty Outcomes." Paper prepared for the American Sociological Association, San
Unger, D. G. and M. B. Sussman, eds. 1990. Marriage and Family Review. New York:
Williams, T. and W. Kornblum. 1985. Growing Up Poor. Lexington, MA: Lexington Books.
Wilson, W. J. 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public
Policy. Chicago: University of Chicago Press.
Zelizer, V. 1985. Pricing the Priceless Child. New York: Basic Books.
Zigler, E. and H. Weiss. 1985. "Family Support Systems: An Ecological Approach to Child
Development," Pp. 166-205 in Children. Youth. and Families: The Action-Research
Relationship, edited by R. Rapoport. Cambridge: Cambridge University Press.
The Community Context of Violent Crime

Robert J. Sampson

Violent crime is a serious and apparently growing problem facing our na-
tion-particularly urban centers. Given the site of the 1990 annual American
Sociological Association meeting, it is appropriate to note that the Washing-
ton, D.C., murder rate reached an all-time high last year. Violent crime rates
in cities such as New York, Chicago, and Philadelphia are also running at
record highs (Hinds 1990; Recktenwald and Morrison 1990; James 1991).
Moreover, the leading cause of death among urban black males is homicide
(Fingerhut and Kleinman 1990, p. 3292), and the lifetime risk of being
murdered is as high as 1 in 21 for black males compared to 1 in 131 for white
males (U.S. Department of Justice 1985). As a sobering report in the New
England Journal ofMedicine recently revealed, a resident of rural Bangladesh
has a greater chance of surviving to age 40 than does a black male in Harlem
(McCord and Freeman 1990). Clearly, the time is ripe for a better understand-
ing of the roots of urban violence.
In an attempt to meet this challenge I advance in this chapter a commu-
nity-level perspective on violence. Although violence has not been an
under-studied phenomenon in American criminology, the study of the
causes and consequences of criminal violence in community context has
not been a major priority. Most of the more than 2,000 studies of violence
published since 1945 (see Bridges and Weis 1989, p. 14) have been
descriptive and focused on individual-level correlates of violent offend-
ing. As a result, extant reviews emphasize individual-level factors (e.g.,
age, race, and gender) or particular classifications such as family violence

AUTHOR'S NOTE: This chapter is drawn in part from a commissioned review, coauthored
with Janet L. Lauritsen, for the Panel on the Understanding and Control of Violence of the
National Research Council, National Academy ofSciences.


and criminal career violence. And at all levels of analysis the link between
offending and victimization in personal violence has been neglected
(Sampson and Lauritsen forthcoming). My goal is therefore to show how
a community-level perspective on both victimization and offending may
substantially increase our theoretical understanding of violence.
In keeping with the theme of the 1990 ASA meeting-"Sociology and
the Public Agenda"-I also believe a community-level framework has
much to offer in the way of public policy. The public agenda of crime
control is in particularly bad shape, guided by punitive and expensive
public policies predicated on false assumptions. Indeed, when millions of
federal dollars are spent on drug policies and incarceration programs that
ignore key pieces of knowledge on the structural and cultural sources of
violence, it seems time for alternative voices to be heard. In a nutshell, I
propose that a community-level perspective offers just such new insights
into constructive public policies, while at the same time exposing the
faulty premises and negative consequences of current programs.
To assess the implications of a community-level perspective on violence
for future research and policy, we must first come to grips with the empirical
foundations and limitations of past work. In a recent paper commissioned by
the Panel on the Understanding and Control of Violence of the National
Academy of Sciences, Janet Lauritsen and I reviewed in depth the empirical
literature on individual, situational, and community-level sources of serious
interpersonal violence (i.e., assault, homicide, robbery, and rape). As might
be expected, such a task was not easy and the result was rather long and
detailed. I Therefore, in this chapter I simply highlight the major continuities
running throughout past community-level research. Readers interested in the
detailed empirical foundation upon which this synthesis is based are referred
to the longer version which will be published as part of the Panel report
(Sampson and Lauritsen forthcoming).


The macrosocial or community-level ofexplanation asks what it is about

community structures and cultures that produce differential rates of crime
(Short 1985; Byrne and Sampson 1986; Bursik 1988). For example, what
characteristics of communities are associated with high rates of violence?
Are communities safe or unsafe because of the persons who reside in them
or because of community properties themselves? Can changes in commu-
nity structure affect violent crime rates? As implied by these questions,
the goal of macrolevel research is not to explain individual involvement
in criminal behavior but to isolate characteristics of communities, cities,
Robert J. Sampson 261

or societies that lead to high rates of criminality (Short 1985; Byrne and
Sampson 1986).
Within this framework I focus primarily on studies that examine neigh-
borhoods or other intraurban units of analysis. Cities and metropolitan
areas are large, highly aggregated, and heterogeneous units with politi-
cally defined and hence artificial ecological boundaries. Although in-
traurban units of analysis (e.g., census tracts, wards, block-groups) are
imperfect proxies for local community, they possess more ecological
integrity (e.g., natural boundaries, social homogeneity) than cities or
Standard Metropolitan Statistical Areas (SMSAs) and are more closely
linked to the causal processes assumed to underlie the etiology of crime
(see also Bursik 1988).

The Shaw and McKay Model

Although their focus was primarily on delinquency, the research of Shaw
and McKay forms the infrastructure for modem American studies of the
ecology of violence. In their classic work Juvenile Delinquency and Urban
Areas, originally published in 1942, Shaw and McKay (1969) argued that
three structural factors-low economic status, ethnic heterogeneity, and
residential mobility-led to the disruption of local community social organi-
zation, which in tum accounted for variations in crime and delinquency.
Shaw and McKay (1969) also demonstrated that high rates of delin-
quency persisted in certain areas over many years, regardless of population
turnover. More than any other, this finding led them to reject individual-
istic explanations of delinquency and focus on the processes by which
delinquent patterns of behavior were transmitted across generations in
areas of social disorganization and weak social controls (1969, p. 320).
Shaw and McKay, then, located the causes of crime in the structural
context of local communities and changes taking place in the wider urban
ecology. From this viewpoint the "ecological fallacy" (inferring individ-
ual-level relations from aggregate data) is irrelevant because the unit of
analysis is the community itself.
Despite Shaw and McKay's focus on nondifferentiated delinquency
rates in ecological research, two studies in the mid-20th century did
underscore the role of community-level socioeconomic status in delineat-
ing patterns of violence. Bullock (1955) examined the ecological distribu-
tion of homicide occurrence rates among the census tracts of Houston
between 1945 and 1949. Like Shaw and McKay, but focusing on more
serious crime, Bullock demonstrated that homicides were disproportionately
concentrated in areas characterized by physical deterioration of dwelling
units, unemployment, low economic status, and low median education.

Similarly, Bensing and Schroeder (1960) documented that homicides were

disproportionately concentrated in a small number of areas in Cleveland.
They described the high-rate areas as those with the "lowest socioeco-
nomic status, most undesirable neighborhood conditions, greatest finan-
cial dependency, the most acute problems of space and crowded housing
conditions, the least stability of the population, the greatest social malad-
justment and family and individual adjustment problems, and the poorest
health" (1960, p. 184). Bensing and Schroeder's description of Cleveland
homicide areas reads like Shaw and McKay's account of delinquency areas
in Chicago some 30 years earlier.

Recent Community-level Studies of Violence

After a hiatus in the 1960s, the past two decades have witnessed a sharp
increase in research focused on variations in urban crime rates (see also
the reviews in Byrne and Sampson 1986; Bursik 1988).2 Based on a
comprehensive review of community-level research, the following key
factors were identified.

Poverty and Deprivation

Not surprisingly, a large proportion of recent neighborhood-based studies

of violence have emphasized poverty. Unlike Shaw and McKay, though, the
majority of these employed multivariate methods that seek to estimate the
effects of economic structure independent of other factors such as population
composition. Overall, the results have been decidedly mixed-some studies
show a direct relationship between poverty and violence (e.g., Block 1979;
Curry and Spergel 1988) whereas others show a weak or insignificant inde-
pendent effect (Messner and Tardiff 1986; Sampson 1985, 1986; for a com-
plete review see Sampson and Lauritsen forthcoming).
Some evidence also suggests that the effect of poverty is conditional on
community contexts of mobility and change. For example, in an evaluation
of Shaw and McKay's (1969) basic model, Smith and Jarjoura (1988)
discovered a significant interaction between mobility and low income in
explaining violence across 57 neighborhoods in three cities. Mobility was
positively associated with violent crime rates in poorer neighborhoods but
not in more affluent areas. The main effects of mobility and income were
not significant when this interaction term was in the model. Smith and
Jarjoura (1988) concluded that communities characterized by rapid popu-
lation turnover and high levels of poverty have significantly higher violent
crime rates than mobile areas that are more affluent, or poor areas that are
stable. This point is crucial, and I thus tum to the second major factor
Robert J. Sampson 263

highlighted by Shaw and McKay that may help to shed light on the
meaning of poverty in ecological context.

Mobility and Community Change

One of the fundamental claims made by Shaw and McKay (1969) was that
population change and turnover had negative consequences for the social
control of delinquency. A high rate of mobility, especially in an area of
decreasing population, was inferred to foster institutional disruption and
weakened community controls (Kornhauser 1978). The existing research on
mobility has not been as extensive as that on economic status, but it has been
revealing. For example, Block's (1979, p. 50) study of Chicago's community
areas revealed large negative correlations between stability and the violent
crimes of homicide, robbery, and aggravated assault.
Victimization data from the National Crime Survey show that residen-
tial mobility has significant positive effects on rates of violent victimiza-
tion (Sampson 1985, pp. 29-30). Even after controlling for other neighbor-
hood-level correlates of victimization, rates of violent victimization for
residents of high-mobility neighborhoods are at least double those of
residents in low-mobility areas (Sampson 1985, p. 30; 1986, p. 44). As
noted above, Smith and Jarjoura (1988) also found a significant positive
effect of neighborhood mobility (percentage of households occupied by
persons who lived there less than 3 years) on rates of robbery and assault
victimization in low-income neighborhoods. This interaction of mobility
and poverty was the largest determinant of violence (1988, p. 41). In
community contexts of poverty at least, mobility appears to have impor-
tant consequences for violence.
Although mobility rates of an area have been linked to violence rates in
cross-sectional designs, until recently few studies in the Shaw and McKay
tradition have actually investigated processes of neighborhood change in
a longitudinal perspective. Recognizing this gap, Taylor and Covington
(1988) published an important study of changes in community structure
and the violent crimes of murder and aggravated assault for 277 Baltimore
city neighborhoods in the period 1970-1980. The two dimensions of
ecological change examined were economic status and family status. Their
thesis was that neighborhoods experiencing declines in relative economic
status and stability should also experience increases in relative violence
(1988, p. 561). In support ofthis notion, Taylor and Covington found that
increasing entrenchment of the urban underclass in the form of increasing
poverty and minority concentration was linked to increases in violence. In
other words, in the city's lower income neighborhoods that changed for
the worse in terms of an increasing concentration of ghetto poverty,

violence rose sharply. Gentrifying neighborhoods that were experiencing

unexpected changes in owner-occupied housing, one-unit structures, and
family status also experienced large increases in violence. Taylor and
Covington (1988, p. 583) interpreted these patterns to mean that the
process of change itself led to increasing violence.
In short, extant research suggests that rates of mobility and neighbor-
hood change are related to community violence. What seems most salient
about change is its linkage to the downward spiral of a neighborhood
becoming increasingly "underclass" (see also Schuerman and Kobrin
1986; Wilson 1987). The notion of an interaction between poverty con-
centration and residential change is discussed in more detail in a later
section, as are problems with casual interpretations of this phenomenon.

Heterogeneity and Racial Composition

Although heterogeneity has always been accorded a central role in Shaw

and McKay's research (e.g., Kornhauser 1978), in their writings Shaw and
McKay ([1942] 1969, p. 153) referred mostly to population composition.
This is not surprising because their data showed that delinquency rates
were higher in predominantly black and foreign-born areas than in areas
of maximum heterogeneity. For example, in Shaw and McKay's (1969, p.
155) data the delinquency rate in areas with more than 70% black/foreign-
born was more than double the rate in areas of maximum heterogeneity
(e.g., 50%-59% black/foreign-born).
Later research on violence has also examined racial composition-usu-
ally percentage black-rather than racial heterogeneity. A consistent find-
ing has been that percentage black is positively and strongly related to
rates of violence. Block (1979) showed that homicide rates had significant
and large correlations with percentage black, as did Messner and Tardiff
(1986), Sampson (1985), Roncek (1981), and Smith and Jarjoura (1988).
The dispute arises over the direct effect of racial composition on violence.
Though some report that percentage black remains strong even after
controlling for other factors (e.g., Roncek 1981), others find a sharply
attenuated effect of race once other factors are accounted for (e.g., Block
1979; Curry and Spergel 1988; Messner and Tardiff 1986; Sampson 1985).
Thus, although percentage black and heterogeneity are strong and per-
vasive correlates of violent crime rates, there is reason to doubt whether
racial composition has unique explanatory power. Additionally, the theo-
retical status of racial composition at the community level is question-
able-race is a salient individual-level predictor of violent offending,
especially robbery and murder (Sampson and Lauritsen forthcoming). If
community-level variations between racial composition and violent crime
Robert J. Sampson 265

merely reflect the aggregation of individual-level effects, the value of

community-level research is called into question.

Housing and Population Density

Although infrequently studied, recent research has highlighted the poten-

tial role that the physical structure and density of housing may play in
understanding patterns of violent crime. Roncek (1981) found that the per-
centage of units in multiunit housing structures was a consistent and strong
predictor of block-level variations in violent crime in Cleveland and San
Diego. Land area in acres, population size, and the percentage of primary
individual households also had significant and substantively important effects
on violence despite age and race composition. As Roncek (1981, p. 88)
summarizes, "the most dangerous city blocks are relatively large in popula-
tion and area with high concentrations of primary individuals and apartment
housing." In a similar vein, Schuerman and Kobrin (1986, p. 97) found that
increases in multiplex dwellings and renter-occupied housing were major
predictors of increases in crime rates in Los Angeles neighborhoods.
Roncek (1981, p. 88) argues that such findings are consistent with the
idea of anonymity because primary-individual households and high pro-
portions of multiunit structures are posited to increase the number of
residents who do not know one another. For example, as the number of
households sharing common living space increases, residents are less able
to recognize their neighbors, to be concerned for them, or engage in
guardianship behaviors (Roncek 1981, p. 88). These arguments were
supported in studies of violent victimization using neighborhood charac-
teristics data and the National Crime Survey (Sampson 1985). Several
studies also report a significant and large association between population
density and violent crime despite controlling for a host of social and
economic variables (see Sampson and Lauritsen forthcoming). Housing
and population density thus appear to increase the level of violent crime
regardless of compositional factors.

Family Structure

Largely ignored in early ecological research, several recent studies have

examined the community-level consequences of family "disruption" (e.g.,
divorce rates; female-headed families with children). Family disruption
has been posited to facilitate crime by decreasing community networks of
informal social control. Examples of informal social control include
neighbors observing or questioning strangers, watching over each others'
property, assuming responsibility for supervision of general youth activi-

ties, and intervening in local disturbances (see e.g., Sampson and Groves
1989; Taylor et a1. 1984). This conceptualization focuses on the commu-
nity-wide effects of family structure and does not require that it is the
children of divorced or separated parents that are engaging in crime. For
instance, youth in stable family areas, regardless of their own family
situation, have more controls placed on their leisure time activities, par-
ticularly with peer groups (Sullivan 1989, p. 178). Because delinquency
is a group phenomenon (Reiss 1986a), neighborhood family structure is
likely to be important in determining the extent to which neighborhood
youth are provided the opportunities to form a peer-control system free of
the supervision or knowledge of adults (Reiss 1986b).
Felson and Cohen (1980) also note the potential influence of family
structure not just on the control of offenders, but on the control of criminal
targets and opportunities. They argue that traditional theories of crime
emphasize the criminal motivation of offenders without considering ade-
quately the circumstances in which criminal acts occur. Predatory crime
requires the convergence in time and space of offenders, suitable targets,
and the absence of effective guardianship. The spatial and temporal struc-
ture of family activity patterns plays an important role in determining the
rate at which motivated offenders encounter criminal opportunities. For
example, those who live alone (e.g., single women) are more likely to be
out alone (e.g., going to work, restaurants, night clubs, etc.) than married
persons and are thus more vulnerable to personal crimes of violence (e.g.,
rape, robbery). In support of their theory, Felson and Cohen (1980)
demonstrated that primary-individual households had a significant and
positive effect on crime trends in the United States in the period 1950-
1972, independent of factors presumed to reflect the supply of motivated
offenders (e.g., unemployment, age composition).
The salience of family structure has been supported at the neighbor-
hood-level with respect to violence. Studies reporting a large and positive
relationship between measures of family disruption (usually percentage
female-headed families or divorce rate) and rates of violence include
Block (1979), Messner and Tardiff (1986), Sampson (1985, 1986), Roncek
(1981), Smith and Jarjoura (1988), and Schuerman and Kobrin (1986).
What is especially striking is the strength of family disruption in predicting
violence in multivariate models. For example, Sampson (1985, 1986)
found that rates of victimization were 2-3 times higher among residents of
neighborhoods with high levels of family disruption compared to low
levels, regardless of alternative predictors of violent victimization such as
percentage black and poverty. In fact, the percentage of female-headed
families helped to explain in large part the relationship between percent-
age black and crime. Namely, percentage black and percentage female-
Robert J. Sampson 267

headed families are positively and significantly related; however, when

percentage female-headed families is controlled, percentage black is not
significantly related to violent victimization (1985, p. 27).
Similarly, Messner and Tardiff (1986) found that when percentage
divorced and percentage poverty were controlled, the relationship between
percentage black and homicide rates was insignificant. And Smith and
Jarjoura (1988) report that family structure, especially percentage single-
parent families, helps account for the association between race and violent
crime at the community level: racial composition was not significantly
related to violent crime in multivariate models once percentage single-par-
ent families was included.
Overall, then, the correlation of family structure with violence is not
simply the result of other factors we typically consider, such as poverty,
race, and density. Rather, the effect of family disruption appears to be
independent and large. I now explicate the potential intervening constructs
that may help to explain why family disruption and other community
structural factors are related to violence.

Community Social Disorganization

The theory of social disorganization refers to the inability of a community

structure to realize the common values of its residents and maintain effective
social controls (Kornhauser 1978, p. 120; Bursik 1988). The structural
dimensions of community social disorganization can be conceptualized by
the prevalence and interdependence of social networks in a community-both
informal (e.g., density of acquaintanceship; intergenerational ties) and formal
(e.g., organizational participation)-and in the span of collective supervision
that the community directs toward local problems (Sampson and Groves
1989). This approach is grounded in a systemic model, where the local
community is viewed as a complex system of friendship and kinship net-
works, and formal and informal social ties rooted in family life and ongoing
socialization processes (Sampson 1988). Social organization and social dis-
organization are thus seen as different ends of the same continuum ofsystemic
networks of community social control. As Bursik (1988) notes, when formu-
lated in this way social disorganization is clearly separable not only from the
processes that may lead to it (e.g., poverty, mobility), but also from the degree
of criminal behavior that may be a result.
A major dimension of social disorganization is the ability of a community
to supervise and control teenage peer-groups-especially gangs (Sampson
and Groves 1989). It is well documented that delinquency is primarily a
group phenomenon (Thrasher 1963; Shaw and McKay 1969; Reiss 1986a),
and hence, the capacity of the community to control group-level dynamics

is a key theoretical mechanism linking community characteristics with

crime. A central fact underlying Shaw and McKay's research was that the
majority of gangs developed from the unsupervised, spontaneous play-
group (Thrasher 1963, p. 25). Shaw and McKay (1969) thus argued that
residents of cohesive communities were better able to control the teenage
behaviors that set the context for group-related violence (see also Short
1963, p. xxiv). Examples of such controls include supervision of leisure-
time youth activities, intervention in street-corner congregation (Thrasher
1963, p. 339; Shaw and McKay 1969, pp. 176-85), and challenging youth
"who seem to be up to no good" (Skogan 1986, p. 217; Taylor et al. 1984,
p. 326). Socially disorganized communities with extensive street-corner
peer groups are also expected to have higher rates of adult violence,
especially among younger adults who still have ties to youth gangs (Thrasher
A second dimension of community social organization is local friend-
ship and acquaintanceship networks. Systemic theory holds that locality-
based social networks constitute the core social fabric of human-ecologi-
cal communities (Sampson 1988). When residents form local social ties
their capacity for community social control is increased because they are
better able to recognize strangers, and are more apt to engage in guardian-
ship behavior against victimization (Taylor et al. 1984, p. 307; Skogan
1986, p. 216). Also, the greater the density and multiplexity of interper-
sonal networks in a community, the greater the constraint on deviant
behavior within the network.
A third component of social organization is the rate of local participa-
tion in formal and voluntary organizations. Community organizations
reflect the structural embodiment oflocal community solidarity, and hence
the instability and isolation of community institutions are key factors
underlying the structural dimension of social disorganization. As such,
Kornhauser (1978, p. 79) argues that when the links among community
institutions are weak the capacity of a community to defend its local
interests is also weakened. Shaw and McKay (1969, pp. 184-85), and more
recently Taylor et al. (1984) and Simcha-Fagan and Schwartz (1986, p.
688), have also argued that a weak community organizational base atten-
uates local social-control functions.
It is difficult to study social disorganization directly, but at least two
recent studies provide empirical support for the theory's structural hypoth-
eses. First, Taylor et al. (1984) examined variations in violent crime across
63 street blocks in Baltimore in 1978. Based on interviews with 687
household respondents, Taylor et al. (1984, p. 316) constructed block-
level measures of what they termed "social ties" and "near home respon-
sibilities." The former measured the proportion of respondents who
Robert J. Sampson 269

belonged to an organization to which co-residents also belonged, and the

latter measure tapped the extent to which respondents felt responsible for
what happened in the area surrounding their home. Both of these dimen-
sions of informal social control were significantly and negatively related
to community-level variations in violence (1984, p. 320). Additionally,
Taylor et al. (1984, p. 317) showed that blocks with higher neighborhood
identification (measured by the proportion of residents who were able to
provide a neighborhood name) had significantly lower rates of violence.
These results support the social-disorganization hypothesis that levels of
organizational participation and informal social control (especially of
local activities by neighborhood youth) inhibit community-level rates of
violence (see also Taylor et al. 1984, p. 326).
Second, Sampson and Groves (1989) analyzed the British Crime Survey
(BCS), a nationwide survey of England and Wales conducted in 1982 and
1984. Using samples drawn from more than 200 geographical units that
approximated the concept of "local community," survey-generated responses
were aggregated within each of the areas and structural variables constructed
(e.g., means, percentages). The results (1989, p. 789) showed that the preva-
lence of unsupervised peer-groups in a community (e.g., teenagers hanging
out making a nuisance) had the largest effect on rates of robbery and stranger-
violence victimization in 1982. The density of local friendship networks-
measured by the proportion of residents with half or more of their friends
living in the neighborhood-and the level of organizational participation by
residents also had significant negative effects on violent victimization. The
largest effect on personal violence offending rates in 1982 was unsupervised
peer groups (1989, p. 793). These general findings were replicated in the 1984
BCS, thereby suggesting that communities characterized by sparse friendship
networks, unsupervised teenage peer-groups, and low organizational partici-
pation experience disproportionately high rates of violence.
Moreover, for both survey years variations in structural dimensions of
community social disorganization mediated in large part the effects of
community socioeconomic status, residential mobility, ethnic heterogene-
ity, and family disruption in a manner consistent with social disorganiza-
tion theory. For example, mobility had significant inverse effects on
friendship networks, family disruption was the largest predictor of unsu-
pervised peer groups, and socioeconomic status had the largest inverse
effects on organizational participation in 1982. When combined with the
results of research on gang delinquency that point to the salience of
informal and formal community structures in controlling the formation of
gangs (Thrasher 1963; Sullivan 1989), the empirical data suggest that the
structural elements of social disorganization have relevance for explaining
macrolevel variations in violence.

Community Cultures and Ethnography

Social disorganization theory also focuses on how the ecological segrega-

tion of communities gives rise to what Kornhauser (1978, p. 75) terms cultural
disorganization-the attenuation of societal cultural values. Poverty, mobil-
ity, heterogeneity, and other structural features of urban communities are
hypothesized to impede communication and obstruct the quest for common
values, thereby fostering cultural diversity of nondelinquent values. Accord-
ingly, a second component of Shaw and McKay's theory was that disorga-
nized communities spawned delinquent gangs with their own subcultures and
norms perpetuated through cultural transmission.
Despite their relative infrequency, ethnographic studies support the
notion that structurally disorganized communities are conducive to the
emergence of cultural value systems and attitudes that legitimate, or least
provide a basis of tolerance, for crime and deviance. For example, Suttles's
(1968) account of the social order of a Chicago neighborhood character-
ized by poverty and heterogeneity supports Thrasher's (1963) emphasis
on age, sex, ethnicity, and territory as markers for the ordered segmenta-
tion of slum culture. Suttles found that single-sex, age-graded primary
groups of the same ethnicity and territory emerged in response to threats
of conflict and community-wide disorder and mistrust. Although the com-
munity subcultures Suttles (1968) discovered were provincial, tentative,
and incomplete (Kornhauser 1978, p. 18), they nonetheless undermined
societal values against delinquency and violence.
Similarly, Anderson's (1978) ethnography of a bar in Chicago's South-
side black ghetto shows how primary values coexisted alongside residual
values associated with deviant subcultures (e.g., hoodlums) such as "tough-
ness," "getting big money," "going for bad," and "having fun" (1978, pp.
129-30; 152-58). In Anderson's analysis, lower class residents do not so
much "stretch" mainstream values as "create their own particular stan-
dards of social conduct along variant lines open to them" (1978, p. 210).
In this context the use of violence is not valued as a primary goal, but is
still expected and tolerated as a fact of life (1978, p. 134). Much like
Suttles (1968) and Horowitz (1987), Anderson's ethnographic research
suggests that in certain community contexts the wider cultural values are
simply not relevant-they become "unviable."
Whether or not community subcultures are authentic or merely "shadow
cultures" (Liebow 1967) cannot be resolved here (see also Kornhauser
1978). Yet that seems less important than acknowledging that community
contexts shape what I term cognitive landscapes or ecologically structured
norms (e.g., normative ecologies) regarding appropriate standards and
expectations of conduct. That is, in structurally disorganized slum
Robert J. Sampson 271

communities it appears that a system of values emerges in which violence,

disorder, and drug use are less than fervently condemned and hence
expected as part of everyday life. These ecologically structured social
perceptions and tolerances in tum appear to influence the probability of
violent outcomes and other deviant behavior (e.g., drug use). In this regard
I believe Kornhauser's attack on subcultural theories misses the point. By
attempting to assess whether subcultural values are authentic in some
deep, quasi-religious sense (1978, pp. 1-20), she loses sight of the pro-
cesses by which cognitive landscapes rooted in normative ecologies may
influence behavior. Indeed, the idea that dominant values become existen-
tially irrelevant in certain community contexts is a powerful one, albeit
one that has not had the research exploitation it deserves.'
The foregoing review suggests that structural and cultural dimensions of
community social disorganization have mediating effects on violent crime.
Having reviewed the exogenous community-level correlates of violence and
the intervening features of social disorganization, I tum now to a brief
summary of ecological findings based on interurban units of analysis.


Probably because of the greater ease of data availability, recent literature

on violence in relation to characteristics of metropolitan-level or interurban
units of analysis such as cities and Standard Metropolitan Statistical Areas
(SMSAs) is much more extensive than research at the local community level.
Unfortunately, the high level of aggregation at the city and SMSA levels
makes this area of research less central to theoretical concerns regarding
community-level processes underlying violence (for more discussion see
Sampson and Lauritsen forthcoming). However, it is useful to highlight two
recent macrolevel studies that are largely consistent with those at the com-
munity level and that provide further insights into patterns of violence.
In the first, Land' et al. (1990) analyzed the relationships among the
following structural covariates represented in 21 macrolevel studies of
homicide: population size, population density, percentage of the popula-
tion that is black, percentage youth (e.g, percentage age 15-29), percentage
male divorce rate, percentage of children not living with both parents,
median family income, percentage families living in poverty, the Gini
index of income inequality, the unemployment rate, and a dichotomous
variable indicating those cities or metropolitan areas located in the South.
Using principal components analysis, two clusters of variables were found
to consistently "hang together" over time and space (i.e., 1960, 1970, and
1980; cities, SMSAs, and states). The first factor was termed a population

structure component and consisted of population size and population

density. The second factor was labeled resource deprivation/affluence,
and included three income variables-median income, percentage of fam-
ilies below the poverty line, and the Gini index of income inequality-in
addition to percentage black and percentage of children not living with
both parents. Although these variables tap somewhat different concepts,
Land et al, (1990) found they could not be separated empirically. On the
other hand, four variables-percentage divorced males, percentage age
15-29, unemployment rate, and South-were found to vary independently.
Land et al. (1990) examined how resource deprivation/affluence and
population structure along with these four single variables explained
homicide rates across time and space. As Land et al. (1990, p. 947)
summarize the findings: "three structural indexes/covariates-population
structure, resource deprivation/affluence, and percent of the male popula-
tion divorced-now exhibit statistically significant relationships to the
homicide rate in the theoretically expected direction across all time peri-
ods and levels of analysis." This result suggests that, all else equal, cities
with a large poor population in conjunction with a high percentage of black
residents and single-parent families with children have disproportionately
high homicide rates. Similarly, homicide rates are larger in cities, states,
and regions with large population size and high male divorce rates. In
magnitude, resource deprivation has the largest effect, generally followed
by the male divorce rate and population structure.
The large effect of resource deprivation/affluence on homicide is sub-
stantively consistent with Wilson's (1987) recent conceptualization of
concentration effects. Wilson argues that the social transformation of the
inner city has resulted in a disproportionate concentration of the most
disadvantaged segments of the urban black population-s-especially poor,
female-headed black families with children. In particular, opposition from
organized community groups to public housing projects coupled with
political decisions to neglect rehabilitation of single-family residential
units, has led to increasing segregation of poor blacks (Sampson 1990).
Urban minorities have also been vulnerable to structural economic changes
related to the de-industrialization of central cities (e.g., shift from goods-
producing to service-producing industries; increasing polarization of the
labor market into low-wage and high-wage sectors; and relocation of
manufacturing out of the inner city). This social milieu differs signifi-
cantly from the environment that existed in inner cities in previous decades
(Wilson 1987, p. 58). Namely, the exodus of middle- and upper income
black families from the inner city removes an important social buffer that
could deflect the full impact of prolonged joblessness and industrial
transformation (Hagedorn 1988). Wilson's (1987, p. 56) thesis is based on
Robert J. Sampson 273

the assumption that the basic institutions of an area (e.g., churches,

schools, stores, recreational facilities) are more likely to remain viable if
the core of their support comes from more economically stable families in
inner-city neighborhoods.
Wilson's (1987) concept of social isolation in areas of concentrated
poverty is consistent with the Land et al. (1990) findings and also those of
Taylor and Covington (1988) noted earlier, who found that increasing
entrenchment of the urban "underclass" was associated with large in-
creases in violence. Land et al.' s (1990) results also go beyond Wilson by
suggesting that the clustering of economic and social indicators appears
not only in 1980 and in neighborhoods of large cities, but also for the two
previous decennial periods and at the level of macrosocial units as a whole.
Moreover, Land et al. present evidence in support of Wilson's argument
that concentration effects grew more severe from 1970 to 1980 in large
cities-"the numerical values of the component loadings of percent pov-
erty, percent black, and percent of children under 18 not living with both
parents are larger in 1980 than 1970" (1990, p. 945). Therefore, indicators
of urban disadvantaged populations appear to be increasing in their eco-
logical concentration.

Racial Disaggregation
The Land et al. (1990) approach is important because it provides a summary
index of the effects of poverty and relative deprivation on homicide. How-
ever, the resource deprivation index still confounds measures of racial com-
position, and hence does not allow the examination of poverty and family
structure independent of race. This is problematic because racial differences
in poverty and family disruption are so strong that the worst urban contexts
in which whites reside are considerably better off than the average context of
black communities (Sampson 1987, p. 354). Thus, regardless of whether or
not a black juvenile is raised in an intact or "broken" home, or rich or poor
home, he or she will not grow up in a community context similar to whites
with regard to family structure and poverty.
One way to examine these contextual issues is through racial disaggre-
gation of both the crime rate and the explanatory variables of theoretical
interest. This approach was used in research that examined racially disag-
gregated rates of homicide and robbery by juveniles and adults in more
than 150 U.S. cities in 1980 (Sampson 1987). Substantively, the theory
tested focused on the exogenous factors of black male joblessness and
economic deprivation and their effects on violent crime as mediated by
black family disruption. The results supported the main hypothesis and
showed that the scarcity of employed black males relative to black women

was directly related to the prevalence of families headed by females in

black communities (Wilson 1987). In tum, black family disruption was
substantially related to rates of black murder and robbery, especially by
juveniles. These effects were independent of income, region, density, city
size, and welfare benefits. The finding that family disruption had stronger
effects on juvenile violence than on adult violence, in conjunction with
the inconsistent findings of previous research on individual-level delin-
quency and broken homes, supports the idea that the effects of family
structure are related to macro level patterns of social control and guardian-
ship, especially regarding youth and their peers (see also Felson and Cohen
1980; Sampson and Groves 1989). Moreover, the results suggest why
unemployment and economic deprivation have had weak or inconsistent
effects on violence rates in past research-although joblessness or even
poverty may not have direct effects on crime, they do have significant
effects on family disruption, which in tum predicts violence.
Despite a tremendous difference in mean levels of family disruption
among black and white communities, the percentage of white families
headed by a female also had a large positive effect on white juvenile and
white adult violence. In fact, the predictors of white robbery were shown
to be in large part identical in sign and magnitude to those for blacks.
Therefore, the effect of family disruption on black crime was independent
of commonly cited alternative explanations (e.g., poverty, region, density,
age composition), and could not be attributed to unique cultural factors
within the black community given the similar effect of family disruption
on white crime.


In synthesizing current knowledge on the community-level risk factors in

violence, we must acknowledge that numerous problems plague this sort of
research---especially inferences about the causal status of community char-
acteristics. Among other limitations, the use of varying and sometimes highly
aggregated units of analysis, potentially biased sources of information on
violence (e.g., official data), feedback or reciprocal effects from crime itself,
widely varying analytical techniques, and high correlations among indepen-
dent variables all hinder the attempt to forge general conclusions.
Furthermore, prior research has mostly inferred the existence of inter-
vening community processes, even though the correlation of violence with
ecological characteristics is consistent with many different theoretical
perspectives. For example, the typical aggregate-level study shows that
percentage black, poverty, and family disruption predict violence rates.
Robert J. Sampson 275

Although useful as a preliminary test, this strategy does not go much

beyond the steps taken by Shaw and McKay more than 40 years ago. As
Kornhauser (1978, p. 82) argues, most criminological theories take as their
point of departure the same independent variables (e.g., socioeconomic
status). The variables that intervene between community structure and
violence are at issue, and to test competing theories adequately one must
establish the relationship to violence of the interpretive variables each
theory implies.
Unfortunately, previous macro level research has relied primarily on
census data that rarely provide measures for the social variables hypoth-
esized to mediate the relationship between community structure and vio-
lence. Reliance on census data has also meant that culture is commonly
ignored in macro level research even though a key element of macrolevel
explanation is how community cultures and value systems produce differ-
ential rates of crime. Some studies have examined quantitative dimensions
of informal social control (see, e.g., Maccoby et at. 1958; Simcha-Fagan
and Schwartz 1986), but they have been limited to a few communities,
precluding comprehensive multivariate analysis and generalization. Con-
sequently, empirical examination of between-community differences in
social disorganization and other community-level processes have been
rarely undertaken (but see Taylor et at. 1984; Sampson and Groves 1989).
Another major problem underlying macrolevel research is the mistaken
assumption that the unit of analysis automatically determines the level of
causal explanation. Macrosocial investigations usually claim that the
ecological fallacy is irrelevant because an interest in individuals is dis-
avowed. The problem with this strategy is that an apparent ecological or
structural effect may in fact arise from individual-level causal processes.
For example, an observed macro level result such as the correlation of
median income or percentage black with violence rates may simply repre-
sent the aggregation of relationships occurring at lower levels of social
structure and not a manifestation of processes taking place at the level of
the community as a whole. The research on individual-level risk factors
demonstrates the potential power of compositional confounding by factors
such as race and social class (Sampson and Lauritsen forthcoming). Even
though rates of crime may not be used to make inferences about individ-
uals, individuals surely commit the crimes that constitute the rates.
Consider further the basic facts on delinquency. Research has consis-
tently demonstrated the early onset of delinquency and its long-term
stability (Glueck and Glueck 1950; Robins 1966; Sampson and Laub
1990). These general differences among individuals that are stable over
time have direct implications for an ecological study of crime. Longitudi-
nal research suggests that delinquent tendencies are fairly well established

at early ages-even at age 8 or so, and certainly by the early teens (Glueck
and Glueck 1950). Antisocial children tend to fight, steal, become truant,
drop out of school, drift in and out of unemployment, live in lower class
areas, and go on to commit adult crime (Sampson and Laub 1990). The
causal nature of the relationship between achieved adult characteristics
(e.g., employment status) and adult crime is thus fraught with methodolog-
ical difficulties. In fact, in Deviant Children Grown Up, Lee Robins (1966)
offered the provocative hypothesis that antisocial behavior predicts class
status more efficiently than class status predicts antisocial behavior.
If area differences in violence rates result from the characteristics of
individuals selectively located in those communities (Kornhauser 1978),
we must question the findings derived from macrolevel criminology. Is the
correlation of concentrated poverty with violence caused by an aggrega-
tion of individual-level effects of class, a genuine community-level effect,
or is it simply a differential selection of individuals into communities
based on prior (e.g., antisocial) behavior? Or is it that common third
factors cause individuals to both commit violence and perform poorly in
the occupational sphere? And if violent and antisocial tendencies are
formed at early (preteen) ages, what plausible roles can community labor
markets and economic stratification play in understanding violence?
Simply put, macrolevel research is not immune to questions concerning
the level at which causal relations operate. The level at which a causal
relation occurs is a complex issue that is not solved simply by the nature
of how variables are measured or the unit for which they are measured. To
make matters more complicated, the concrete actions of individuals also
feed back to shape the collective environment (Tienda 1989). Thus the unit
of analysis does not define the level of causal explanation, and the infor-
mation contained in aggregate data is not necessarily generated by macro-
sociological processes.


The confluence of selective aggregation into communities, cross-level

misspecification (e.g., compositional effects), multicollinearity and non-
experimental designs, a static conceptualization of community structure,
the early onset of many forms of violence, and indirect measurement of
community characteristics suggest that community-level research is still
in its infancy. Nevertheless, this conclusion should not lead us to despair.
I have been deliberately critical in assessing a community-level perspec-
tive on violence as a vehicle to suggest new directions and insights for
research and policy. In particular, I believe that a community-level per-
spective not only improves our ability to distinguish causal effects on
Robert J. Sampson 277

violence and hence adjudicate among competing theories, but that it

enlarges our conceptual lens to assess the implications of current policies
for the public agenda.
We should recognize also that the research limitations just noted are no
worse than those found in individual-level research. The difference is that the
assumptions embodied in individual-level research are usually accepted at
face value. Consider cross-level misspecification in the form of the "individ-
ualistic fallacy"-the assumption that individual-level correlations are nec-
essarily generated by individual-level causal relations. Research conducted
on individuals rarely questions whether obtained results might be spurious
and confounded with community-level processes.
A good example is race, an important individual-level correlate of
violence (Sampson and Lauritsen forthcoming). It is commonplace to
search for individual-level explanations for the race-violence linkage
(including the ubiquitous reference to social class differences). By con-
trast, Wilson's (1987) thesis of concentration effects suggests a possible
contextual explanation for the race-violence link among individuals. Al-
though approximately 70% of all poor whites lived in nonpoverty areas in
the five largest U.S. central cities in 1980, only 15% of poor blacks did.
Moreover, whereas only 7% of poor whites lived in extreme poverty areas,
almost 40% of poor blacks lived in such areas (Wilson 1987, p. 58). In
New York City, 70% of poor blacks live in poverty neighborhoods; by
contrast, 70% of poor whites live in nonpoverty neighborhoods (Sullivan
1989, p. 230).
The consequences of these differential ecological distributions by race
raise the substantively plausible hypothesis that individual-level correla-
tions by race (and also class) may be systematically confounded with
important differences in community contexts. As Wilson argues:

simple comparisons between poor whites and poor blacks would be confounded
with the fact that poor whites reside in areas which are ecologically and econom-
ically very different from poor blacks. Any observed relationships involving race
would reflect, to some unknown degree, the relatively superior ecological niche
many poor whites occupy with respect to jobs, marriage opportunities. and
exposure to conventional role models. (1987. pp. 58-60)

Regardless of a black's individual-level family or economic situation, the

average community of residence differs dramatically from that of a simi-
larly situated white (Sampson 1987). Therefore, the relationship between
race and violence may be largely accounted for by community context
(e.g., segregation, concentration of family disruption and joblessness;
social isolation; sparse social networks and resources).

Race and violence is but one example of the potentially misleading infer-
ences that might be drawn in our attempt to explain violence when individuals
are stripped of community and structural context. Others include family
disruption and residential mobility. Because of the typical individual-level
research design, alternative explanations are never subject to a critical test.
Although space limitations preclude detail, I therefore conclude with discus-
sion of community-based research strategies and insights that may help to
resolve outstanding questions and provide policy inputs.


A major implication of this chapter is the need for criminological research

on violence to more effectively consider the community level of analysis. One
approach to this need is contextual analysis where information on communi-
ties is combined with individual-level data to explain violence. This approach
eliminates the tendency in criminological research to ignore either the micro
or macro levels based on disciplinary biases and data availability. Most prior
research on violence involves study of either individual effects or commu-
nity-level effects-almost no research has examined both (Reiss 1986b).
Most individual-level research is thus inadequate because it neglects variation
in community characteristics, whereas community-level research fails to take
account of selection effects.
To meet these goals requires new initiatives in research design and
conceptualization. Perhaps the most pressing issue is the importance of
studying neighborhood and individual change as a means to distinguish
contextual from selection effects. Understanding mechanisms producing
alternative neighborhood trajectories is essential to avoid confounding
selection effects with neighborhood effects (Tienda 1989, p. 18). Control-
ling for background characteristics is not enough, for contextual analysis
per se does not deal with selection processes. As Tienda (1989) argues,
multilevel models that simply combine person and place characteristics
without regard to the potential endogeneity of neighborhood characteris-
tics are flawed. Therefore, longitudinal designs that follow not only
changes in the structure, composition, and organization of communities
but also the individuals who reside there are needed to establish the unique
contribution of individuals and communities to crime and delinquency
(Reiss 1986b, p. 29).
The longitudinal study of communities is important for making causal
inferences even for purely macrolevel processes. There are reasons to
expect that violence has feedback effects on major dimensions of commu-
nity structure, and several key hypotheses concerning the link between
Robert J. Sampson 279

communities and rates of violence were explicitly formulated in terms of

change (e.g., increasing concentration of poverty; effect of residential
turnover on disruption of social networks). To disentangle community
processes and the potential reciprocal effects of violence requires longi-
tudinal designs where temporal ordering can be established and change
processes explicitly modeled.
Another major research need is for direct examination of the social
processes and interactions through which community effects are transmit-
ted. One measurement strategy is to design survey instruments that tap
community-level patterns of informal social control, friendship networks,
collective orientations (e.g., subcultural values), rates of organizational
participation, and social resources. Some preliminary attempts have been
made (e.g., Taylor et al. 1984; Simcha-Fagan and Schwartz 1986; Samp-
son and Groves 1989), but we need units of analysis defined by the
research question, not by governmental agencies.
A different approach is ethnographic analysis, for we can learn much
about community structure and violence from ethnographic studies that
we cannot in multivariate analyses of census data and survey measures.
Given the importance of community cultures and value systems in crimi-
nological theories of violence, it is problematic that modern criminology
has reified multivariate methodology as the engine of scientific progress.
I believe a more balanced perspective calls for multivariate research to be
supplemented with community ethnography. John Hagedorn's People and
Folks: Gangs, Crime and the Underclass in a Rustbelt City (1988) and
Mercer Sullivan's Getting Paid: Youth Crime and Work in the Inner City
(1989) are but two recent examples of how rich and provocative commu-
nity ethnographic research can be.


Last, but certainly not least, a substantive. focus on community-level

processes has constructive implications for public policy. The reason is
that many of the hypothesized community-level sources of violence noted
throughout this chapter (e.g., residential instability, concentration of pov-
erty and family disruption, high-density public housing projects; attenua-
tion of social networks) are determined, both directly and indirectly, by
the policy decisions of public officials. Take, for example, municipal code
enforcement and local governmental policies toward neighborhood dete-
rioration. In Making the Second Ghetto: Race and Housing in Chicago,
1940-1960, Hirsch (1983) documents in great detail how lax enforcement
of city housing codes played a major role in accelerating the deterioration

of inner-city Chicago neighborhoods. More recently, Daley and Meislin

(1988) have argued that inadequate city policies on code enforcement and
repair of city properties contributed to the systematic decline of New York
City's housing stock, and consequently, entire neighborhoods. When con-
sidered with the practices of redlining and disinvestment by banks and
"block-busting" by real estate agents (Skogan 1986), local policies toward
code enforcement-that on the surface are far removed from crime-have
in all likelihood contributed to crime through neighborhood deterioration,
forced migration, and instability.
The provision of city municipal services for public health and fire safety-
decisions presumably made with little if any thought to crime and violence-
also appear to have been salient in the social disintegration of poor commu-
nities. As Wallace and Wallace (1990) argue based on an analysis of the
"planned shrinkage" of New York City fire and health services in recent
decades: "The consequences of withdrawing municipal services from poor
neighborhoods, the resulting outbreaks of contagious urban decay and forced
migration which shred essential social networks and cause social disintegra-
tion, have become a highly significant contributor to decline in public health
among the poor" (1990, p. 427). The loss of social integration and networks
from planned shrinkage of services may increase behavioral patterns of
violence that may themselves become "convoluted with processes of urban
decay likely to further disrupt social networks and cause further social
disintegration" (1990, p. 427). This pattern of destabilizing feedback (see
Skogan 1986) appears central to an understanding of the role of governmental
policies in fostering the downward spiral of high crime areas.
Decisions by government on public housing paint a similar picture.
Bursik (1989) has shown that the planned construction of new public
housing projects in Chicago in the 1970s was associated with increased
rates of population turnover, which in turn were related to increases in
crime-independent of racial composition. More generally, Skogan (1986,
p. 206) notes how urban renewal and forced migration contributed to the
wholesale uprooting of many urban communities, especially the extent to
which freeway networks driven through the hearts of many cities in the
1950s destroyed viable, low-income communities. The instability forced
upon low-income minority neighborhoods by governmental urban-renewal
decisions was profound. In Atlanta, one in six residents was dislocated
through urban renewal, the great majority of whom were poor (Logan and
Molotch 1987, p. 114). Nationwide, fully 20% of all central city housing
units occupied by blacks were lost in the period 1960-1970. As Logan and
Molotch (1987, p. 114) observe, this displacement does not even include
that brought about by more routine market forces (evictions, rent in-
creases, commercial development).
Robert J. Sampson 281

Perhaps most disturbing, Wilson (1987, pp. 20-62) documents the negative
consequences of policy decisions to concentrate minorities and the poor in
public housing. Opposition from organized community groups to the building
of public housing in their neighborhoods, de facto federal policy to tolerate
extensive segregation against blacks in urban housing markets, and the
decision by local governments to neglect the rehabilitation of existing resi-
dential units (many of them single-family homes), have led to massive,
segregated housing projects that have become ghettos for the minorities and
disadvantaged (see also Sampson 1990). The social transformation of the
inner city, triggered in large part by governmental policy, has thus led to the
disproportionate concentration of the most disadvantaged segments of the
urban black population in a few areas. The result is that even given the same
objective socioeconomic status, blacks and whites face vastly different envi-
ronments in which to live, work, and raise their children (Massey and
Kanaiaupuni 1990, p. 18). When linked to the probable effects of concentrated
urban poverty and family disruption on community social organization as
reviewed above, governmental housing policies clearly become relevant to
crime control policy. Moreover, when segregation and concentrated poverty
represent structural constraints embodied in public policy, concerns raised
earlier about self-selection confounding the interpretation of community-
level effects on violence are considerably diminished. As Massey and
Kanaiaupuni have argued, public housing represents a federally funded,
physically permanent institution for the isolation of black families by race
and class, and must therefore be considered an important structural constraint
on ecological area of residence (1990, p. 20).
In short, government matters, and we need to incorporate explicitly the
political economy of place (Logan and Molotch 1987) into theories and
policies of community-level social organization. The reason is that what
seem to be "non-crime" policies-for example, where or if to build a
housing project; enforcement of municipal codes; reduction in essential
municipal services; rehabilitation of existing residential units, the dis-
bursement of the disadvantaged---can have important indirect effects on
violence. As detailed above, many community characteristics implicated
in violence such as residential instability, concentration of poor, female-
headed families with children, multiunit housing projects, and disrupted
social networks, appear to stem rather directly from planned governmental
policies at local, state, and federal levels. This conceptualization diverges
from the natural market assumptions of the early social ecologists (e.g.,
Shaw and McKay 1969) by considering the role of political decisions in
shaping community structure.
On the positive side, the implication of this community-level perspective
is that policy-manipulable options may help reverse the tide of community

social disintegration. Among others, these might include resident manage-

ment ofpublic housing (to increase stability), tenant buy-outs (to increase
home ownership and commitment to locale), rehabilitation of existing
low-income housing (to preserve area stability, especially single-family
homes), disbursement ofpublic housing (versus concentration), and strict
code enforcement (to fight deterioration). Moreover, there are recent
examples that such policies are viable and in fact have stabilizing effects
on communities and hence potential for crime reduction (see, e.g., the
programs described in Reiss 1989 and Sampson 1990; more generally, see
Logan and Molotch 1987).
Finally, a community-level perspective shows the unintended conse-
quences of official social control policies that focus on atomized individ-
uals. The major thrust of current crime control policy-whether aimed at
drugs or violence-is to lock up or execute offenders. Such policies
assume that either individual deterrence or incapacitation will reduce
further violence. A community-level or structural perspective on crime
suggests that we should not be surprised to see crime rates ignore these
threats. Violence rates have recently risen in many of our nation's citie