The Spiral of Silence: New Perspectives on Communication and Public Opinion [1 ed.] 0415509327, 9780415509329

Table of contents :
Cover
The Spiral of Silence
Title Page
Copyright Page
Table of Contents
Tables and Figures
Contributors
1 The Legacy of Spiral of Silence Theory: An Introduction
2 Public Opinion and Social Control
Part I Overview, Evidence, Methods
3 Paradigm Shifts in the Study of Media Effects
4 Three Contexts of the Spiral of Silence Theory
5 Methodological Conundrums in Spiral of Silence Research
6 Speaking in Spirals: An Updated Meta-Analysis of the Spiral of Silence
Part II Media Fragmentation, Internet, and Public Opinion
7 Normative Implications of the Spiral of Silence: Hearing the Other Side in the 21st Century
8 Media and Public Opinion in a Fragmented Society
9 Public Opinion Expression in Online Environments
10 Linking Social Network Analysis to the Spiral of Silence, Coorientation, and Political Discussion: The Intersection of Political Perceptions and Political Communication
Part III Allied Research Traditions
11 Social Norms, Spirals of Silence and Framing Theory: An Argument for Considering Cross-Cultural Differences in Media Effects Research
12 The Intersection of Third-Person Effect and Spiral of Silence
13 Pluralistic Ignorance and the Spiral of Silence Meet: Mutual Lessons
14 Opinion Leadership and Public Opinion: Where Weak/Strong Media Paradigms Converge
15 Vernacular Rhetoric, Considered Public Opinion and Modes of Publicity
Part IV Intercultural Contexts
16 World Public Opinion
17 Cultural Orientation and the Spiral of Silence
18 The Kuuki Theory and the Spiral of Silence
A Personal Tribute
19 Elisabeth Noelle-Neumann and the Mysteries of Solitude
Author Index
Subject Index

Citation preview

THE SPIRAL OF SILENCE

Since its original articulation in the early 1970s, the spiral of silence theory has become one of the most studied theories of communication and public opinion. It has been tested in varied sociopolitical contexts, with different issues and across communication systems around the world. Attracting the interest of scholars from communication, political science, sociology, public opinion, and psychology, it has become both the subject of tempestuous academic debate as well as a mainstay in courses on communication theory globally. Reflecting substantial new thinking, this collection provides a comprehensive examination of the spiral of silence theory, offering a synthesis of prior research as well as a solid platform for future study. It addresses various ideological and methodological criticisms of the theory, links the theory with allied areas of scholarship, and provides analyses of empirical tests. Contributors join together to present a breadth of disciplinary and international perspectives. As a distinctive and innovative examination of this influential theory, this volume serves as a key resource for future research and scholarship in communication, public opinion, and political science. Wolfgang Donsbach is Professor of Communication at the Department of Media and Communication at Dresden University of Technology, Germany. Charles T. Salmon is Professor and Director of Graduate Research Programmes in the Wee Kim Wee School of Communication and Information at Nanyang Technological University, Singapore. Yariv Tsfati is Associate Professor at the Department of Communication, University of Haifa, Israel.

This page intentionally left blank

THE SPIRAL OF SILENCE New Perspectives on Communication and Public Opinion

Edited by Wolfgang Donsbach, Charles T. Salmon, and Yariv Tsfati

First published 2014 by Routledge 711 Third Avenue, New York, NY 10017 And by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2014 Taylor & Francis The right of Wolfgang Donsbach, Charles T. Salmon, and Yariv Tsfati to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging in Publication Data The spiral of silence : new perspectives on communication and public opinion / edited by Wolfgang Donsbach, Charles T. Salmon, Yariv Tsfati. pages cm 1. Public opinion. 2. Mass media and public opinion. I. Donsbach, Wolfgang, 1949– II. Salmon, Charles T. (Charles Thomas) III. Tsfati, Yariv. HM1236.S665 2014 303.3’8—dc23 2013036612 ISBN: 978-0-415-50932-9 (Paperback) ISBN: 978-0-415-50931-2 (Hardback) ISBN: 978-0-203-12500-7 (eBook) Typeset in Bembo and Stone Sans by EvS Communication Networx, Inc.

CONTENTS

Tables and Figures Contributors 1 The Legacy of Spiral of Silence Theory: An Introduction Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon 2 Public Opinion and Social Control Elisabeth Noelle-Neumann

ix xi 1

19

PART I

Overview, Evidence, Methods

33

3 Paradigm Shifts in the Study of Media Effects Denis McQuail

35

4 Three Contexts of the Spiral of Silence Theory Hans Mathias Kepplinger

44

5 Methodological Conundrums in Spiral of Silence Research Jörg Matthes and Andrew F. Hayes

54

6 Speaking in Spirals: An Updated Meta-Analysis of the Spiral of Silence Carroll J. Glynn and Michael E. Huge

65

vi

Contents

PART II

Media Fragmentation, Internet, and Public Opinion 7 Normative Implications of the Spiral of Silence: Hearing the Other Side in the 21st Century Diana C. Mutz and Laura Silver

73 75

8 Media and Public Opinion in a Fragmented Society Patricia Moy and Muzammil M. Hussain

92

9 Public Opinion Expression in Online Environments Patrick Rössler and Anne Schulz

101

10 Linking Social Network Analysis to the Spiral of Silence, Coorientation, and Political Discussion: The Intersection of Political Perceptions and Political Communication William P. Eveland, Jr.

119

PART III

Allied Research Traditions 11 Social Norms, Spirals of Silence and Framing Theory: An Argument for Considering Cross-Cultural Differences in Media Effects Research Timothy K. F. Fung and Dietram A. Scheufele 12 The Intersection of Third-Person Effect and Spiral of Silence Albert C. Gunther

129

131

145

13 Pluralistic Ignorance and the Spiral of Silence Meet: Mutual Lessons Jacob Shamir

153

14 Opinion Leadership and Public Opinion: Where Weak/Strong Media Paradigms Converge Gabriel Weimann

161

15 Vernacular Rhetoric, Considered Public Opinion and Modes of Publicity Gerard A. Hauser

169

Contents

vii

PART IV

Intercultural Contexts

177

16 World Public Opinion Frank Louis Rusciano

179

17 Cultural Orientation and the Spiral of Silence Sonny Rosenthal and Benjamin Hill Detenber

187

18 The Kuuki Theory and the Spiral of Silence Youichi Ito

201

A Personal Tribute

211

19 Elisabeth Noelle-Neumann and the Mysteries of Solitude Mihaly Csikszentmihalyi

213

Author Index Subject Index

218 225

This page intentionally left blank

TABLES AND FIGURES

Table 6.1

Correlations and Z-scores for Studies Contributing Data to the Meta-Analysis 68 Figure 7.1 Four ideal type media environments 84 Figure 9.1 Basic elements of the SoST, adapted to Esser’s model of sociological explanation (1999) 103 Figure 9.2 Dimensions of anonymity and identifiability in online and offl ine environments 107 Figure 9.3 The cognitive dimension of the SIDE model 110 Figure 9.4 Sources of norm orientation, depending on a person’s identifiability 111 Figure 9.5 Strength of norm orientation considering the dimension of anonymity 112 Figure 9.6 Strength of norm orientation considering both SIDE dimensions 113 Figure 9.7 Expectations of sanctions in online environments 114 Figure 10.1 Comparing data structures across research approaches 123 Figure 10.2 Multilevel spiral of silence research using whole network data 125

This page intentionally left blank

CONTRIBUTORS

Mihaly Csikszentmihalyi is Professor of Psychology and management at Cla-

remont Graduate University and was the former head of the department of psychology at the University of Chicago. He is the director of the Quality of Life Research Center (QLRC). Csikszentmihalyi is a member of the American Academy of Education, the American Academy of Arts and Sciences, and the National Academy of Leisure Studies. He published several books, e.g., Good Business: Flow, Leadership and the Making of Meaning. Benjamin Hill Detenber earned his PhD at Stanford University. He is cur-

rently an Associate Professor and Chair of the Wee Kim Wee School of Communication and Information at Nanyang Technological University, Singapore. His research interests include the social and psychological significance of media technologies, message processing, public opinion and political communication, and media ethics and education. His research has appeared in the Journal of Communication, Media Psychology, New Media & Society, Communication Research, and elsewhere. Wolfgang Donsbach is Professor of Communication at the Department of

Media and Communication at Dresden University of Technology, Germany. He has taught at the universities of Dortmund, Mainz, and Berlin in Germany and was a visiting professor at Columbia University, Syracuse University, University of Navarra, and Harvard University. He was president of WAPOR (1995–1996) and ICA (2004–2005). Since 2010 he is an ICA Fellow. His main research interests are in journalism, political communication, and public opinion.

xii Contributors

William P. Eveland, Jr. is Professor of Communication and (by courtesy) Political Science at The Ohio State University. His primary contributions to political communication research have been in the area of learning from the news media and communication technology and the understanding of communication influence on social reality perceptions. His current research examines the impact of social context, social networks, and the content of political conversation on political knowledge, perceptions, and participation. Timothy K. F. Fung, PhD, University of Wisconsin‒Madison, is an Assistant Professor in the Department of Communication Studies at Hong Kong Baptist University. His research interests include cultural influence on information processing and judgment, risk communication, and political communication. His work has received two top student paper awards from Association for Education in Journalism and Mass Communication (AEJMC). His research has appeared in Journal of Health Communication, Mass Communication & Society, New Media & Society, and Journal of Information Technology and Politics. Carroll J. Glynn, PhD, University of Wisconsin–Madison, is Professor and Director of the School of Communication, Ohio State University, Columbus, Ohio. Her research interests focus on understanding public opinion from a social and nor mative perspective. Albert C. Gunther, PhD, Stanford University, is Professor in the School of

Journalism and Mass Communication at the University of Wisconsin-Madison. His research interests focus on the psychology of the mass media audience— primarily in the context of scientific controversies or public health issues. Gunther publishes in most major communication journals and has received funding from numerous sources including the USDA, NSF, NIH, and the Robert Wood Johnson Foundation. Gerard A. Hauser is College Professor Emeritus of Distinction at the University of Colorado Boulder. His book publications include Introduction to Rhetorical Theory, Vernacular Voices: The Rhetoric of Publics and Public Sphere (recipient of NCA’s Marie Hochmuth Nichols Book Award), and Prisoners of Conscience: Moral Vernaculars of Political Agency. He is a Distinguished Scholar of the National Communication Association and a Fellow of the Rhetoric Society of America. He is Editor of Philosophy and Rhetoric. Andrew F. Hayes, PhD, Cornell University, is author of An Introduction to Medi-

ation, Moderation, and Conditional Process Analysis, Statistical Methods for Communication Science, coeditor of the Sage Sourcebook of Advanced Data Analysis Methods for Communication Research,  and has published  numerous  journal articles and book chapters in research methods, data analysis, and public opinion. He is one

Contributors

xiii

of the founding editors of  Communication Methods and Measures  and serves as editor-in-chief through 2015. Michael E. Huge, MA, Ohio State University, is a Research Associate with

School of Communication at Ohio State University. His current research focuses on perceptions of public opinion, how those perceptions drive behaviors, and how perceptions are shaped by media consumption. He has also published research on media portrayals of science issues, political contests, and political behavior. Muzammil M. Hussain, PhD, University of Washington, is Assistant Professor

of Global Media Studies at the University of Michigan’s Department of Communication Studies. His research focuses on comparative digital politics, and he is co-author of Democracy’s Fourth Wave? Digital Media and the Arab Spring (Oxford University Press, 2013). Youichi Ito is Dean and Professor at the Graduate School of Global Communication and Culture, Akita International University in Akita City, Japan. He is also a professor emeritus at Keio University in Tokyo.  He served as President of the Japan Society of Information and Communication Research from 2004 through 2008.    He has been an ICA fellow since 2009.    His main research interests are political communication, international communication, and information society research. Hans Mathias Kepplinger, Professor Dr. phil. was assistant lecturer at the Uni-

versity of Mainz and Heisenberg scholarship holder of the German Science Foundation. He has been a research fellow at the University of California, Berkeley, guest professor at the Southern Illinois University and at the Harvard University, and guest lecturer at the University of Tunis. Kepplinger was Director of the Institut fuer Publizistik at the University of Mainz, Dean of the Faculty of Social Sciences at the University of Mainz, and member of the Senate of the University of Mainz. He is corresponding editor of the European Journal of Communication and member of the advisory board of Political Communication Research and Journal of Communication. Jörg Matthes, PhD, University of Zurich, is Professor of Communication Sci-

ence at the Department of Communication, University of Vienna, Austria. His research focuses on advertising effects, the process of public opinion formation, news framing, and empirical methods. Since 2011, he has chaired the Methods division of the German Communication Association (DGPuK), and he is Associate Editor of the journal Communication Methods & Measures. From 2005–2010, he was also chair of the Methods division of the Swiss Communication Association (SGKM).

xiv

Contributors

Denis McQuail is Professor Emeritus and Honorary Fellow at the Amsterdam School of Communication Research, University of Amsterdam. His publications and research have had particular reference to mass media effects and audiences, media policy and theory of media. Most recent publications include: Normative Theory of the Media (with others; University of Illinois Press, 2009); McQuail’s Mass Communication Theory, 6th edition (Sage, 2010); and Journalism and Society (Sage, 2013). Patricia Moy, PhD, University of Wisconsin, is the Christy Cressey Professor of Communication and Adjunct Professor of Political Science at the University of Washington. Her research focuses on communication and citizenship, public opinion, and media effects. She is editor of Public Opinion Quarterly, editor-inchief of Oxford Bibliographies: Communication, and Vice President of the World Association for Public Opinion Research. Diana C. Mutz holds the Samuel Stouffer Chair in Political Science and Com-

munication at the University of Pennsylvania, where she serves as Director of the Institute for the Study of Citizens and Politics at the Annenberg Public Policy Center. In 2011, she received the Lifetime Career Achievement Award in Political Communication. She was inducted as a Fellow of the American Academy of Arts and Sciences in 2008. Mutz is the author of numerous journal articles and three books including Impersonal Influence, Hearing the Other Side, and Population-based Survey Experiments. Elisabeth Noelle-Neumann (1916‒2010) was the founder (1947) and long-

standing director of the Allensbach Institute. She taught at the universities of Berlin, Mainz, and Munich and was a visiting professor at the University of Chicago. She was president of WAPOR (1979‒1980) and a founding editor of the International Journal of Public Opinion Research. Noelle-Neumann was an important media researcher and political adviser in Germany. Her spiral of silence theory is one of the most influential communication theories. Sonny Rosenthal is a postdoctoral research fellow in the Wee Kim Wee School

of Communication and Information at Nanyang Technological University. He earned his PhD in advertising from the University of Texas at Austin, where he worked under a fellowship from the National Science Foundation. His research interests include social change, public understanding of science, environmental sustainability, and research methodology. Patrick Rössler is Professor of Communication and holds a chair for empirical

research/methods at the Department of Media and Communication at the University of Erfurt, where he is currently serving as Vice President for Research. In 2004 he had a DAAD-visiting professorship at the Annenberg School for

Contributors

xv

Communication, University of Southern California. He was the former head of the German Association of Communication (DGPuK) and president of the International Federation of Communication Associations (IFCA). Rössler has edited several books, articles, and papers. Frank Louis Rusciano, PhD, University of Chicago, is Professor of Political Science and Director of Global Studies at Rider University, a Fulbright Scholar University of Ulster Policy Studies Award recipient, and a three-time Alexander von Humboldt Fellow. His books include Isolation and Paradox: Defining “the Public” in Modern Political Analysis (Greenwood, 1989), World Opinion and the Emerging International Order (Praeger, 1998), and Global Rage after the Cold War (Palgrave, 2006). Charles T. Salmon is Professor and Director of Graduate Research Programmes

in the Wee Kim Wee School of Communication and Information at Nanyang Technological University, Singapore. Formerly he was the Dean of the College of Communication Arts and Sciences at Michigan State University, where he held the Ellis N. Brandt Chair. He has been a Rockefeller Foundation Scholar in Bellagio, Italy, and a Fulbright Scholar in Tel Aviv, Israel. His books include Public Opinion and the Communication of Consent (with Theodore Glasser; Guilford Press, 1995). Dietram A. Scheufele holds the John E. Ross Chair in Science Communication at the University of Wisconsin, Madison, and is Co-PI of the Center for Nanotechnology in Society at Arizona State University. An elected fellow of the American Association for the Advancement of Science and the Wisconsin Academy of Sciences, Arts & Letters, he is listed by Microsoft Academic Search as one of the 10 most cited researchers in the communication discipline. His most recent work focuses on the role that social media and other emerging modes of communication play in our society.  Scheufele has been a tenured faculty member at Cornell University, a Shorenstein fellow at Harvard University, and a DAAD Visiting Professor at the Technische Universität Dresden. Anne Schulz, BA, is a graduated of the University of Erfurt). Currently she is

MA Student and Graduate Research Assistant at the IPMZ Institute of Mass Communication and Media Research at the University of Zurich. Schulz was one assistant for the PSP-Study spiral of silence online and won the Best Paper Award of the DGPuK-occupational group reception and effect research in 2011. Her research focuses on public opinion formation, online communication, media reception, and media effects. In her master’s thesis she is dealing with the effects of consonant news reporting on hostile media perception.

xvi Contributors

Jacob Shamir  has recently retired from the Hebrew University of Jerusalem,

where he was a Professor in the Department of Communication and Journalism. His research focuses on public opinion, political communication, media and confl ict resolution. He has published numerous articles and several books in these areas, including Palestinian and Israeli Public Opinion: The Public Imperative in the Second Intifada  (Indiana University Press, 2010) and  The Anatomy of Public Opinion (University of Michigan Press, 2010). Laura Silver is a PhD student in communications and political science at the Annenberg School and at the Department of Political Science at the University of Pennsylvania. Her research focuses on comparative political communication, public opinion, and perceptions of foreign countries and foreign policy. Yariv Tsfati, PhD, University of Pennsylvania, is Associate Professor at the

Department of Communication, University of Haifa, Israel. His research focuses on various facets of public opinion, in particular on trust in media, the third person effect, and campaign effects. He is a past Vice-Chair (2009‒2010) and Chair (2011‒2012) of the Political Communication Division of the International Communication Association (ICA). He currently serves as editor for the International Journal for Public Opinion Research. Gabriel Weimann is a Professor of Communication at the Department of

Communication at Haifa University, Israel.  His research interests include the study of media effects, political campaigns, new media technologies and their social impact, persuasion and influence, media and public opinion, modern terrorism and the mass media. He published eight books: Communicating Unreality  (Sage, 2000);  The Influentials: People Who Influence People  (State University of New York Press, 1995); The Theater of Terror (Longman, 1994); Hate on Trial (Mosaic, 1986); The Singaporean Enigma (Tzivonim, 2001); Terror On the Internet  (USIP Press, 2006);  Freedom and Terror (Routledge, 2011); and  Social Research in Israel  (Tzivonim, 2013). His more than 160 papers and research reports have been published in numerous scientific journals and books.

1 THE LEGACY OF SPIRAL OF SILENCE THEORY An Introduction Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

Elisabeth Noelle-Neumann passed away on March 25, 2010, at the age of 93. Her death ended a highly visible career as a social scientist, an entrepreneur, a political consultant, and a journalist not only in Germany but internationally as well. She was a professor of communications (at the University of Mainz) and one of the fi rst to introduce empirical methods to German communication research. She was the owner of the Institut für Demoskopie Allensbach, an economically successful and methodologically innovative survey and marketing research fi rm. She was a political consultant who, from the very beginning of post-war Germany, supplied evidence on the Germans’ public opinion to the chancellors from Adenauer to Kohl and Merkel. And she was a journalist, in her fi rst career during the Third Reich, and afterwards as what she called a “survey research correspondent” (demoskopischer Korrespondent). But her name will primarily be remembered, at least among academics, as a theorist and specifically as the author of the spiral of silence theory. There have been many controversial discussions in publications, at conferences and in seminars about this theory. But there is certainly accordance, even among the most ferocious critics of this theory, that it has been one of the most influential of all theories developed in communication research and political communication over the last half century. “Influential” can, of course, mean very many different things. What makes a theory influential? We believe that there are five criteria: recognition, acceptance, integration, evidence, and practical relevance. How does the spiral of silence theory fare on each of them? This book’s objective is to address this question. Each of the chapters will speak to at least one of these aspects from different perspectives and with different results. Here we will explain these dimensions, summarize the issues and the evidence, and link them with the chapters that follow.

2

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

Recognition Statistical Evidence One measure of a theory’s recognition by a field is the number of times it is cited, with the caveat that recognition does not necessarily imply acceptance. Indeed, a review of the spiral of silence 20 years ago concluded that the theory was spawning as many critiques as empirical studies (Salmon & Moh, 1992). As of the time of writing the present chapter, Google Scholar showed that Noelle-Neumann’s most cited journal article in English (Noelle-Neumann, 1974) has been cited more than 700 times, while her English language book (Noelle-Neumann, 1984) had been cited more than 1,200 times. To put these figures in an order-of-magnitude context, Google Scholar at the same time listed more than 40,000 citations for Everett Rogers’ Diffusion of Innovations, more than 4,000 citations for McCombs and Shaw’s (1972) seminal journal article on agenda setting, and more than 400 citations for Ball-Rokeach’s and DeFleur’s (1976) original article on media dependency theory. To get a better sense of the impact of the spiral of silence within and outside the discipline of communication research, we examined a sample of 500 works citing Noelle-Neumann (1974) in Google Scholar. Examining the national, disciplinary, and intellectual identity of the authors citing Noelle-Neumann is telling when one wishes to discuss the impact of spiral of silence theory. Of the 142 communication journal articles citing “spiral” in our sample of 500 items (that included in addition books, dissertations, conference papers, and journal articles from other disciplines), 66 (47%) were empirical quantitative papers, while the rest did not include quantitative data. Only 41 (29%) were published in traditional political communication and public opinion outlets, such as Public Opinion Quarterly, International Journal of Public Opinion Research, and Political Communication. Very many of the articles citing Noelle-Neumann’s 1974 Journal of Communication article and appearing in general communication journals such as Journal of Communication, Communication Theory, or Communication Research dealt with media effects in the political context. But this is not to say in any way that “spiral” is only useful for political communication scholars. Out of the 142 citations to Noelle-Neumann in communication journals in our sample, 11 appeared in health communication journals, 3 in Science Communication, and 3 in the Journal of Media and Religion. The spiral of silence was also cited in Strategic Communication, Media Psychology, Media, War & Conflict, and Visual Communication Quarterly. Despite its reputation as an “administrative” theory (i.e. one that is of a practical value for influencing the political and social process), “spiral” has also been cited by journals highlighting critical, cultural, and rhetorical traditions such as Critical Studies in Media Communication, Media Culture & Society, Communication Culture & Critique, Discourse & Society, and Quarterly Journal of Speech. Together, these references accounted for 11% of the citings to the theory in communication journals in our sample.

The Legacy of Spiral of Silence Theory

3

In terms of geographic reach, the 555 authors and coauthors of our sample of 500 citing items originate from 32 countries (based on their affi liation at the time of publication). Most authors citing the spiral of silence (63%) are North American (333 out of the 555 authors are affi liated with U.S. institutions). European scholars account for 23% of citations. Eight percent of the citations came from authors based in Australia and New Zealand, 3% came from Asia, 2% came from the Middle East, 1% from Central or South America, and only 2 citations from Africa. This distribution is quite similar to the global distribution of authors publishing in major journals in the discipline at large (Lauf, 2005). In any case, as far as one can tell from an examination of citation patterns, spiral of silence is much more than a German theory; only 17 (about 3%) of the articles citing spiral of silence were authored by scholars affi liated with German institutions. Reasons for the Theory’s Recognition Why has the spiral of silence theory received such varied recognition in the field? We see mainly four reasons: Noelle-Neumann’s publication strategy, the role of the theory within a historic paradigm change in the field, the theory’s relationship to other theories, and its provocative character. First, Noelle-Neumann practiced something that, in the 1970s and 80s, was highly uncommon for German and other European scholars whose fi rst language was not English: she published her work almost simultaneously in her own language and in English, the only way one could (and still can) get recognition beyond the boundaries of one’s own culture. Her book on the theory fi rst appeared in 1980. With Piper Publishing she chose not a primarily academic publisher but one who was known for offering serious non-fiction publications for a broader intellectual audience. Given her at that time well-established connections with the University of Chicago (where she taught as a visiting professor), she managed to get the English translation out already in 1984 as hardcover and 2 years later as paperback edition with the prestigious University of Chicago Press. Although later translated into 11 languages, it is this English language publication that brought recognition to the spiral of silence theory. She followed up on the book publication with two further English language articles in widely read publications, thus increasing awareness of her theory, although the tone of the academic discussions had already turned somehow controversial (Noelle-Neumann, 1985, 1991). Noelle-Neumann’s decision to seek international outlets for disseminating her ideas and fi ndings early on was motivated by academic professionalism and strategy. Being well connected in international associations of the field like the World Association for Public Opinion Research (WAPOR), the American Association for Public Opinion Research (AAPOR), or the International Association for Mass Communication Research (IAMCR, today called Media

4

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

and Communication Research), it was quite normal for her to think globally. Her peers were not restricted to the (then still small) community of German communication researchers. She traveled to conferences and talked to scholars all over the world, and therefore it was very normal for her to also address these international colleagues in her publications, which meant she needed to publish in English. Given the strong beliefs she had about the validity of her theory, it is probably also fair to say that she employed the English language publications as a strategy to get attention for her theory, recognition, and reputation. Second, as mentioned above, Noelle-Neumann’s challenged the then-dominant paradigm of minimal media effects. As Denis McQuail writes in his chapter in this book, Noelle-Neumann not only “provided the slogan for the banner of paradigm change with the title of her article ‘Return to the concept of powerful mass media’ but offered theory and methods for investigating the posited effects.” The spiral of silence theory, agenda setting, and cultivation, although very different in terms of reasoning and complexity, were well received by the field that had struggled for so long with results from media effects research that were, fi rst, running against the researchers’ intuitive assumptions, and, second, made the whole field somehow socially less relevant. This identity crisis of communications as a field (Donsbach, 2006) and particularly its media effects research certainly contributed to the early awareness that the spiral of silence theory received among communication scholars. In her original articulation of the theory, Noelle-Neumann (1973) described mass media as ubiquitous and consonant, and media content as reflecting the political leanings of journalists who, as a group, were more liberal than the average citizen. She juxtaposed this view of mass media with the individual’s interpersonal communication behavior in social settings (speaking up or being silent). This points to a third reason for the awareness that the theory received, namely that it combines interpersonal communication with mass communication, macro and micro levels of analysis, content and audiences, sociology, and psychology. The theory offers insights for journalism scholars interested in media bias as well as for media psychologists interested in media effects, for cognitive psychologists concerned with individual perception, and for social psychologists interested in collective conformity. It carries important normative implications, but also offers hardcore statistical evidence, and practical relevance for the conduct and publication of public opinion polls. It is relevant for scholars studying communication and elections but also for those focusing on deliberative democracy. The fact that Noelle-Neumann highlighted the role of the spiral of silence in the transition from theories of limited effects back to theories of powerful media makes it a useful citation for scholars working on the history of communication research as well. A fourth reason for attention to the spiral of silence theory is likely attributable to controversial publicity surrounding Noelle-Neumann herself and in particular her writings from the 1930s and 40s while studying in the United

The Legacy of Spiral of Silence Theory

5

States, completing her dissertation at the University of Berlin, and working for the newspaper Das Reich during World War II. This controversy stems from a line of criticism (e.g., Bogart, 1991; Simpson, 1996) that concluded that the impetus for the spiral of silence theory arose from Noelle-Neumann’s work and insights during the Nazi era rather than from her work and insights as a social scientist and pollster in post-war Germany. Critic Leo Bogart (1991) accused Noelle-Neumann of supporting Nazi ideology and exhorted the scholarly community to shun and hence “silence” the spiral of silence theory. Christopher Simpson (1997) set up a Web-based archive to make Noelle-Neumann’s most controversial writings from 1935 to 1945 a matter of public record, raising questions about her involvement in Nazi propaganda efforts. These criticisms and a subsequent high-profi le termination of Noelle-Neumann’s visiting professorship at the University of Chicago directed further attention, albeit negative, to her and to her research. The criticisms rarely, if ever, focused on contemporary scientific merits of the theory itself, but rather raised a series of critical questions about Noelle-Neumann’s beliefs, responsibility, and writings as a citizen, student, journalist, and researcher living and working in the Nazi era and regime, and ultimately authoring a theory in which fear of one’s environment—rather than enlightenment or empowerment—is viewed as the motivation for opinion expression versus silence. Criticisms and defenses of Noelle-Neumann’s early professional career and writings have been aired and debated in academic conferences (most notably, the 1997 convention of the International Communication Association), the popular press (e.g., Honan, 1997; Miner, 1991), and the pages of scholarly journals (e.g., Bogart, 1991; Kepplinger, 1997; Simpson, 1997). Acceptance The spiral of silence theory is a bold theory. It claims to have roots in at least six more or less separate fields and it presents a provocation in at least four ways. Let us start with the intellectual roots. Intellectual Roots The fi rst tradition to which Noelle-Neumann traced her theory’s tradition back is descriptions of public opinion, if not with the term itself with the phenomenon, in political philosophy and literature. Already in her 1980 (in German) and 1984 (in English) book she spent several chapters presenting evidence for the claim that what she sees as public opinion (i.e., the socio-psychological phenomenon) had always been there, be it in the writings of political thinkers like Aristotle, Locke, and Hume or writers like the 18th-century Frenchman Choderlos de Laclos or the 20th-century Swiss novelist Max Frisch. Others who had dealt with public opinion, predominantly German sociologist Jürgen

6

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

Habermas in his “Structural Transformation of the Public Sphere” (1962/1989), would just have dropped the ball by neglecting the socio-psychological side (that had been there for centuries) and reducing public opinion to its role of a social institution juxtaposed to government. The second root is theories of social perception. As Noelle-Neumann’s theory claims that people have an ability to assess majority and minority opinions in their social environment, it fi nds itself in the neighborhood of other theories that either claim or refute this trait. The basic idea goes back to Cooley (1902/1983) who hypothesized that a person’s “self-idea” has three elements: an imagination of what we are to another person, an imagination of that person’s judgment of us, and some sort of self-feeling such as pride or mortification (see Eveland & Glynn, 2008). Being able to make such assessments or, as NoelleNeumann coined it, having a “quasi-statistical sense,” docked on to general theories of pluralistic ignorance (most of which were developed at the same time, see Shamir’s Chapter 13, this volume) as well as those that explain the role of media in social perception such as to third-person effect (Gunther, Chapter 12, this volume), the hostile media phenomenon (see also Chapter 12 as well as Mutz & Silver’s Chapter 7, this volume), and even Gerbner’s cultivation theory (see McQuail’s Chapter 3, this volume). A third and related root is the construct of social control. Again, NoelleNeumann criticized the social sciences for having misled the field by renaming the most important aspect of “public opinion” to “social control” and thereby, first, separating it from the political realm, and second, dysfunctionally amputating the construct of public opinion. What sociologists like Edward Ross would describe as social control would be nothing else but the effects of public opinion, restricted to an apolitical sphere. Nevertheless, the sociological literature offered to her plenty of examples how the social control mechanisms were exerted. A fourth root of the theory (and again related to the previous) is the social nature of man. In Noelle-Neumann’s writing this has three dimensions: our need for the presence of others, judgments in coordination and communication with others, and our dependence on the judgments of others about ourselves. The need for the presence of others is documented primarily in the “fear of loneliness” that can be traced back anthropologically and explained by evolutionary theory (see Csikszentmihalyi’s Chapter 19 in this volume). The role of others for perceptions and more so judgments had evidence in sociopsychological experiments like Sherif ’s or Asch’s studies on group decisionmaking (see Fung & Scheufele, Chapter 11, this volume) or, more recently by Hardin and Higgins’ (1996) theory of “shared reality.” The strongest and most visible indicator of our emotional dependence on how others look at us is research on embarrassment, taken by Noelle-Neumann as a cross-cultural indicator for the social nature of man and the stress that this nature gets into when challenged by public opinion (see Chapter 18 by Ito and Chapter 7 by Mutz & Silver, this volume).

The Legacy of Spiral of Silence Theory

7

The fi fth root of spiral of silence theory is election research. Being herself a pollster and in the business of election forecasts (through the Allensbach Institute), Noelle-Neumann was concerned with reasons for opinion changes and election outcomes. She reports that the “enigma” that she fi rst found in the 1965 German general elections (and which she describes early on in her 1984 and 1989 books) is the key to her subsequent theory: Voting intentions stayed unchanged while the expectation about the presumed winner of the election (climate of opinion) had changed dramatically. Here, Noelle-Neumann developed the hypothesis that perceptions of others’ opinions might influence one’s own behavior thus explaining a phenomenon that had already been around in election research (i.e. the “last minute swing”). The fi nal root of the theory is, of course, communication research, and here particularly media effects research. Elisabeth Noelle-Neumann was very much annoyed by the inability of communication research to fi nd clear evidence for media effects (see McQuail’s Chapter 3 and Kepplinger’s Chapter 4, this volume). The role of the media in the spiral of silence theory (i.e., being one source for the perception of climates of opinion), was only one of several paths that she pursued in order to prove that the minimal effects paradigm was not the last word. Nevertheless, most of her projects centered round the question by what means the media can convey impressions of what is a majority/accepted and what is a majority/unaccepted opinion on a given issue. Provocations While it becomes clear that the complexity of Noelle-Neumann’s system of hypotheses offered many options for integration with other theories (see the section below), it also created many gateways for criticism. The most obvious one is the theory’s breadth (i.e. its nature of a macro theory), incorporating several of the existing theories in the social sciences. For many, this theory has carried a certain posture of arrogance, claiming that its author knew better and pointing out where other authors had misled the field. Moreover, NoelleNeumann claimed that the evidence for what she holds true was there throughout scientific and philosophical history, making her arguments even harder to digest by other authors. Another feature of this boldness is the claim that the theory works across all cultures. This ran counter to many other theories, particularly popular in the second half of the 20th century, that human behavior is very much bound to specific cultural factors and that the “hard-wiredness” of man is very limited. The second provocation is what is perceived by many commentators as Noelle-Neumann’s concept of the individual. The spiral of silence theory assumes that an individual withholds his or her opinion when confronted with a dissonant climate of opinion. This, of course, does not comply with the ideal of the citizen in democratic theory. But Noelle-Neumann faced the same problem of

8

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

the gap between norm and evidence as others in the business of media effects. The fact that the paradigm of minimal media effects, stipulated by the concept of selective exposure in the 1940s dominated communication research for more than 30 years (and often against existing evidence) can only be explained by the norm-reality phenomenon. This paradigm not only legitimized bigmoney media but also saved the image of man in society. As Katz (1987) wrote, “Lazarsfeld and company concluded that it is a good thing for democracy that people can fend off media influence and implied that the crowd may be less lonely and less vulnerable than mass society theorists had led us to believe” (S26–S27). The problem with Noelle-Neumann’s theory was and is that it had opposite evidence, that it showed man not to be as ideal as conceptualized in democratic theory. The problem of many commentators, on the other side, was the fact that they mistook empirical evidence as the normative conception of the author. Spiral of silence theory also provoked because it was embedded in NoelleNeumann’s paradigm of powerful mass media. While many authors (see McQuail’s Chapter 3, this volume) praise Noelle-Neumann for having contributed to a paradigm change and re-orientation of media effects research, the new paradigm was not embraced by all actors. Those scholars whose work stood for the opposite paradigm felt challenged by this new assumption. Journalists and media moguls who had fended off societal and political demands for more responsibility with the “no effects” notion, feared that new discussions about the role and the control of media would arise. Many heated discussions on Noelle-Neumann’s publications (particularly her publications on the role of television in elections) can only be explained on the basis of journalists’ problem of legitimacy. Finally, the theory also provoked because of its applicability and application to practical politics. At least this was the case in Germany when the conservative Christian Democrats (to whom Noelle-Neumann was a public opinion consultant) focused their strategy in the 1976 campaign on the outspokenness of their followers. Many social scientists fi nd problems with practical applications of theories and particularly with scholars getting personally involved, at least when such consultancy concerns conservative parties. Critique These provocations inherent in the spiral of silence theory probably have stimulated more scholars to grapple with it theoretically and empirically than would have been done with other theories. This motivation is still alive today—as this book and a recent special issue of the International Journal of Public Opinion show (see particularly the article by Lang & Lang, 2012). This is not the place to give an account of all the theoretical and empirical evaluations of the theory. Matthes and Hayes (Chapter 5, this volume) summarize what they call “the

The Legacy of Spiral of Silence Theory

9

methodological conundrums” in spiral of silence research and Glynn and Huge (Chapter 6, this volume) present the results of a meta-analysis of empirical studies based on the theory. Many of the other chapters, when relating spiral of silence to other theories, deal in one way or another with theoretical and conceptual problems inherent in the theory. Not one study so far has been able to put the spiral of silence theory in full to an empirical test. The theory’s architecture requires a multi-method and a longitudinal design using topical (and often volatile) political issues with a high emotional impetus. Applying panel surveys plus content analyses over a longer period of time (and controlling for many other variables) is a challenge for social researchers. As a consequence, most studies have tested only fragments of the theory (usually the relationship between perceived public opinion and outspokenness), disregarding longitudinal effects on the aggregate and/or the role of the media (Salmon & Glynn, 2009). Thus, what we hold empirically in hand is slim (see the chapters mentioned above). Theoretical criticism appeared early on and concentrated in the 1980s and ’90s. It mainly focused on four dimensions. The fi rst and most often cited is the role of reference groups and social networks. Noelle-Neumann’s fi xation on the anonymous public would have blinded her for the equal if not stronger power of reference groups when it comes to social perception and outspokenness. There are only a few studies that have tested this influence comparatively and the results are not conclusive (see Glynn & Park, 1997; Oshagan, 1996). It is clear that Noelle-Neuman, in her fi rst layout of the theory, did not deal explicitly with the competing roles of publics and reference groups. Another camp of critics challenged Noelle-Neumann’s notion of the “quasistatistical organ” (she later called it “sense”) and supplied evidence (and explanations) as to why people cannot perceive the distribution of opinions in society (see Shamir’s Chapter 13, this volume). Indeed, at prima facie empirical results on a “pluralistic ignorance” (i.e., a misperception of the true distribution of opinions) seem to run counter to Noelle-Neumann’s hypotheses (see, for instance, O’Gorman & Garry, 1976). However, the media effects part of spiral of silence theory does indeed include such a perception—caused by a slanted coverage of the news media. Noelle-Neumann’s basic hypothesis was not so much about the accuracy of perceptions but the fact that people (a) are motivated to screen their social environment (see the theory’s roots mentioned above), (b) make assessments about what is a majority opinion, and (c) recognize changes. Thus, although Noelle-Neumann somehow failed to really clarify her notion of a “quasi-statistical sense” in terms of other theories—such as false consensus effect, social projection, looking glass perception, overestimate of consensus, false idiosyncrasy effect, disowning projection (see McLeod, Detenber, & Eveland, 2001)—their existence does not jeopardize the basic assumption. There are just many reasons why people perceive their social environment the way they do, the media being just one of many factors influencing the result. The

10

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

key hypothesis in the spiral of silence theory—that people’s readiness to talk about their own opinion is influenced by this perception—stays unaffected by such theories in its neighborhood. A third avenue of criticism has been Noelle-Neumann’s concentration on the “fear of isolation” as motivating engine for human behavior in social settings. Some of this criticism is based on the author’s transfer of evidence from small group experiments to the behaviors in the public. There would exist competing motivations to fear of isolation such as the individual’s personal interest (as in cost-benefit-calculations; see Taylor, 1982), and there could be other reasons for conformity such as attraction to or identification with a person or group (see, for instance, Salmon & Kline, 1985). Finally, several authors criticized that an individual’s personal characteristics or traits would easily overrule fear of isolation and decide about whether a person would prefer to speak out or stay silent (see, for instance, Lasorsa, 1991). Most of the arguments brought forward by other authors were certainly valid in the sense that they differentiated the complex field of causal relationships in humans’ behaviors, particularly in social perception and interpersonal communication. There is and there will never be one single construct explaining this behavior, be it the assumptions made in the spiral of silence theory or by any other theory. The fact that so many authors felt stimulated to challenge Noelle-Neumann’s theory had a lot to do with the way she presented her theory: bold and stereotypical. However, by often neglecting necessary subtleties and differentiations in her own writing Noelle-Neumann managed at the same time to draw attention to her theory. This “reduction of complexity” (Lippmann, 1922) in social reality accomplished that her basic ideas were communicated to the scientific community and beyond. She often admitted the validity of the many theories “competing” with her spiral of silence theory but held that only the competitive explanation of variance would bring us closer to understanding human behavior. Integration Another dimension for assessing a theory’s influence lies not in the fact that it is mentioned or cited, but in the theory’s epistemological, methodological and substantive linkage to existing theories or those that are developed simultaneously. Spiral of silence is linked in many ways to other lines of communication research as the chapters in our volume demonstrate. As Gunther argues in Chapter 12, both spiral of silence and the third-person effect traditions separate self from others and propose that media are powerful in shaping what we think others think. As Gunther notes, Davison (1983) recognized the resemblance between the third person effect and the spiral of silence in his original third-person effect paper. Similarly, in Chapter 13, Shamir reviews NoelleNeumann’s claims about perception of majority opinion with predictions and

The Legacy of Spiral of Silence Theory

11

fi ndings from research on pluralistic ignorance, and examines the compatibility of these two streams of research. In Chapter 10, Eveland integrates spiral of silence with concepts from McLeod and Chaffee’s (1973) coorientation model and with research on political discussion. Both “spiral” and coorientation focus on the incongruence between perception and accuracy. Despite some divergences in their focus, both “spiral” and research on political conversation concur that most political conversation takes place with likeminded individuals. While there are clear lines of convergence between them, Eveland argues that the three lines of research could benefit from further integration both theoretically and empirically but also methodologically. All three lines of research, Eveland argues, could benefit from the insights and methodologies developed by social network approaches. In Chapter 3, McQuail positions the spiral of silence within a tradition highlighting the dominance of media and the inability of audiences to resist their influence. As Weimann argues in Chapter 14, this tradition is supposedly alienated from the notions of the two-step flow and opinion leadership, typically said to highlight the power of the audience and the limited influence of media. Weimann feels that separating spiral of silence and the two-step flow into divergent traditions does not do justice to both theories. Beyond the obvious shared interest in the intersection of interpersonal and mass-mediated influences, both research traditions converge in their fi ndings, Weimann argues. He notes that research on the two-step flow fi nds that the opinion leaders enhance rather than limit the power of media over audiences. The convergence between the theories is also evidenced, according to Weimann, by Elisabeth NoelleNeumann’s successful attempts to develop a scale measuring opinion leadership and to integrate them into her research. The connection between spiral of silence to pluralistic ignorance and the third person effect and even its connections to the two-step flow theory and to coorientation research are more obvious (and are manifested more frequently in cross-citations) than its connection with framing research. As Fung and Scheufele argue in Chapter 11, although the spiral of silence and framing effect theories belong to different unconnected research traditions, both share the idea that socially-negotiated norms or schemas are critical for explaining the influence of socio-cultural environments on individual responses to opinion disagreement and on mediated message interpretation. Critics of the spiral of silence theory often imply that it does not work outside the German context. The cross-cultural applicability of the theory is discussed in depth in the chapters by Detenber and Rosenthal (Chapter 17) and Fung and Scheufele (Chapter 11) in this volume, and both chapters argue that the theory works better in collectivistic, as compared to individualistic, cultures. But while there is some debate about the applicability of “spiral” across contexts, there is no doubt that the theory had influenced scholars over the world as described earlier in this chapter.

12 Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

Evidence Two of the spiral of silence’s main arguments have enjoyed considerable amounts of research in the field of communication. First, Noelle-Neumann’s prediction that exposure to media shapes audience assessment of the social climate of opinion has been tested in both quasi-experimental (Mutz & Soss, 1997) and experimental settings (e.g., Perry & Gonzenbach, 1997). Results supported the hypotheses, but effect sizes were rather small. Second, the hypothesis that perception of supportive opinion climate is associated with opinion expression, has been tested in dozens of studies. In Chapter 6, Glynn and Huge located 55 such studies and used meta-analysis to assess the strength of the association across studies. Their results indicated that the average effect size is quite small (overall Pearson r = .053, SD = 0.58, 95% CI = .042 to .064) but still statistically significant (p < .0001, fail-safe N = 584). Even if we take the upper band of the confidence interval, perceived opinion climate explains less than 1% of the variance in willingness to express opinion. In addition to this small effect size, Matthes and Hayes (Chapter 5, this volume) argue that both typical independent and dependent measures lack reliability and validity: Fear of isolation is typically assumed to be constant across respondents, perceived opinion climate is measured combining two single-item ordinal measures that lack reliability information, and willingness to express opinion is typically measured using hypothetical questions that do not present respondents with a realistic threat of social isolation in public situations. Matthes and Hayes further argue that almost all spiral of silence studies fail to combine content analytic data capturing the mediated opinion climate, and that the minority of studies examining media content do so at the aggregate level without taking into account individual differences in media exposure. This is related to an additional problem highlighted by the authors regarding some of Noelle-Neummann’s evidence: the ecological fallacy. The aggregate designs used by Noelle-Neumann do provide us with a sense about how the variables change over time at the aggregate level, but as Matthes and Hayes claim “we are unable to draw any conclusions about a spiraling process at the individual level.” Beyond these severe methodological problems, two additional issues loom: cross-cultural applicability, and causality. Matthes and Hayes claim that the applicability of Noelle-Neumann’s predictions in individualistic cultures should be questioned. This concern is echoed to some extent by the theoretical arguments in Detenberg and Rosenthal’s Chapter 17. Fung and Scheufele likewise make the argument that “spiral” is a culturally specific theory and demonstrate that other theories of media effects in our field (in particular framing) are likewise culturally specific. Both Detenberg and Rosenthal and Fung and Scheufele present very strong theoretical arguments regarding why people in collectivistic societies would be more sensitive to perceived opinion and social norms, compared to people in individualistic societies. Having said that,

The Legacy of Spiral of Silence Theory

13

Matthes and Hayes are correct in noting a serious problem in the few studies comparing spiral of silence effects across cultures: the failure to fi nd effects in individualistic societies might have resulted from the use of different issues in the different contexts. If we take into account Noelle-Neumann’s requirement about the issues at stake—they should be likewise controversial and raise moral or ethical issues to the same extent across context—it seems very difficult to design a comparative spiral of silence study due to these requirements, and hence despite strong theoretical arguments, our ability to infer from existing data that spiral of silence is better suited to collectivistic cultures is limited. The most daunting methodological concern raised by Matthes and Hayes has to do with time and causality. How could we know that it is opinion climate that shapes opinion expression and not the other way around, when most studies testing the spiral are cross-sectional? Matthes and Hayes argue that even the longitudinal studies fail to fully examine the spiraling process proposed by Noelle-Neumann. Taking into account the complex scenario necessary to examine the spiral of silence, Matthes and Hayes conclude that the full scope of the theory has not been comprehensively tested—and it probably never will be. With such a shaky empirical record, why has the spiral of silence continued to garner so much attention? It is hard to think of any other “media effects” theory that has been criticized, if not refuted, so many times by so many scholars that is still so widely studied. Part of the endurance of spiral of silence theory stems from the way it invigorated and stimulated research, partly because of its relevance in many subfields of communication, and partly because of the fact Noelle-Neumann was a leader in the tradition transforming communication research from naïve and rather simple theories of powerful audiences to theories offering a more dynamic and complex association between people and the media. In any case, though the theory is characterized by a number of methodological conundrums, it has remained a rich source of new questions and new variables in communication and public opinion research processes. Practical Relevance We mentioned earlier that one of the provocations of spiral of silence theory was and is its applicability in practical politics. Moreover, the author herself used her theory to give advice to a political party in campaigns (see NoelleNeumann, 1999; Petersen, 2007). The spiral of silence theory certainly is a child of modern social research in as much it makes clear predictions about the impact of some variables on some others. “Nothing is as practical as a good theory” is a famous quote that can be traced back to several sources (see Donsbach & Brade, 2011). It basically means that “theory” is no antagonism to practice but that instead proven intersubjective evidence is the best foundation for it. Assertions about the construction of reality are always assertions on the role of plan, and meaningful acting, some of it in the realm of politics (Streeck, 2008,

14

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

p. 6). Thus, a connection between the social sciences and practical politics is not a new phenomenon. On the contrary, one of the reasons for the development of the empirical social sciences was the goal to make political decisions more evidence-based, a “scientification of politics” caused by the emerging social problems brought through industrialization in the 19th century (Maus, 1973). Then, this new concept of social research and its tools like surveys and statistics had to be established against metaphysical notions of human beings and society. In the 20th century, a new battlefield arose between the empirical social sciences, epistemologically based on Popper’s philosophy of science, and critical theory. Critical theorists deliberately wanted to be “practical” in the sense of changing social reality to the better but denied this potential to, what they called, “administrative research,” the latter being in their view restricted to “social technology” applications. However, Adorno, Horkheimer, and Habermas mingled two dimensions: objective and method—or, in Reichenbach’s words—context of discovery and context of justification (Reichenbach, 1938). The intersubjective discovery of causal relationships between social variables is independent of its possible applications. Theories or laws can be applied for the sake of preserving existing power structures or for overthrowing the same structures. Again, “nothing is as practical as a good theory,” particularly when one wants to change the world. And, over time, social science theories have become ever more precise, being able to incorporate more and more intervening variables and antecedent conditions. The expectation of outcomes, of practical relevance, is the major reason why modern societies spend enormous amounts of money for funding social research. Nevertheless, many social scientists feel discomfort with applying theory to practice, at least when the researcher him- or herself gets involved in this application. There are mainly two reasons for this discomfort: (a) the assumption that research loses its independence when it becomes part of policy-making, and (b) the perception that researchers support the wrong goals. In the case of spiral of silence theory, both reasons apply. The fact that political parties, at least in Germany, used the spiral of silence/silent majority notion for developing campaign strategies made Noelle-Neumann’s theory much more visible than most other theories in social and electoral behavior. Research in other fields— such as the issues of smoking (Shanahan, Scheufele, Yang, & Hizi, 2004) and of homosexuals in the military (Gonzenbach, King, & Jablonski, 1999)—revealed additional areas of applications in highly debated areas. As for the “wrong goals,” it is probably not too far-fetched to surmise that some scholars had problems with Noelle-Neumann, particularly advising the conservative party in Germany. In the United States, two thirds of the social scientists describe themselves as “liberal” and only 18% as “conservative” (Zipp & Fenwick, 2006). We have no equivalent data on German scholars, but the distribution of political worldviews should not be too different there or in other parts of the world. Thus, the perceived legitimacy of practical applications of

The Legacy of Spiral of Silence Theory

15

social research is often more a question of cui bono than of general epistemological norms and values. Future Challenges Yet another challenge to spiral of silence research lies in the changing media landscape. The mass communication environment is becoming markedly different than the one assumed by Noelle-Neumann when formulating her theory in the early 1970s. Noelle-Neumann assumed consonance (i.e., a unified mass communication environment projecting a consistent picture of the opinion climate). Noelle-Neumann further assumed that audiences are unwilling and incapable to selectively expose themselves to likeminded media. Unwilling— because their quasi-statistical sense motivates them to scan the environment at large (and not an ideological niche) in search of cues for the opinion climate. Incapable—because, given a consonant media landscape, they simply did not have the option to selectively expose themselves to ideologically-congruent new media outlets. These assumptions are challenged by two interconnected changes in the media landscape, noted by Moy and Hussain in Chapter 8, and by Mutz and Silver in Chapter 7. The fi rst has to do with the development (mainly in the United States but also elsewhere) of extremely popular ideological television news outlets serving either conservative or liberal audiences, coupled with the rise of likewise popular and ideologically slanted political talk radio, and televised political comedy. The second is the rise of online communication with endless possibilities for ideologically similar or incongruent communication and ample indication for the opinion climate that bypasses mainstream media information and provide audiences (minorities included) with a sense that they are not alone and with information about concrete people who think like them (see also Metzger, 2009). As noted by Rössler and Schulz in Chapter 9, the anonymity provided by the Internet, reducing fears about opinion expression and social isolation, is a further development that may make at least some of Noelle-Neumann’s prediction seem obsolete. How the predictions evident in her theory will fare in this new fragmentized and polarized environment is a question for future empirical research. But another, more important set of questions has to do with the continued relevance of the theory for future generations of scholars. Will Noelle-Neumann’s questions, concepts, and methods continue to attract scholars from different fields and subfields and from all over the academic world in light of changes in the media landscape and the social processes that accompany these changes? Online communication has been here for a while and so have Fox News, MSNBC, and political talk radio. One of the ways to explore the continued relevance of the spiral of silence is to inspect past trends in citations of spiral of

16 Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

silence over time, and in particular, examine if citations to the theory decreased over the past decade. Among the articles in our sample, spiral of silence has been cited on average 10 times per year between 1984 and 1993. This figure had increased to approximately 18 cites/year between 1994 and 2003, and to approximately 23 cites/ year between 2004 and 2012 (when the popular Internet was already in its second decade, when Fox News was already extremely popular, and when media scholars were already talking about “the end of mass communication”; Chaffee & Metzger, 2001). The past is never a perfect predictor of future trends, but it is often not a bad one, and thus if the trend of citations to Noelle-Neumann captured by our sample is correct, the number of articles citing the spiral of silence should continue to increase in future decades. The fact that scholars increasingly refer to Noelle-Neumann’s early works despite the fact that the media landscape is becoming increasingly remote from the one assumed by Noelle-Neumann is telling. As long as scholars will be interested in what we think others think, as long as they will try to connect interpersonal to mediated (if not “mass”) communication, as long as they will be investigating societies with real or perceived minorities and majorities, as long as they will focus not only on whether or not people hold opinions but also on whether they are willing to publicly express these opinions, as long as scholars will be interested in publicity and conformity and public opinion as the glue that ties us all together—they will always fi nd the useful and popular toolkit of concepts, methodologies, hypotheses, and operationalizations offered by the spiral of silence theory at their disposal. References Ball-Rokeach, S. J., & DeFleur, M. (1976). A dependency model of mass-media effects. Communication Research, 3, 13–21. Bogart, L. (1991). The pollster & the Nazis. Commentary, 92(2), 47–49. Chaffee, S. H., & Metzger, M. J. (2001). The end of “mass” communication? Mass Communication & Society, 4(4), 365–379. Cooley, C. H. (1902/1983). Human nature and the social order. New Brunswick, NJ: Transaction Books. Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47, 1–15. Donsbach, W. (2006). The identity of communication research. Journal of Communication, 56(4), 437–448. Donsbach, W., & Brade, A. M. (2011). Nothing is as practical as a good theory: What communication research can offer to the practice of political communication. The International Journal of Press/Politics, 16(4), 508–522. Eveland, W. P., & Glynn, C. J. (2008). Theories on the perception of social reality. In W. Donsbach & M. W. Traugott (Eds.), Handbook of public opinion research (pp. 155–163). London, England: Sage. Glynn, C. J., & Park, J. E. (1997). Reference groups, opinion intensity and public opinion expression. International Journal of Public Opinion Research, 9(3), 233–247.

The Legacy of Spiral of Silence Theory

17

Gonzenbach, W. J., King, C., & Jablonski, P. (1999). Homosexuals and the military: An analysis of the spiral of silence. Howard Journal of Communication, 4, 281–296. Habermas, J. (1962/1989). The structural transformation of the public sphere: An inquiry into a category of bourgeois society, polity. Cambridge, MA: The MIT Press. Hardin, C. D., & Higgins, E. T. (1996). Shared reality. How social verification makes the subjective objective. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition (Vol. 3, pp. 28–84). New York, NY: Guilford. Honan, W. H. (1997). U.S. professor’s criticism of German scholar’s work stirs controversy. New York Times. Retrieved from http://www.nytimes.com/1997/08/27/us/us-professor-s-criticism-of-german-scholar-s-work-stirs-controversy.html Katz, E. (1987). Communication research since Lazarsfeld. Public Opinion Quarterly ,51, S25–S45. Kepplinger, H. M. (1997). Political correctness and academic principles: A reply to Simpson. Journal of Communication, 47(4), 102–117. Lang, K., & Lang, G. E. (2012). What is this thing we call public opinion? Reflections on the spiral of silence. International Journal of Public Opinion Research, 24(3), 368–386. Lasorsa, D. L. (1991). Political outspokenness: Factors working against the spiral of silence. Journalism Quarterly, 68(1/2), 131–140. Lauf, E. (2005). National diversity of major international journals in the field of communication. Journal of Communication, 55(1), 139–151. Lippmann, W. (1922). Public opinion. New York, NY: Harcourt Brace. Maus, H. (1973). Zur Vorgeschichte der empirischen Sozialforschung [A prehistory of social research]. In R. König (Ed.), Handbuch der empirischen Sozialforschung [Handbook of empirical social research] (vol. 1, 3rd ed., pp. 21–56). Stuttgart, Germany: dtv. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of the mass media. Public Opinion Quarterly, 36, 176–187. McLeod, D. M., Detenber, B., & Eveland, W. P. (2001). Behind the third-person-effect: Differentiating perceptual processes for self and other. Journal of Communication, 54(4), 678–695. McLeod J., & Chaffee, S. H. (1973). Interpersonal approaches to communication research. American Behavioral Scientist, 16, 469–499. Metzger, M. (2009). The study of media effects in the era of internet communication. In R. Nabi & M. B. Oliver (Eds.), The Sage handbook of media effects and processes (pp. 561–576). Thousand Oaks, CA: Sage. Miner, M. (1991). Woman with a past. The Reader, December 12. Retrieved from http://www. chicagoreader.com/chicago/woman-with-a-past/Content?oid=878794 Mutz, D., & Soss, J. (1997). Reading public opinion. The influences of news coverage on perceptions of public sentiments. Public Opinion Quarterly, 61, 431–451. Noelle-Neumann, E. (1973). Return to the concept of powerful mass media. Studies of Broadcasting, 9, 67–112. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24(2), 43–51. Noelle-Neumann, E. (1980). Die Schweigespirale. Öffentliche Meinung — unsere soziale Haut [The spiral of silence. Public opinion — our social skin]. München, Germany: Piper. Noelle-Neumann, E. (1984). The spiral of silence: Public opinion — our social skin. Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1985). The spiral of silence. A response. In K. R. Sanders, L. L. Kaid, & D. Nimmo (Eds.), Political communication yearbook (pp. 66–94). Carbondale: Southern Illinois University Press. Noelle-Neumann, E. (1991). The theory of public opinion. The concept of the spiral of silence. In J. A. Anderson (Ed.), Communication yearbook 14 (pp. 256–287). Newbury Park, CA: Sage. Noelle-Neumann, E. (1999). Wahlkampf seit 1995. In drei Stufen zum Wahlsieg [Campaigning since 1995. In three steps to victory]. In E. Noelle-Neumann, H. M. Kepplinger, & W. Donsbach (Eds.), Kampa. Meinungsklima und Medienwirkung im Bundestagswahlkampf 1998

18

Wolfgang Donsbach, Yariv Tsfati, and Charles T. Salmon

[Kampa. Climate of opinion and media effects in the German general elections 1998] (pp. 172–180). Freiburg, Germany: Alber. O‘Gorman, H. J., & Garry, S. L. (1976). Pluralistic ignorance — A replication and extension. Public Opinion Quarterly, 40(4), 449–458. Oshagan, H. (1996). Reference group influence on opinion expression. International Journal of Public Opinion Research, 8(4), 335–354. Perry, S. D., & Gonzenbach, W. J. (1997). Effects of news exemplification extended: Considerations of controversiality and perceived future opinion. Journal of Broadcasting & Electronic Media, 47(2), 229–244. Petersen, T. (2007). Helmut Kohl’s Wahlkämpfe [Helmut Kohl’s campaigns]. In N. Jackob (Ed.), Wahlkämpfe in Deutschland — Fallstudien zur Wahlkampfkommunikation 1912–2005 [Election campaigns in Germany — Case studies of campaign communication] (pp. 194–214). Wiesbaden, Germany: VS Verlag. Reichenbach, H. (1938). Experience and prediction. An analysis of the foundation and structure of knowledge. Chicago, IL: University of Chicago Press. Salmon, C. T., & Glynn, C. J. (2009). The spiral of silence: Communication and public opinion as social control. In D. Stacks & M. Salwen (Eds.), An integrated approach to communication theory and research, 2nd ed. (pp. 153‒168). Hillsdale, NJ: Erlbaum. Salmon, C. T., & Kline, F. G. (1985). The spiral of silence ten years later. An examination and evaluation. In K. R. Sanders, L. L. Kaid, & D. Nimmo (Eds.), Political communication yearbook 1984 (pp. 3–29). Carbondale: Southern Illinois University Press. Salmon, C. T., & Moh, C. Y. (1992). The spiral of silence: Linking individual and society through communication. In D. Kennamer (Ed.), Public opinion, the press, and public policy (pp. 145–162). Westport, CT: Praeger. Shanahan, J., Scheufele, D. A., Yang, F., & Hizi, S. (2004). Cultivation and spiral of silence effects: The case of smoking. Mass Communication and Society, 4, 413–428. Simpson, C. (1996). Elisabeth Noelle-Neumann’s “spiral of silence” and the historical context of communication theory. Journal of Communication, 46, 149–173. Simpson, C. (1997). Elisabeth Noelle-Neumann background documents 1935–45. Retrieved from http://www1.american.edu/radiowave/noelle/noelle.htm Streeck, W. (2008). Von der gesteuerten Demokratie zum selbststeuernden Kapitalismus: Die Sozialwissenschaften in der Liberalisierung [From steered democracy to self-steering capitalism: The social sciences in the process of liberalization]. Köln, Germany: Max-Planck-Institut für Gesellschaftsforschung. Taylor, G. D. (1982). Pluralistic ignorance and the spiral of silence: A formal analysis. Public Opinion Quarterly, 46(3), 311–335. Zipp, J. F., & Fenwick, R. (2006). IS the academy a liberal hegemony? The political orientations and educational values of professors. Public Opinion Quarterly, 70(3), 304–326.

2 PUBLIC OPINION AND SOCIAL CONTROL Elisabeth Noelle-Neumann

Foreword by the Editors This text represents the edited and slightly shortened English version of Elisabeth NoelleNeumann’s inaugural lecture at the Johannes Gutenberg University Mainz on November 9, 1965. The translation has been supplied by John Fosberry. In the German academic system, newly appointed full professors are invited to give an inaugural lecture to a wider audience of the university presenting their main research and teaching agenda. NoelleNeumann was appointed for the newly founded chair of communication (in German “Publizistik”) the previous year after having taught for four years as an adjunct professor at the Free University Berlin. The text “Public Opinion and Social Control” is her first comprehensive formulation of ideas that later found their way into her socio-psychological theory of public opinion. So far it has been only published in German (Noelle, 1966). The text is published here with the permission of Noelle-Neumann’s heir Dr. Ralph Erich Schmidt. Omissions from the original text are indicated by: … Introduction Public Opinion—in some mysterious way this term retained its aura of suspense. Yet it is the fate of literary or learned treatises which venture to ventilate the concept of public opinion that they disappoint the reader or listener. When they demonstrate that “public opinion” is nonexistent, that it is a fiction, they are unconvincing. “The term cannot be killed,” complains Dovifat (1962, p. 108). Jürgen Habermas (1962, p. 13) wrote: “… not only does colloquial usage cling to it; even the sciences and particularly jurisprudence, politics and sociology are patently incapable of replacing traditional categories such as … ‘public opinion’ with more precise classifications.” And when, conversely, a defi nition

20 Elisabeth Noelle-Neumann

is attempted, the outcome is no better. After a two-day discussion and the reading of four papers on public opinion at a symposium of German scholars of communication in 1961, one of the participants said, “I still don’t know what public opinion is” (referenced in Löffler, 1962, p. 76). What is the significance of this obstinacy with which people cling to the term and of the feeling of disappointment after having discussed defi nitions? It means that the term “public opinion” stands for reality and that the defi nitions do not yet fully describe that reality. Thus, the situation is still the one as described by Oncken as early as 1904: “The mutable and indeterminate cannot be better understood by encaging it in a formula … In the fi nal analysis everybody, when asked, will know exactly what Public Opinion means” (Oncken, 1914, p. 236). But we cannot let the matter rest there, if only for constitutional and political reasons. As Landshut wrote, the constitutional institutions of social mass democracy count on an intact public opinion because the latter is still the only accepted basis for the legitimation of political power: “As the principal of its own verity, the modern state presupposes sovereignty of the people, and the sovereignty, in its turn, is assumed to be identical with public opinion. Without this imputation, without the substitution of public opinion as the origin of all authority for decisions that are binding on the population as a whole, modern democracy lacks the substance of its own verity” (Landshut, 1953, p. 586). And for scientific reasons, too, this question cannot simply be dropped. In this respect I need only refer to my own field of specialization, i.e., communication research: How can we examine what influence is exerted by the mass media on public opinion and, conversely, how public opinion fi nds expression in journalism, if we have diff use conceptions of what public opinion is?

… The assuredness most people have that something like public opinion really exists, that it is not a fiction, becomes evident when public opinion surveys are employed to observe certain socio-psychological processes, namely the processes of changes in the population´s opinion. Let me quote an example from the regular opinion surveys made by the Institut für Demoskopie Allensbach. A few weeks after Ludwig Erhard had become chancellor of the Federal Republic of Germany with an overwhelming popular approval, the organization of war victims and surviving dependents demanded an increase of their pensions. Erhard, determined to adhere to the principles of his government program, i.e., limitation of the budget despite demands by pressure groups, refused. Thereupon the number of his opponents grew within a few weeks by 5% or two million. It was not a specific group that rallied against him—say, the old people or the poorer families—but in each sub-group of age or region the proportion of his opponents had increased. Such changes in the frequency distribution of

Public Opinion and Social Control

21

opinions are sensed by the politician and by many people with an unconscious statistical perception even without surveys. These real experiences—illustrated at this juncture by just one example—are in my view the actual cause why the term “public opinion,” once it had been coined, was never given up. But now a host of questions arise which make the term “public opinion” almost intolerable. We will divide these questions into three categories. 1. Do such unarticulated moods of the population really deserve the name “public opinion”? Must we not have quite different demands on the power of judgment of those who bring forth public opinion—that good which is of “constitutional rank”? Who are the legitimate bearers of public opinion? 2. How is the term “public” to be interpreted in this context? “Non-public opinion”—to adopt the words of Habermas (1962, p. 112)—could be a quite apt description of such displeasure with Erhard as was shown by a large segment of the population in the example above. 3. How should a government react to this sort of public opinion? Should it give way to it? Or ignore it? What influence should public opinion have in a representative democracy on government action? The last question is essentially a query as to what functions public opinion has in a democracy, and it is this aspect of the public opinion phenomenon which will be dealt with here. We will attempt to define public opinion by its function—by what it does, by its effects—rather than by endeavoring to describe what it is. We will then also obtain some indications of its nature. Public Opinion Becomes Social Control If the existence of public opinion makes itself felt so perceptibly as was assorted earlier on, it is rather difficult to understand why scientific debate did not fi nd an accurate defi nition of that reality to which the term “public opinion” is applied. My hypothesis is: Thinking about public opinion has suffered from the fact that the link between two effects of public opinion was disrupted—its effects on governments and its effects on the individual member of society. “Public opinion” had become more or less exclusively the term with which the relationship to governments, its political aspect, was studied while the function of public opinion for the individual was studied under another name: “social control.” In 1690, when John Locke concisely and impressively described the phenomenon which was later known as “social control,” the term “public opinion” had not yet been coined. He wrote about the “law of opinion” (Locke, 1690, p. 475), linking up two elements, i.e., the fleeting, somehow intangible element “opinion” and the majestically powerful elements which was later so effectively combined by Rousseau when he coined “public opinion.” Between 1896 and 1898, when Edward A. Ross published in the American Journal of

22

Elisabeth Noelle-Neumann

Sociology the papers in which he created the term “social control” in its current meaning, he probably did not know John Locke’s “Essay Concerning Human Understanding” (see Ross, 1901, 1941). But both doubtlessly describe the same process by which society imposes conformity on its members. John Locke: “Thus the measure of what is everywhere called and esteemed virtue and vice is the approbation or dislike, praise or blame, which by secret and tacit consent, established itself in the several societies … in the world: whereby several actions come to fi nd credit or disgrace amongst them, according the judgment, maxims, or fashion of the place” (1690, p. 477). And Locke further: “But no man escapes the punishment of their censure and dislike, who offends against the fashion and opinion of the company he keeps, and would recommend himself to. Nor is there one of ten thousand, who is stiff and insensible enough, to bear up under the constant dislike and condemnation of his own club. He must be of a strange and unusual constitution who can content himself to live in constant disgrace and disrepute with his own particular society. Solitude many men have sought, and been reconciled to: but nobody that has the least thought or sense of a man about him can live in society under the constant dislike and ill opinion of his familiars, and those he converses with. This is a burden too heavy for human sufferance …” (p. 479). And Edward A. Ross, 200 years later, without references to Locke, wrote: “The course, vital man may ignore the social stigma. The cultivated man may take refuge from the scorn of his neighbors in the opinion of other times and circles, but for the mass of men, the blame and the praise of their community are the very lords of life” (1941, p. 90). “It is not so much the dread of what an angry public may do that disarms the modern American, as it is shear inability to stand unmoved in the rush of total hostile comment, to endure a life perpetually at variance with the conscience and feeling of those about him” (p. 105). “Only the criminal or moral hero cares not how others may think of him” (p. 104). In Ross’ work these sentences are still to be found under the chapter heading “Public Opinion.” In later texts on “social control” the term “public opinion” is not used at all or only quite coincidentally. The connection had been broken. In fact, however, even though the term “public opinion” is no longer used, no source can tell us better than the theory of social control about the function public opinion has in relation to the individual member of society. Paul Landis (1956, p. 12) states simply: It “is concerned … with … orderliness….” Expressed in more sociological terms: It is a means of social integration, social stability. It constrains the individual to adapt to prevailing conceptions of orderly behavior, it enforces—to varying degrees in different societies—conformity. That is its function. It is by no means the only instrument of social control. But—as compared to law, a much more visible, institutionalized form of social control—Ross praises it as being “very flexible” and “cheap” (Ross, 1941, p. 94). Rudolph von Ihering, a German professor of law, emphasizes: “The

Public Opinion and Social Control

23

advantage of psychological compulsion lies in the unlimited bounds of its sphere of effect. Like the air it penetrates everywhere, into the innermost recesses of the home, and to the steps of the throne, to places where physical compulsion would fail completely…” (1883, p. 180).

… Ihering and another German law professor, Franz von Holtzendorff (1879), who wrote about the same set of questions in the eighties—more than a decade before Ross—probably pioneered the development of the modern concept of social control. Ross knew and quoted “Der Zweck im Recht” (The Purpose of Law), a paper published by Ihering in 1883, in which the latter wrote “Order in society therefore means interdependence of its members, observance of the rules of behavior imposed upon them by the plan of the whole” (p. 176). Social Sanctions The violator of social norms meets with the censure, reproach, and disapprobation of the public. That disapprobation “is the expression not of a mere, theoretical judgment as in the case of a wrong conclusion, a mathematical error, a poor work of art, but the conscious or unconscious practical reaction of the consensus to its violation, a defense of its own security” (Ihering, 1883, p. 325). These are modern sociological conceptions. At the same time, however, for Ihering the relationship to public opinion was still alive. He refers to it as the “taskmaster of the ethical” (p. 340). Social control must not be regarded only as compulsion, as a negative force. Religion, tradition, ethics and customs, social roles, even fashion are positive forces of social integration. But when these positive rules, these “unwritten laws” (Socrates) are violated by deviant behavior of individuals or groups, public opinion is mobilized. Since, as an instrument of social control, it tends to preserve the status quo, stability, and conformity, it would drive society to torpor, if there were no opposing forces, no “moral heroes” (as Ross puts it) or the “angry young men” of modern usage, no artists, inventors, scientists, reformers who accomplish changes by direct personal impact on their environment, by their works or by the formation of subcultures. The innovators and reformers, to whom artists belong by their very nature, cannot regard public opinion as anything but repulsive. Friedrich Schlegel (the German author and translator of Shakespeare, 1767–1845) describes public opinion as a monster: “It appeared swollen with poison, its transparent skin shimmering in every color, and the entrails writhing like worms” (Schlegel, 1799, p. 40). The function of public opinion as towards members of society is solely to limit their individual elbow-room.1 The language is clear enough when, for example, John Locke speaks of “private censure” (1690, p. 480), when reference

24 Elisabeth Noelle-Neumann

is made to the “tribunal” of public opinion ( Jeremy Bentham, see Bowring, 1843, pp. 41‒45), or works on social control contain extensive chapters on “sanctions.” The measures of social sanction are then subdivided into three categories: physical sanctions, economic sanctions and, most important of all, psychological sanctions (LaPiere, 1954), the beginning of which may perhaps consist in discontinuing daily greetings and, at the end of which, to use Edward Ross’ words, the “dead member drops from the social body” (1941, p. 92). Public Opinion and Government Censure, sanctions—many would doubt that concepts can apply to the relationship between public opinion and the governing body. What can be said of public opinion as a “correlate of government” (Schmidtchen, 1959, p. 255)? Scholars of social control have assessed the effect of public opinion on the individual member of society by observing reality. On the other side, attempts to elucidate the nature and function of public opinion within the state have been exclusively made via historical or philological analyses. Up to a few decades ago, the interplay between public opinion and political power was scarcely accessible for scientific, systematic observation, and empirical study. The methods of representative public opinion polls had not been invented. Even now, by far the greater part of this work of investigation and observation still lies ahead of us. This work—and this is the second hypothesis I would like to advance— should be based on the assumption that public opinion has the same function in relation to political and economic power as it has in relation to the individual member of society, i.e. the function of integration, although the modus operandi is, of course, quite different. Public opinion as a correlate to government prevents the gap between government and the governed from becoming too large. There are many ways in which a dangerous gap might emerge. A government may act contrary to strong prevailing attitudes on the part of the population. This can happen either knowingly and out of the political leaders’ conviction that they possess superior political farsightedness, or unknowingly because they have no knowledge of the intensity of certain sentiments and views of the population. Or, influential groups may give rise to critical opinions or demands to which the government does not give way… That gap can be closed by the government by responding to the demands or at least by tactical evasions, or by convincing the people sooner or later that its actions were wise. These efforts are made under pressure from public opinion. Public opinion appears to government as a force that has to be reckoned with. Now let us turn back to the decline in support for Chancellor Erhard when he refused to increase the war victims` pensions in the fall of 1963. We can now answer the question of why the reaction of public opinion was so strong at that time: because war victims were involved, because the population has a

Public Opinion and Social Control

25

deep-seated feeling of obligation to victims of the war and sees their claims in a different light than those of others. This can be illustrated, for instance, by a survey on social justice in 1962. The survey showed how sensitive people are to differences in income, and that it is generally demanded that the affluent should give, but not receive any fi nancial support whatever. A test question on allowances for children read: “The position at present is that an allowance is paid from the third child onwards, regardless of how much the father earns. Do you personally feel it is just or unjust that every family receives children’s allowances?” “Just” said 37%, “unjust” 54%, and 9% were undecided. But this pattern does not apply to the case of war victims. We asked: “Opinions differ as to whether all war victims should receive a pension. Some say that people who have good earnings or other resources ought not to receive a war victim’s pension. Others say that if a person is a victim of the war, he should receive a pension even if he is in an economically sound position. What do you think?” Eight in 10 Germans were in favor of “a pension in any case” while only 18% said that the affluent should not get any pension … Boundary Conditions for the Emergence of a Public Opinion Obviously, on the question of increasing war victims’ pensions, the sentiments of the population confl icted sharply with the decision of Erhard´s government. Such confl icts are not easy to foresee. As mentioned at the beginning (in describing the decline in support of Erhard), the existence of public opinion becomes apparent when the attitudes of the population change. Perhaps it is more correct to say: Public opinion exists solely in a state of change, excitement, disturbance, when there is danger of disintegration. Or, it emerges occasionally by a sudden strengthening of integration when there is a particularly high degree of unanimity of government and the governed, for instance, in the case of a national triumph. When opinion is in a state of rest, it is preferable to use the term “climate of opinion,” which was coined in the 17th century by Joseph Glanvill (1661).2 The nature of that climate, the latent sentiments, are not always easy for the politician to perceive. Should Erhard have given way when public opinion turned against him? What influence should public opinion in a representative democracy have on government action? This question was raised at the beginning. From the results of analysis obtained so far, it appears certain to me that it would be wrong for a head of a government simply to ignore such a trend and to regard such course as a forceful method of conducting government business. That would be tantamount to denying that public opinion has any function in a democracy. It seems right to me to give way, when the grounds for the resistance of public opinion have moral weight, as in the case under discussion, and when the other circumstances of an issue permit the government to yield without substantial political or economic damage… But let us consider another, hypothetical case.

26

Elisabeth Noelle-Neumann

Let us assume that the federal government of Germany resolves to set a new course in foreign policy, an anti-American course. There can be no doubt that public opinion would be up arms against it, for it has been demonstrated many times that the feeling of friendship for America is currently almost unshakeable and that even the inactivity of the United States in the face of the building of the Berlin wall on August 13, 1961, did not change this.3 It would be a misunderstanding to assume that public opinion—which, as demonstrated, makes rules for the individual—also sets out to make rules for the rulers. Society probably demands imaginative, forceful leadership which, when it steers away from what was formerly desired and expected, must convince the public and so restore integration. In fact, as Landis (1956) suggested, public opinion measures with two different yardsticks and applies social control less stringently to leaders. Landis quotes William Gladstone’s saying: “If a man steals a loaf of bread, we send him to prison, if he steals a railway, we send him to parliament” (p. 404). For this reason it was probably a mistake to assume that the Profumo scandal4 was to blame for the loss of votes suffered by the British Conservatives, which fi nally lead to the Labour Party’s victory in the general elections in autumn 1964. Henry Durant, director of the British Gallup Institute, provides cogent proof that the real reason was the marked rise in the unemployment rate, which in the severe winter of 1962/63 stood at 3.25% and could not be significantly reduced up to the elections.

… Landis considers this use of two different yardsticks a phenomenon which is hard to interpret. But he has probably hit upon the truth with his own supposition that, in the case of political leaders, society is concerned about more important things than exemplary conduct in the bourgeois sense. Towards a Definition of Public Opinion At this point I would like to attempt a defi nition of public opinion. Public opinion is a force which acts in two directions. It integrates government, on the one hand, and the individual, on the other, with society. It arises out of the latent climate of opinion and consolidates when disturbances occur, when cohesion is in jeopardy. This is not the only form in which it manifests itself, but the most important one.5 Robert Merton, to whom I described this conception of the functional effectiveness of public opinion, drew the outline of an hour-glass to symbolize his interpretation of this thought. The context in which the term “public opinion” is found particularly often—“under the pressure of public opinion,” “the demand of public opinion”—demonstrate that public opinion is not a still life but a genuine force. The various ways in which it acts upon the individual and the leaders differ in

Public Opinion and Social Control

27

a characteristic manner: the individual is threatened with social isolation, the leader is threatened with a loss of power. Many scholars who have dealt with the concept of public opinion have limited it entirely to the political sphere, to its relationship with government. Among the modern authors, for example, Hans Speier (1956), who defi nes public opinion as free and public communications by citizens to their government in matters that are of importance for the community, follows this line, as do Ernst Fraenkel (1962), Siegfried Landshut (1953), and Wilhelm Hennis (1957a). In fact, however, the double meaning of the term has never been lost. JeanJacques Rousseau , who was the fi rst to use the term in 1750, expressly described public opinion as an instrument of social control: “He who judges morals also judges honor, and he who judges honor takes (public) opinion as his measure” (Rousseau, 1782, p. 224; translated from French). Here Rousseau uses language very similar to that of John Locke when he describes his law of opinion and reputation: “The measure of what is … virtue and vice is … the judgment … of that place” (Locke, 1690, p. 477). “For Rousseau public opinion is precisely what modern sociology calls an instrument of conformity,” writes Hennis (1957b). On the other hand, Rousseau calls public opinion a correlate to government “volonté générale.” In mentioning Rousseau at this point, I would like to add that I have intentionally not gone more closely into the question of who fi rst used the term “public opinion.” The question is not of great importance. The phenomenon was described long before the term was coined. That is why now even Machiavelli (1496–1527; see Katz, Cartwright, Eldersveld, & McClung Lee, 1954) and Cromwell (1599–1649; see Dovifat, 1962) are quoted on public opinion, and why the views of the Greeks and the Romans on public opinion are described (Holtzendorff, 1879).6 At the end of the 18th century, public opinion reappears in the sense of an instrument of social control in the famous novel Les Liaisons Dangereuses by the French author Choderlos de Laclos. A woman of the world writes to a young lady, advising her not to consort with a man of bad reputation: “You think him capable of a change for the better, let us go further and assume that this miracle has really happened—would not public opinion still be against him, and should not that be enough to make you settle your relation with him accordingly?” 7 In the 19th century, public opinion as an instrument of social control even appears in the role of an operetta chorus leader. In Jacques Offenbach’s “Orpheus in the Underworld” (1858), public opinion introduces itself in the prologue in front of the curtain: “Who am I?—The chorus of elders/I replace most ably/I am public opinion/Symbolic only in appearance/I criticize severely.” And in the middle of the 20th century, we also fi nd it as an implement of social control in our (German) Penal Code, in which Articles 186 and 187 lay down that the allegation even on an unimportant fact is sufficient to constitute

28

Elisabeth Noelle-Neumann

slander or libel, provided that the allegation is calculated “to lower the person concerned in the judgment of public opinion.” After reading modern German writers on public opinion, the use of such a formulation in the text or a law strikes one as quite strange. Here are preserved the views of 19th-century German jurists, the socio-psychological perspicacity that was so praiseworthy in Ihering’s work on the purpose in the law (“Der Zweck im Recht”) and also distinguishes the inquiry into Wesen und Wert der Öffentlichen Meinung (the nature and value of public opinion) published by Franz von Holtzendorff in 1879. There he writes of public opinion: “Alienated from its most important function as a zealous court of morals, public opinion was occupied largely with questions of formal constitutional law in Germany” (p. 74) … “We shall attain a true conception of public opinion only if we start from the generally admitted fact—sometimes extolled, sometimes lamented—that is a force which makes itself felt in the life of the people and the state in relation to both private persons and the government itself ” (pp. 46–47). What thematic boundaries are set for public opinion? The answer can be derived from the defi nition. The domain of public opinion must exclude all objects, all issues on which no confl ict can arise either between government and the governed, or between society and the individual. If its boundaries are set in this manner, the ambit of public opinion is almost universal. And what justification is there for the attribute “public” in the concept “public opinion”? The answer lies in the function of public opinion. That function relates to matters of common concern, not private matters, it relates to the cohesion of state and society (see also Habermas, 1962, pp. 15–18). Many authors have interpreted the attribute “public” in “public opinion” as “public reasoning,” for example, as in a newspaper editorial, a radio or television commentary, or the discussions in the political parties or quasi-official circles of highly educated and politically interested people, thoughtful and with a feeling of responsibility for the common wealth. It is this particular element of intelligent, political discussion that invigorates democracy which, says Habermas (1962, pp. 110–111), was meant by the “rigorously exact term ‘public opinion’” as used in France as early as the end of the 18th century. Should such public discussion, in the light of the concept that I have just cited, no longer be included under the heading of public opinion? I maintain that there is absolutely no necessity to use the term “public opinion” for this “public reasoning.” The role played by unhindered, free political discussion in forming opinion in a democracy remains entirely unaffected by whether public debate itself is classified as “public opinion.” This is expressed in a ruling of the Federal Constitutional Court in Germany: The formation of public opinion is a process to which all vital forces contribute “according to their actual moment” (Rulings of the Federal Constitutional Court, 1958, p. 104). The press in particular contributes to that process in a variety of ways. Sometimes it calls public opinion into being and moulds it, sometimes it mirrors it, and then again it sometimes

Public Opinion and Social Control

29

criticizes it. The object of the study of journalism is to do away with the generalization that the press embodies public opinion and to sound out and study these various modes of operation. The concept of public opinion cannot be limited to certain groups of people—say, informed citizens or responsible, thinking people—whose opinions supposedly constitute public opinion, although this has been tried over and over again. As Löffler (1962, pp. 78–79) correctly has stated, it can be limited only in space and time. So we can speak of public opinion in Hamburg or of public opinion in the Federal Republic of Germany or of public opinion in France or perhaps even of a world public opinion, and we can date it: in autumn 1965, or in 1950, and so forth. Authors dealing with public opinion have been particularly concerned with the quality of public opinion: Is public opinion clairvoyant—a signpost to the future that gives the politician important advice—or is it foolish and shortsighted? It is evident from what has been said here that it is neither the one nor the other. Public opinion has no intrinsic quality. It is solely a force that is pushing for integration, a social process that ensures cohesion of the system as a whole. Whether public opinion follows a far-sighted politician, a short-sighted politician, or a demagogue depends on the powers of persuasion of those men and on the quality of journalism, and also on a factor closely linked with the latter, the general level of the peoples power of judgment in a given country. Finally, it is also dependent on whether a population lives in good (or at least tolerable) conditions or in a state of desperate crisis. The Task for Empirical Research Of course, the goal of empirical social science is to apprehend and express key concepts such as public opinion so that they are accessible to systematic observation, experimentation, and forecasting. The conception presented here satisfies these requirements, as already has been demonstrated by the political issues mentioned above. Leo Crespi (1965), who carries out surveys for the U.S. government, affirms that the outbreak of revolutionary unrest in South American countries could be predicted from the results of some survey questions. This is not unbelievable, if one proceeds from the fact that public opinion does not tolerate all too large a gap between the rulers and the ruled. Even in modern dictatorships, with all the means at their disposal, the rulers fear the opening of too wide a gap and do everything in their power to keep public opinion busy with functions of social control. After all that has been said, many questions remain. How is public opinion formed? Certainly not by a simple increase or decrease in the number of people who hold a certain opinion. The force, the pressure of public opinion evidently develops from people’s notions that many others also think in the same or some other way, that is, from certain notions as to the distribution of opinions in

30

Elisabeth Noelle-Neumann

their environment. The capability of a statistical perception of people probably triggers the dynamic process of public opinion.8 Of what elements is the climate of opinion compassed? How does the politician bring about agreement of public opinion with his aims, how does he achieve integration? How can public opinion, always initially a conservative force, be won over to new ideas? Can public opinion be corrupted, manipulated against common interests? In order to work on such questions, we need a clarified concept of public opinion. I have endeavored here to make my contribution to that clarification. Notes 1. Society also awards its members, but by other mechanisms, not so much by “popularity” and “esteem,” the poor and inconstant rewards that public opinion can bestow. 2. The term was revived by Alfred North Whitehead and since been used quite frequently in scientific and political life (Merton, 1964, p. 135, footnote 6; p. 470). 3. Comment by editors: The author presents data from the Allensbach Archives on the question if Germany should stay closely allied to the United States. In 1961 40% were in favor, in 1965—after the building of the Berlin wall—even 47%. Questions on sympathy for the American people and on the decision with which country Germany should cooperate most closely yielded the similar results. 4. Comment by editors: John Profumo then was the British Secretary of War. After an aff air with a callgirl, he had to resign in 1963. 5. Other forms which must be analyzed are marked positive reaction to the government, e.g., during a period of nation-wide festive mood, say, during the British Queen’s tour of Germany in Summer 1965, positive change, improvement of integration bring the force of public opinion into action. 6. “For the Greeks, public opinion, presented outside its usual sphere of operation through democracy, was the fleeting moods of the demos, the light side of which found expression in the comedies of Aristophanes, or the popularity approved views of the wise men, which were voiced by the chorus in tragedies” (p. 13). 7. Comment by editors: cited by the author from a German translation of the book. 8. That notions about opinions of others constitute a part of public opinion is explained by Hofstätter (1949, pp. 53–55).

References Bowring, J. (Ed.). (1843). The works of Jeremy Bentham (Vol. 9). Edinburgh: William Tait. Crespi, L. (1965, April). United States Information Agency research: Objectives and problems. Paper presented at the First International Conference on Social Science Research, Williamstown, MA. Dovifat, E. (1962). Zeitungslehre [ Journalism] (4th ed.). Berlin, Germany: de Gruyter. Fraenkel, E. (1962). Öffentliche Meinung und Internationale Politik [Public Opinion and International Politics]. Tübingen, Germany: J. C. B. Mohr. Glanvill, J. (1661). The vanity of dogmatizing: Or, Confidence in opinions. Manifested in a discourse of the shortness and uncertainty of our knowledge, and its causes; with some refl exions on peripateticism; and an apology for philosophy. London, England: E. C. for H. Eversden at the Grey-Hound in St. Pauls-Church-Yard. Habermas, J. (1962). Strukturwandel der Öffentlichkeit. Untersuchungen zu einer Kategorie der bürgerli-

Public Opinion and Social Control

31

chen Gesellschaft [Structural Changes of the Public. Studies on a Category of Society]. Neuwied, Germany: Hermann Leuchterhand. Hennis, W. (1957a). Meinungsforschung und Repräsentative Demokratie [Opinion Research and Representative Democracy]. Tübingen, Germany: J. C. B. Mohr. Hennis, W. (1957b). Der Begriff der Öffentlichen Meinung bei Rousseau [The Concept of Public Opinion in Rousseau´s Works]. Archiv für Rechts- und Sozialphilosophie, 43, 111–115. Hofstätter, P. R. (1949). Die Psychologie der öffentlichen Meinung [The Psychology of Public Opinion]. Vienna, Austria: Wilhelm Braumüller. Holtzendorff, F. von (1879). Wesen und Werth der Öffentlichen Meinung [Nature and Value of Public Opinion] (2nd ed.). Munich, Germany: M. Rieger’sche Universitäts-Buchhandlung (Gustav Himmer). Ihering, R. von (1883). Der Zweck im Recht (Vol. 2.). Leipzig, Germany: Breitkopf & Härtel. Katz, D., Cartwright, D., Eldersveld, S., & McClung Lee, A. (1954). Public opinion and propaganda. New York, NY: The Dryden Press. Landis, P. H. (1956). Social control. Social organization and disorganization in process (Rev. ed.). New York, NY: Lippincott Company. Landshut, S. (1953). Volkssouveränität und öffentliche Meinung [Sovereignity of the People and Public Opinion]. In D. S. Constantopoulos (Ed.), Gegenwartsprobleme des internationalen Rechtes und der Rechtsphilosophie : Festschrift für Rudolf Laun zu seinem 70. Geburtstag [Current problems of international law and philosophy of law: Festschrift for Rudolf Laun for his 70th birthday] (pp. 579–586). Hamburg, Germany: Girardet. LaPiere, R. T. (1954). A theory of social control. New York, NY: McGrawHill. Locke, J. (1690). An essay concerning human understanding. Collated and annotated, with prolegomena, biographical, critical, and historical, by Alexander Campbell Fraser. Oxford, England: Clarendon Press. Löffler (Ed.). (1962). Die Öffentliche Meinung: Publizistik als Medium und Faktor der öffentlichen Meinung [Public Opinion: Journalism as Medium and Factor of Public Opinion]. Publication series of the Deutsche Gesellschaft für Publizistik, 4. Munich, Germany: C. H. Beck. Merton, R. K. (1964). Social theory and social structure (9th ed.). Glencoe, IL: The Free Press. Noelle, E. (1966). Öffentliche Meinung und Soziale Kontrolle [Public Opinion and Social Control]. Tübingen, Germany: Mohr. Oncken, H. (1914). Politik, Geschichtsschreibung und öffentliche Meinung [Politics, Histography and Public Opinion]. In H. Oncken (Ed.), Historisch-politische Aufsätze und Reden [Historical-political Essays and Speeches] (Vol. 1, pp. 203–243). Munich, Germany: R. Oldenburg. Ross, E. A. (1901). Social control. A survey of the foundation of order. New York, NY: Macmillan. Ross, E. A. (1941). Recollections of a pioneer in sociology. Social Forces, 20(2), 32–35. Rousseau, J.-J. (1782). Collection compete des Oeuvres. Vol. 2: Du Contrat Social ou Principes du Droit Politique [The Social Contract or Principles of Political Law]. Geneva, Switzerland. Rulings of the Federal Constitutional Court (1958), Vol. 8, p. 104. Schlegel, F. (1799). Lucinde. Berlin, Germany: Heinrich Fröhlich. Schmidtchen, G. (1959). Die befragte Nation. Über den Einfluss der Meinungsforschung auf die Politik [The Interviewed Nation. On the influence of opinion research on politics]. Freiburg, Germany: Rombach. Speier, H. (1956). Historical development of public opinion. American Journal of Sociology, 55(4), 376–388.

This page intentionally left blank

PART I

Overview, Evidence, Methods

This page intentionally left blank

3 PARADIGM SHIFTS IN THE STUDY OF MEDIA EFFECTS Denis McQuail

This chapter deals very briefly with a very large topic. It seeks, in particular, to locate spiral of silence theory within the debates that continue to resurface, concerning the strength and diversity of media effects and ways of conceptualizing the processes at work. These debates have often involved a choice between a powerful and a minimal model of influence. The term paradigm is chosen because the choice has large implications for research as well as for an estimate of consequences for society. We are not only dealing with shifts over time, with the implication that older approaches are replaced by newer and improved ones. It is also a question of a widening of options with alternative lines of thinking being opened up that do not necessarily replace earlier ideas. This reflects three other relevant aspects of the case: that the term media effects covers a wide spectrum of phenomena, for which no single paradigm might be suitable; that the nature of effects and the process by which they are attained (or occur) is dependent on the nature of the media in question (their technology and organization); and that changing circumstances of societal and world environment play a part. The breadth of the terrain opened up is reflected in one concise defi nition of the term media effects: “The consequences or outcomes of the working of, or exposure to, mass media, whether or not intended. They can be sought at different levels of social analysis [for instance, society, group, individual]. There are many types of effect, but it is usual to distinguish at least between effects that are behavioral, attitudinal (or affective) and cognitive. Effects are distinct from ‘effectiveness,’ which relates to the efficiency of achieving a given communication effect” (McQuail, 2010, p. 556). In the following sketch of alternative paradigms, several of these variable features come into play.

36

Denis McQuail

The Starting Point: Mass Media and Mass Persuasion The earliest conception of media effects had its origins in widespread assumptions about the communicative power of the new “industrialized” forms of public communication that developed in succession during the latter 19th and early 20th centuries, especially the large circulation newspaper, then fi lm, and radio broadcasting. These media-technological developments were closely related to changes in society in the direction of centralization, mobility of labor, urbanization, and new forms of mass politics. The concept of “mass society” was coined (Mills, 1951) to capture the essence of emerging modern society, in which the mass media were a key element, both rooted in and also reinforcing the features just summarized. Accordingly, a process of mass communication was identified and with it a version of the predominant type of effect that could be expected from extensive exposure to the new media of the time. Key elements of the mass society were being acted on at an earlier date by promoters of media businesses, as well as publicists and propagandists for commercial or political purposes. The advantages of mass media were: their nearuniversal reach; their popularity and often authority; their apparent uniform and instant impact on so many. These assumptions, built into the practice of the media themselves, came to shape the fi rst paradigm, one of powerful effects. It was this that stimulated and framed early attempts at empirical verification by way of the relatively new methods of social science, from the 1920s onwards. The most salient features were: • The adoption of the perspective of the sender or source, thus primarily relating to intended effects; • A focus on mass effects, but by way of direct effects on very many (anonymous) individuals; • A view from above of the masses as generally vulnerable to impulse and suggestion; • An emphasis on persuasion and attitude change, but also on collective behavior (e.g., mass movements, consumption, voting); • A recognized risk of unintended undesirable effects. It was this paradigm that largely defi ned expectations about mass media until corrected in several fundamental respects in the light of research. A Revised Paradigm of Minimal Effects The issues of public concern initially guiding research related especially to: fears of unintended harm to the young or vulnerable from exposure to inappropriate content; concerns about the integrity of established democratic procedures; possibilities of applying the mass media to positive social purposes (in respect of social attitudes, citizenship, health, etc.). The harnessing of communication to Second World War aims of training and propaganda provided

Paradigm Shifts in the Study of Media Effects

37

another incentive to research. What gradually emerged in a consistent way from inquiries based on comparative and before-after designs of possible media influence was the conclusion that the new channels of influence were only one factor contributing to communication effects and that exposure to medium and message in itself has very low predictive value of any effect. Many studies came to the conclusion that outcomes were related to a set of variable conditions, especially relating to message, channel, form and content, to individual differences between receivers, and to the social context in which the communication event takes place. The observation of extensive audience selectivity in both choice and interpretation of content played an important part in this revision. Mass communication did not necessarily mean mass reception. The failure of the powerful media thesis to pass the test of research has several causes apart from lack of initial cogency and unrealistic expectations or even the factors mentioned above. It reflected a better appreciation of the social forces underlying communication. In particular, it was apparent that influences of a more or less enduring kind (attitudes, beliefs, loyalties, information) require support from known and personally trusted others if they are to occur (Katz & Lazarsfeld, 1955). Not least, early research showed that media reinforce and anchor opinion as much as they change it (Lazarsfeld, Berelson, & Gaudet, 1944). Change will not occur unless it accords with some need or predisposition on the part of the receiver of influence. These conditions cannot be easily created or manipulated, hence setting close limits to any media power of persuasion or other influence. By the late 1950s a counter-myth of media impotence or, at most, minimal effects, had become established, based on a wide range of (essentially U.S.) research as summarized in Joseph Klapper’s (1960) review of fi ndings. In fact, as subsequently pointed out, it had not been Klapper’s intention to reject the notion of media influence, but to put the hypothesis on a sounder basis (Chaffee & Hockheimer, 1982; Neuman & Guggenheim, 2011). Powerful Media Rediscovered The paradigm of limited media effects, as outlined, began to be challenged more or less as soon as it was established. This would be a normal event in any process of scientific theorizing and investigation. However, in this instance it should also be seen as a reaction to changing media conditions as well as to realworld events that seemed to change the nature of the problem under study— essentially reformulating the question of media power. On the one hand, the new medium of television, from the 1950s onwards, gained rapidly in reach and attraction over other media, and became the single most dominant form of media, not least in the United States, where most research was still initiated. Television, as typically organized on a national basis, embodied a number of the mass conditions that had not previously been fully realized. It delivered

38

Denis McQuail

a more or less uniform and consistent set of contents to very large audiences (near whole populations) on a basis of widespread social trust and popularity. The opportunities for audiences to be selective were limited and little resistance could be expected to the predominant direction of flow of meaning. A second external influence came from contemporary events, again affecting the United States in particular, where a post-war period of calm prosperity was interrupted by external threats and social unrest. The heating up of the Cold War, the contentious involvement in the Vietnam war, the rise of student radicalism, the civil rights movement, and rapid social change contributed to an increased role for television as the central messenger about events and the world and as the mirror that reflected a picture of society back to itself. Challenges to the established order were occurring during the 1960s and 70s across Europe (East and West) and in other parts of the world, and highly critical, neo-Marxist theories of media and society were becoming widely current. Conventional thinking about the media seemed both narrow and complacent. The issue of media effects was being redefi ned away from questions of short-term planned influence on individuals to larger matters of long-term, cumulative, collective effects (both unintended and intended), not just in the direction of change, but of stabilization and control on behalf of an embattled social order. The newly arising questions were of compelling interest, but they were not easy to answer by means of methods of research in the minimal effects tradition. Much reliance was placed on content analysis designed to expose more or less hidden ideological directions in both journalism and fiction. Largely missing was a broader appreciation of the sociological basis for the formation of public opinion and its social dynamics. A key contributor to the search for a way forward was Elisabeth Noelle-Neumann, who not only provided the slogan for the banner of paradigm change with the title of her article “Return to the Concept of Powerful Mass Media” but offered theory and methods for investigating the posited effects (Noelle-Neumann, 1973). This is not to attribute the general shift in ideas about media power primarily to Elisabeth Noelle-Neumann alone, but her thinking embodies the essence of the paradigm revision under way. Her most specific contribution came with the concept of spiral of silence, which describes the outcome of a situation in which a given population is exposed over time to a persistent and consistent media account of current events and of dominant opinion on critical and controversial issues (Noelle-Neumann, 1974). The primary motivation of a person will be to conform, at least outwardly, so as to avoid discomfort and dissonance (a fear of isolation), an expectation supported by relevant theory of social relations. Over time there is thus a spiraling of opinion change in favor of one set of views. This perspective depends on the assumption that there does tend to be a predominant direction from major media sources in a national society. The essential condition for dominant perspective to shape public perceptions is the

Paradigm Shifts in the Study of Media Effects

39

extent of agreement between, in this case, the main sources of news. At the same time, the dominant view acquires a richer and more varied set of arguments and the language for diff usion by word of mouth. There are many connections between the thinking outlined and a number of other key concepts and discoveries in communication research at around the same time that served to firmly establish the new paradigm of potential media power. Support for the thesis that media are more influential in turbulent times was provided by dependency theory (Ball-Rokeach & DeFleur, 1976) which posited that social system conditions of change, confl ict, or instability increase the dependence of members of society on information and guidance offered by the mass media. Also directly relevant to the new phase of media effect is the concept of agenda-setting formulated fi rst by McCombs and Shaw (1972). The basic proposition is that the order of importance attached to events and persons in the news (time, space, and prominence) has a causal influence on the perceived significance of objects of public attention and thus indirectly on public opinion. There is an evident connection with another influential version of the media effect process—media cultivation (Signiorelli & Morgan, 1980). Research by its main author, George Gerbner, revealed an extraordinary condition of pluralistic ignorance resulting from excessive exposure to a highly selective and misleading television portrayal of social reality. Also supportive of spiral of silence theory was the phenomenon known as a third-party effect fi rst conceptualized by W. P. Davison (1983). This refers to the fact that survey respondents are inclined to attribute to media the power to influence the behavior and opinions of others, while denying any personal effect. It reinforces the view that both the formation and the public expression of a personal view regularly occurs with reference to the probable opinion of others, as personally estimated, more often than not with the media as the main source. The theory also offers general support to the notion of a duality of opinion. Last, but not least, there is a consistency with the ever-popular idea that journalists tend to frame all news stories in respect of their central theme, issues at stake, the back story, and sometimes even evaluative direction where a consensus seems to exist (Entman, 1993). Frame and schema theory (Graber, 1984) supports the view that media consistently promote dominant meanings, even if without ideological or other intention. The connections between these various theories are striking and help to explain why the new orientation towards the power of media was both welcome and has been enduring. It was not, however, the end of the story. While generally robust, this new paradigm was still somewhat restricted to a certain class of events and situations. It was also often dependent on a view of society as dominated by reigning orthodoxies (whether of left or right), as well as by a concentrated media system. The conditions of a mass society thus tend to be assumed.

40

Denis McQuail

A Fourth Stage: Negotiated Media Influence A later approach to media effects emerged that can best be termed social constructivist (Gamson & Modigliani, 1989). In essence this involves a view of the media as having their most significant effects by constructing meanings negotiated with more or less active audiences. The media tend to offer a consistent, even a preferred view of social reality (one that purports to be reliable). This includes both the information provided and the appropriate way of interpreting it, forming value judgments and opinions and reacting to it. These are the ready-made meanings that the media systematically offer to their audiences. It is up to the audience member to decide whether or not to adopt the views offered, although they are often the only material available for forming an opinion on distant matters. The alternative sources could include influences from personal experience or from the social or cultural environment that might even be a basis for active resistance to influence. Thus there is no automatic or direct transfer of meaning but a negotiation between what is offered and what a receiver is inclined to accept. This view of the process is a departure from the all-powerful media paradigm and is also marked by a shift from quantitative and behaviorist methods towards qualitative, deeper, and ethnographic methods. In some respects it embraces ideas that once led to the thesis of minimal effects, especially in recognizing the power of an audience to resist influence. However, we are here concerned not just with resistance but with voluntary change. In general, it is a formulation of the effect process which accords well with a view of mass media as essentially mediating rather than serving as a fi rst cause. This four-phase account of the development of thinking about media effects is only one interpretation. In her overview of the field, Perse (2001) argues that this and similar accounts of the development of effect theory are an oversimplification and may be misleading, especially by not recognizing the differences between various research areas. For instance, researches on children and on political communication have different histories. She proposes to deal with key differences in terms of alternative models of effect of different kinds. The four models she names are those of: • • • •

direct effects; conditional effects (varying according to social and psychological factors); cumulative effects (gradual and long term); cognitive-transactional (with particular reference to schemata and framing).

In fact these models correspond quite closely to the four phases described above, while putting less emphasis on the dimension of time. This accords with the view that the complexity of influences in a modern society requires us to have a corresponding diversity of models and methods of investigation. Potter (2011) also emphasizes the diversity of type of effect, with particular reference

Paradigm Shifts in the Study of Media Effects

41

to the time scale, the relative magnitude and the question of change versus reinforcement. A more fundamental enquiry into the development of thinking about effects (Neuman & Guggenheim, 2011), based on an extensive and deep literature review and a factorial analysis of patterns identifies six stages of effect theory, differentiated by development over time, but also by other characteristics of type of effect and predominant condition or variable. The six stages are: persuasion studies; active audience theories; social context; societal and media theories;. interpretative effects theories; new media theories. These succeed each other in sequence between 1944 and 1996, reflecting accumulating understanding. This necessarily sketchy history of the field also indicates a continuous process of adaptation to changing technology and changing social structures and media systems. A New Stage of Media Effects? We have already reached a point in the quite fundamental reshaping of our means and systems of communication, both private and public, when some of the accumulated approaches to media effects are looking either outdated or inadequate. There are few clear guidelines as yet for predicting how effect theory and research will proceed. The new online media are characterized by great diversity, interactivity, and freedom of publication as well as by new forms of communication relationship. It is no longer possible, as in earlier theory, to treat interpersonal and public communication as two separate spheres for investigation. In societies that are more based on a network form, with innumerable nodes rather than a hierarchy of control, there are at the very least many new possibilities for different patterns of effect. Many of these have already been imagined and tried out, some already investigated. The general drift of early thinking about new media and effects was that abundance and diversity would shift the balance of power from senders and sources to receivers, open access to channels for a myriad of new voices and open the way for new kinds of effect, based on the interactive potential of online media. A new media future has thus often been seen as replacing a mass media past, a paradigm of openness and multiplicity. In making sense of this change, the old opposition of minimal to powerful paradigms has been revived, for instance by Bennett and Iyengar (2008), who see the once concentrated and rather uniform space of public communication in the United States as being too fragmented to sustain the essentials of powerful media theory. However, it is less clear that social structures and media systems have changed in any way as radically as the means of communication available and the dynamics of the old order are still powerfully at work, in the sense of a competitive geo-political frame and the empire-building ambitions of global media corporations with respect to new online media in particular. The emancipatory

42 Denis McQuail

potential of new media receives little encouragement in this environment. Their basic interactive properties plus the networking and seemingly autonomous operation of social media can all be harnessed to larger media campaigns (political or commercial), without reverting to the status of mass media. Thus mass communication as a process can survive the mutation of mass media, as originally characterized, into a wide spectrum of media possibilities for different purposes. Moreover, the overall pattern of Internet usage is beginning to exhibit familiar patterns of domination by a relatively small range of uses, dominant providers and owners, and even by mass audiences (e.g., 800 million subscribers to Facebook). The once fi rm belief in its immunity to control and central supervision is gradually fading. The implication of these remarks is that powerful media theories, such as the spiral of silence, are not at all obsolete in this new age of communication but much rethinking is needed. Nevertheless, despite the valid criticism of naïve histories of media effect and of technology-centric thinking, we should not lose sight of the fundamental dimension of communicative power (or influence). It is not easy to escape from the impression that there remains an underlying choice between two sets of (inter-related) paradigmatic choices: on the one hand, between extensive and minimal power and, on the other, between other-directed and self-chosen influence (control vs. autonomy). The overall expectation is that we may be returning to a phase of minimal effects, as a result of the break-up and diversification of media systems (Bennett & Iyengar, 2008). References Ball-Rokeach, S. J., & DeFleur, M. L. (1976). A dependency model of media effects. Communication Research, 3(1): 3–21. Bennett, W. L., & Iyengar, S. (2008). A new era of minimal effects? Changing foundations of political communication. Journal of Communication, 58, 707–731. Chaffee, S. H., & Hockheimer, J. L. (1982). The beginnings of political communication research in the US: Origins of the limited effects model. In E. M. Rogers & F. Balle (Eds.), The media revolution in Europe and America (pp. 263–284). Norwoood, NJ: Ablex. Davison, W. P. (1983). The third person effect. Public Opinion Quarterly, 47(1): 1–15. Entman, R. M. (1993). Framing bias: towards clarification of a fractured paradigm. Journal of Communication, 43, 51–58. Gamson, W., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power: a constructivist approach. American Journal of Sociology, 95, 1–37. Graber, D. A. (1984). Processing the news. New York, NY: Longman. Katz, E., & Lazarsfeld, P. F. (1955). Personal Influence. New York, NY: Free Press. Klapper, J. (1960). The effects of mass communication. New York, NY: Free Press. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people’s choice. New York, NY: Duell, Sloan and Pearce. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of the press. Public Opinion Quarterly, 36, 176–187. McQuail, D. (2010). McQuail’s theory of mass communication. London, England: Sage. Mills, C. W. (1951). White collar. New York, NY: Oxford University Press.

Paradigm Shifts in the Study of Media Effects

43

Neuman, W. R., & Guggenheim, L. (2011). The evolution of media effects theory: a six stage model of cumulative research. Communication Research, 21, 164–196. Noelle-Neumann, E. (1973). Return to the concept of powerful mass media. Studies of Broadcasting, 9, 66–112. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24, 24–51. Perse, E. M. (2001). Media effects on society. Mahwah, NJ: Erlbaum. Potter, W. J. (2011). Conceptualizing media effects. Journal of Communication, 61, 896–915. Signorielli, N., & Morgan, M. (Eds.). (1980). Cultivation analysis. Newbury Park, CA: Sage.

4 THREE CONTEXTS OF THE SPIRAL OF SILENCE THEORY Hans Mathias Kepplinger

This chapter examines the spiral of silence theory in terms of three interrelated contexts corresponding to professional and personal interests of Elisabeth Noelle-Neumann: a career-long fascination with survey research methodology, a belief in powerful media effects, and involvement in post-war German politics. As an exchange student at the University of Missouri-Columbia School of Journalism, Elisabeth had discovered the method of public opinion polls developed by George Gallup. Being fascinated by it, she decided to change her dissertation topic to Amerikanische Massenbefragungen über Politik und Presse [American Mass Surveys on Politics and the Press] (Noelle, 1940). She had no significant practical experience but based her knowledge on American public opinion research when she conducted her fi rst survey on behalf of the French military government. Its subject was the political attitudes of youth. A year later she and Erich Peter Neumann, whom she married in 1946, founded the Institut für Demoskopie Allensbach, which she regarded as her scientific laboratory. The Allensbach Institute’s projects were to be kept manageable and not be allowed to get too large. Friedrich Tennstädt was one of the fi rst employees and a major contributor to the development of the Allensbacher Werbeträger Analyse (AWA), measuring the use of newspapers, news magazines, and electronic media. The survey collected detailed data on the media usage of 20,000 people at a time. After the Social Democratic Party (SPD) did not accept the Institute’s offer to conduct public opinion polls on current issues in 1949, in 1950 the Institute entered into a contract with the conservative German government to research current public opinion (Noelle-Neumann 2006, p. 173). This contract has been renewed every year since. The German government reports were mostly written by her husband, Erich Peter Neumann, until his

Three Contexts of the Spiral of Silence Theory

45

death in 1973. He also founded the Jahrbuch der öffentlichen Meinung 1947‒1955 [Yearbook of Public Opinion 1947–1955], which has continued to appear at irregular intervals since 1956, albeit with the title Allensbacher Jahrbuch der Demoskopie [Allensbach Yearbook of Public Opinion] since 1976. Noelle-Neumann’s methodologies were above all survey design and procedures. Since the fi rst edition of the Jahrbuch der öffentlichen Meinung 1947‒1955 (Noelle & Neumann, 1956: XI–XLII), an article with the rather unassuming title “Erläuterungen” [Notes] has appeared in each issue providing an introduction on questionnaire construction based on the latest literature and with numerous in-house examples. 1962 and 1965 saw the appearance of her pioneering essays “Über den methodischen Fortschritt in der Umfrageforschung” [On methodological Advances in Public Opinion Research] and “Die Rolle des Experiments in der Publizistikwissenschaft” [The Role of Experiments in Mass Communications Research] (Noelle-Neumann, 1977, pp. 25–42, 43–61). This was followed by experiments on the design of questions published in 1970 in the United States (Noelle-Neumann, 1970) as well as her article on “Empfi ndlichkeit demoskopischer Messinstrumente. Frageformulierung und Fragebogenauf bau” [Sensivity of Instruments in Public Opinion Research: Formulating Questions and Designing Questionnaires] (Noelle-Neumann, 1976a, pp. VII–XXV). Her concepts for the choice of study design, question formulation, and selecting categories were dominated by one paramount goal: fi nding the simplest and most elegant way to increase what can be revealed in studies. All other factors such as the time available, funding, employee skills, and so forth, were of little importance for her. If necessary, studies might have taken longer, funding would have to be raised, and staff hired, and if the project was no longer feasible, then it had to wait until another opportunity presented itself. Sometimes it could take months or years. Noelle-Neumann’s defi nition of quality was, above all else, validity of her surveys. Does the question capture the required information? Does the category communicate what it is supposed to say? It was her conviction that this was only the case if the questions were written in the language and thinking style of the respondents. This required intensive struggling with the core of the problem, with numerous attempts to formulate and fi ne tune the best proposals, run them through trials, fi ne tune them again, and continue to test then until the end product was ready. Noelle-Neumann’s drive to shape valid survey questions from complex research queries had major effects on the organization of field work at the Allensbach Institute. Noelle-Neumann was deeply convinced that opinions on socially significant problems could only be validly researched with a broad variety of visual scaffolding. She would devise cartoons to ask dialogue questions with explanations for opposing views, and differentiation of shades of opinion was clarified with statements written on cards that would be sorted repeatedly. Likewise, respondents were helped to refi ne their viewpoints and make

46 Hans Mathias Kepplinger

them more precise with single or multiple choices. Noelle-Neumann always mistrusted the standard statistical reconstructions of complex attitudes and opinions based on agreeing with simple statements. Their results were, in her opinion, too faint, too unintuitive, and simply unrealistic. In general, she preferred polarizing test questions to establish opposing categories of respondents who can be compared to some artificial indices which might allow localizing respondents on a continuum using statistical techniques. Because the complex types of questions needed to make discrete distinctions among opinions were only possible in face-to-face interviews, she always insisted on them despite the high cost and time required. She declined to follow the popular trend of telephone surveys, and it was only in her fi nal years that she consented to using them for gathering data on simple issues. Noelle-Neumann went against the general trend and practice of most polling organizations on a second question of methodology: she was a devotee of quota samples from the outset of her career until her death (Noelle-Neumann & Petersen, 2005, pp. 236–281). She gave three main arguments and relevant empirical evidence for her position. First, polls conducted by random sampling often do not meet the statistical requirements for a random sample due to low response rates. Second, good (i.e., sufficiently complex) quotas lead to a better random selection than random sampling. Third, forecasts are more accurate on the basis of quota requirements than on the basis of random sampling. NoelleNeumann, however, was not a naive devotee of face-to-face interviews and quota samples. Instead, she had continually and systematically researched the undesirable influence of the interviewer on the selection of interviewees and crafted the results into a catalog of rules to be followed. Among these rules was the requirement that interviewers should be distributed throughout the country in such a way that they could carry out the interviews in their hometowns and that each interviewer would do only a few interviews. She considered testing different versions of questions in split-ballots an essential basis for public opinion research to be considered scientific. Thus, over the decades, the Allensbach Institute tested more than 6,000 questions in alternative versions, of which approximately half related to the formal design of the questionnaires, including the effects of sequence and context of the questions (Petersen, 2002, p. 101). Panel surveys were for Noelle-Neumann the pinnacle of survey research. Trend surveys and analyses on aggregate levels were good and necessary and formed the major focus of her Institute’s work. But, whenever it was possible, she would ask the same people several times in order to investigate the changes in their individual opinions, attitudes, and behaviors. Although the Institute conducted most of its panel surveys to analyze consumer behavior, Noelle-Neumann did repeatedly use them also in political polling, subsequently triggering intense public debate (Noelle-Neumann, 1980, pp. 77–115). The second context for developing the spiral of silence theory is NoelleNeumann’s notions of strong media effects. Noelle-Neumann’s inaugural

Three Contexts of the Spiral of Silence Theory

47

lecture in December 1965 on “Öffentliche Meinung und Soziale Kontrolle” [Public Opinion and Social Control] laid out her thematic and theoretical program which she had pursued single-mindedly for many years (Noelle, 1966). Like her colleague Paul F. Lazarsfeld, Noelle-Neumann considered public opinion leaders to be key players in shaping societal opinion and will. After years of preparation, in 1985 she presented the scale she and her staff had been developing to identify the key opinion leaders (Noelle-Neumann, 2002, pp. 94–130). This scale brought together two major strands of opinion leader research—one sociometric and the other based on personality traits (Weimann, 1991). Noelle-Neumann’s most important contribution to the development of public opinion research methods was her systematic textbook Umfragen in der Massengesellschaft [Polls in Mass Society], published in 1963 with an initial print run of 20,000 copies. It has since been reprinted several times and has been translated into many languages. A significantly revised (with Thomas Petersen) edition appeared in 1996 under the title of Alle, nicht jeder [Everybody, Not Everyone] now in its fourth edition (Noelle-Neumann & Petersen, 2005). In contrast to Paul F. Lazarsfeld, for Noelle-Neumann the important role of opinion leaders in the diffusion of information and opinions does not minimize but maximize the effect of mass media. According to her, most journalists in the dominant media hold liberal attitudes and opinions which more or less influence their coverage; their coverage reinforces and partly changes the views of opinion leaders which hold similar views as the journalists. Both belong to a small segment of the population. Opinion leaders transmit the information and opinions from the media more or less unchanged to their interlocutors. Therefore, much of the effect exerted by opinion leaders has to be attributed to the mass media which feed them with topics, frames, and information. The indirect influence of opinion leaders extends the direct effect of media coverage on readers, listeners, and viewers and slowly turns the views of a minority into the views of the majority. Noelle-Neumann did not only recognize the indirect effect of mass media as outlined above but presented a theory of direct effects especially of television on the population in general. Her major argument was that people do not use television as selectively as newspapers, which were the main focus of interest of Lazarsfeld and his colleagues in the 1940s. Therefore, in the long run, the majority of TV viewers get confronted with large amounts of information and opinions of journalists and public figures which contradict their own attitudes and opinions. This steady confrontation with divergent views changes their own views and underlying value system. This process can be enforced by the pressure of public opinion—perceived majority opinions presented in media coverage and opinions held by colleagues, friends, etc., if the issues covered are related to important values. In 1972, her presentation “Return to the Concept of Powerful Mass Media” at the 20th International Congress of Psychologists in Tokyo led largely to the

48 Hans Mathias Kepplinger

paradigm shift in media effects research which continues to resonate today (Noelle-Neumann 1973a). She published the text in Germany under the title “Kumulation, Konsonanz und Öffentlichkeitseffekt” [Cumulation, Consonance, and Effect of Publicity] (Noelle-Neumann 1973b) which mentions the essential elements of her new approach in three key words: the impact of frequent reports over time, the impact of similar representations across several media outlets, and the influence of the media climate on individual opinion formation. Two years after her speech in Tokyo, she published her ideas under the title “The Spiral of Silence: A Theory of Public Opinion” (Noelle-Neumann, 1974a). In 1980, her Theorie der Schweigespirale [Theory of the Spiral of Silence] appeared as a separate book. An extended version was published in the English language under the title “The Spiral of Silence. Public Opinion – Our Social Skin” (Noelle-Neumann, 1993). Except for the coverage on extraordinary events, the effects of media coverage do not occur immediately but rather within weeks, month, or even years. Therefore, Noelle-Neumann was highly interested in the combination of quantitative content analysis of media coverage on social issues like death penalty, nuclear energy, images of politicians, for example, over long periods of time with related results from opinion polls. Here, her concept of the “dual climate of opinion” comes into play: if the interviewees report that the dominant media coverage contradicts the dominant views of their friends, colleges, and family, in the long run the observed majority will not change media coverage but media coverage will change the observed majority. As soon as the gap between the two climates of opinion has closed, the impact of media coverage—supplemented by the impact of interlocutors—is at its peak. The third context for the development of the spiral of silence theory dealt with her interest in politics, her essays on public issues, and her related experiences. For Noelle-Neumann, who had studied historiography to become a journalist, poll research was not only an instrument to determine the distribution of current opinions. She found polls to be equally important as a chronicle of contemporary history and a historiographical source. In this, she viewed public opinion research as a corrective to the tenor of the media, who, she was convinced, reflected the views of a small section of the social elite more than the opinions of the masses. An important base for her historiographical studies and essays on current issues like the change of social norms and behavior in Germany (Noelle-Neumann, 1978a, 1978b), the mentality of German employees (Noelle-Neumann, 1984), and the split and unification of Germany (Noelle-Neumann, 1991) was the archive of the Institute. The archive includes documentation of all the questions asked by the Institute throughout its long history. Each question record revealed whether it was asked only once or several times, in the same way or modified, as a statistical trend or in panels, and the broader survey to which it belonged. The archive contains approximately 300,000 questions now stored electronically.

Three Contexts of the Spiral of Silence Theory

49

In the 1950s and 60s, Noelle-Neumann became a media celebrity in Germany. The major news magazine Der Spiegel had published two cover stories about her in 1953 and 1957 and thus gave her publicity normally reserved for major politicians. Noelle-Neumann’s television appearances on election nights were national media events anticipated by millions of voters. Her surveys on the current state of the country were not only something new at the time, but also held a mirror before a population scarred by years of self-alienation. The intensive and positive media coverage reached its zenith in the 1960s and then led to extremely polarized opinions about Noelle-Neumann. A turning point was her 1969 lecture at a conference sponsored by the second German TV channel (ZDF) entitled “Der getarnte Elefant. Über die Wirkung des Fernsehens” [The Camouflaged Elephant: About the Effects of Television]. This lecture outlined why the thesis of the 1940s and 50s that the mass media have minimal effects was no longer valid (Noelle-Neumann, 1979, pp. 115– 126). This had the result of pitting Noelle-Neumann against the then-leading academic theory and breaking a taboo of the media against using media to influence public opinion. This criticism took on a political quality during the student movement, which saw Noelle-Neumann as an advocate of the West German political-economic system and a representative of the establishment. In the winter semester 1970/71, some of her opponents occupied the Institute. Noelle-Neumann´s lectures were interrupted by shouts and whistles often so loud that she could barely be heard. In addition, she came under attack by some journalists and academics for her activities during the Third Reich, which had not previously played a role. The criticism originally published in some media outlets and articulated by a small group of opponents among her students grew into a storm of indignation as she documented in the 1976 general election how frequent viewers of political TV broadcasts saw the left leaning SPD as winners and the right leaning CDU/CSU as losers, even though their stated voting intentions showed a headto-head race. On the other hand, those who saw such programs rarely, if ever, kept their own ideas of winners and losers. Noelle-Neumann considered this as a testament to the impact of television programs edited by left leaning journalists, published her fi ndings before the election, and sparked a massive academic and political discussion (Noelle-Neumann, 1980, pp. 77–115). In the 1980s, Noelle-Neumann published numerous analyses of the self-image of the West German population in the time before the fall of the Berlin Wall. A striking example of this was her 1984 essay in the Frankfurter Allgemeine Zeitung entitled “Im Wartesaal der Geschichte” [In the Waiting Room of History] in which, based on trend questions, she revealed the desire of Germans for reunification. These and a number of similar studies before and after reunification in 1990 (Noelle-Neumann, 1991) were criticized by intellectuals who viewed the division of Germany as a just punishment for the crimes of the Nazis and wanted to retain it. One of the admired stars of Germany´s public life had slowly developed

50

Hans Mathias Kepplinger

into one of the most controversial, if not an isolated researcher and essayist. It should be noted that the stepwise change of her role beginning in the late 1960s and cumulating in the mid-1970s was closely linked to the development of the spiral of silence. This is not to say that Noelle-Neumann derived her theory of public opinion from personal experiences—she has been fascinated by this topic long before her public image changed. But her personal observations reinforced her theoretical insights and strengthened her motivation to analyze the process of opinion change empirically. Integrating the Contexts The theory of the spiral of silence is probably the only theory of a German social scientist to have attracted world-wide attention since 1945, and it has since influenced the ideas of many lay people about the processes of opinion formation. Nevertheless, Noelle-Neumann was not an academic theorist in the usual sense. She was much more an explorer, always looking for new and surprising results. At every opportunity—in seminars and conferences, on museum visits, and even at dinner parties—she would have a large stack of notes with her, where she would write down important observations and statements with sweeping strokes and preserve them carefully. There must be more than 100,000 of these notes extant, which form an inexhaustible source for the reconstruction of her life. Public opinion research was for Noelle-Neumann not an evidentiary process, but rather a process of discovery. This is also seen in the development of her theory on the spiral of silence. Public opinion had been her life’s work since her student days in the United States. She also repeatedly pointed to her early, formative ideas on the so-called band-wagon effect. But it did not serve as the initial impetus for the theory of the spiral of silence. Rather, her theory came about by combining two sets of disparate fi ndings gradually until she suddenly saw the solution. One set involved the greater visibility of SPD supporters in the national election of 1972 and the damage infl icted by CDU/CSU election posters as found out in surveys conducted at the time. This inferred a greater commitment and aggressiveness of SPD supporters, coupled with intimidation by supporters of the opposition. The other set, fi rst identified in 1970, concerned the overrepresentation of SPD voters and under-representation of CDU/CSU voters in post-election polling about the votes they had cast in the previous election. This was interpreted as evidence that the sample was biased and was compensated by reweighting the data (Noelle-Neumann, 1974b). The key to the development of the spiral of silence theory rested in the insight that both issues were related: the CDU/CSU supporters were not underrepresented in the polls since 1970. Some of them, however, had denied their vote for the CDU/CSU in the previous election when asked, because they, as the second set of data

Three Contexts of the Spiral of Silence Theory

51

showed, felt under intense social pressure. Her own experiences as public figure validated her empirical fi ndings and theoretical insights and motivated her to conduct a large number of studies to learn to know the causes and conditions of opinion changes in society. Noelle-Neumann fi rst explicated the conditions for her theory only under the influence of others’ research which had not confi rmed her hypotheses, because the circumstances had not complied with her conditions. The fi rst version of the spiral of silence theory made the point that opinion formation is influenced only in certain situations. For example, the opinions surveyed must be about emotionally charged topics which could lead to social isolation. But in the early publications, this was treated more as an aside, because one could infer it from the test questions and the answers. It seemed to be trivial. Therefore, for the fi rst few years, she had not independently tested the conditions of her theory. As a consequence, one could not decide whether the conditions for its application had not been given or whether the theory had not been confi rmed. Later, she developed her “threat test” which eliminated this problem. Noelle-Neumann has since tested her theory on the basis of innumerable subjects, introduced new test questions, and opened up further fields of application. She never subjected them to falsification attempts because she considered Popper’s epistemology to irrelevant in practice. This concerned not only its deductive principle, but also its falsification postulate. The former contradicted her belief in the importance of individual data as a source of discovery, the latter she considered an unrealistic imposition. Her frequent multi-level contrast group analyses were not designed to falsify her theory. She wanted to see if her assumptions derived from her theory would still apply if she introduced new control variables. As much as she was uninterested in a falsification of her theory, she likewise had no interest in drawing and testing some obvious conclusions. These include the hypothesis that the spiral of silence turns faster if, instead of a dual climate of opinion, the tenor of the media agrees with the majority opinion. Even that she would have considered trivial, because in her view, the overlap of the majority opinion with the opinion of journalists was not a problem requiring explanation. Such practical problems of a society were of much greater importance for her than purely theoretical discussions. Therefore, it can be stated that Noelle-Neumann’s theory of the spiral of silence did not develop out of an interest in academic theories, but out of a concern for methodological and social problems, nor it did not result from deductive derivations, but rather intuitive insights, based on a variety of life experiences. That might be one of the reasons that the spiral of silence theory still receives world-wide attention. At the same time, its success is proof that textbooks teach useful basic skills such as the fundamentals of epistemology and methodology, but creative research often follows other principles.

52

Hans Mathias Kepplinger

References Noelle, E. (1940). Amerikanische Massenbefragungen über Politik und Presse [American Mass Surveys on Politics and the Press]. Frankfurt, Germany: Verlag Moritz Diesterweg. Noelle, E. (1966). Öffentliche Meinung und Soziale Kontrolle. [Public opinion and social control]. Tübingen, Germany: J. C. B. Mohr. Noelle-Neumann, E. (1970). Wanted: Rules for wording structured questionnaires. Public Opinion Quarterly, 34, 191–201. Noelle-Neumann, E. (1973a). Return to the concept of powerful mass media. Studies of Broadcasting, 9, 67–112. Noelle-Neumann. E. (1973b). Kumulation, Konsonanz und Öffentlichkeitseffekt [Cumulation, Consonance, and Effect of Publicity]. Publizistik, 18, 26–55. Noelle-Neumann, E. (1974a). The spiral of silence: A theory of public opinion. Journal of Communication, 24, 43–51. Noelle-Neumann, E. (1974b). Wahlentscheidung in der Fernsehdemokratie. Eine sozialpsychologische Interpretation der Bundestagswahl [Voting behavior under the influence of TV. A social-psychological interpretation in the general election]. In D. Just & L. Romain (Eds.), Auf der Suche nach dem mündigen Wähler. Die Wahlentscheidung 1972 und ihre Konsequenzen [In search of the mature voter. The outcome of the 1972 election and its consequences] (pp. 161–205). Bonn, Germany: Schriftenreihe der Bundeszentrale für politische Bildung. Noelle-Neumann, E. (1976a). Die Empfi ndlichkeit demoskopischer Messinstrumente. Frageformulierung und Fragebogenauf bau [Sensibility of Instruments in Public Opinion Research. Formulating Questions and Designing Questionnaire]. In E. Noelle-Neumann (Ed.), Allensbacher Jahrbuch der Demoskopie 1974–1976 [Allensbach Yearbook of Survey Reseach 1974– 1976 ] (pp. VII–XXVI). Wien, Austria: Fritz Molden, S. Noelle-Neumann, E. (1976b).Umfragen in der Massengesellschaft. Einführung in die Methoden der Demoskopie [Polls in Mass Society. Introduction into the methods of opinion polls], (7th ed.). Reinbek bei Hamburg, Germany: Rowohlt. Noelle-Neumann, E. (1978). Werden wir alle Proletarier? Wertewandel in unserer Gesellschaft [Will we all become proletarian?]. Zürich, Switzerland: Edition Interfrom. Noelle-Neumann, E. (1978a). 1967 bis 1992. Die bürgerlichen Werte sinken ab. In E. NoelleNeumann, Werden wir alle Proletarier? Wertewandel in unserer Gesellschaft (pp. 10–43). Zürich: Edition Interfrom, S. Noelle-Neumann, E. (1978b). Verfall der Arbeitsfreude. In E. Noelle-Neumann, Werden wir alle Proletarier? Wertewandel in unserer Gesellschaft (pp. 59–72). Zürich: Edition Interfrom, S. Noelle-Neumann, E. (1979). Öffentlichkeit als Bedrohung. Beiträge zur empirischen Kommunikationsforschung [Negative publicity as threat. Contributions to emprical communications research]. Jürgen Wilke (Ed.). Freiburg, Germany: Karl Alber. Noelle-Neumann, E. (1980). Wahlentscheidung in der Fernsehdemokratie [Voting behavior under the influence of TV]. Freiburg, Germany: Ploetz. Noelle-Neumann, E. (1984). Menschen aller sozialer Schichten brauchen und fi nden Freiraum für persönliche Entscheidungen [People of all social strata need and will fi nd freedom to make indicidual decisions]. In E. Noelle-Neumann & B. Strümpel (Eds.), Macht Arbeit krank? Macht Arbeit glücklich? Eine aktuelle Kontroverse [Does woking make ill? Does it make happy? A topical controversy] (pp. 66–96). München, Germany: Piper. Noelle-Neumann, E. (1991). Demoskopische Geschichtsstunde. Vom Wartesaal der Geschichte zur Deutschen Einheit [A lesson in history based on opinion polls. About the waiting room of the history of Germen unification]. Zürich, Switzerland: Edition Interfrom. Noelle-Neumann, E. (1993). The spiral of silence. Public opinion — Our social skin (2nd ed.). Chicago, IL: The University of Chicago Press. Noelle-Neumann, E. (2002). Die soziale Natur des Menschen. Beiträge zur empirischen Kommunikationsforschung [The social nature of men. Contributions to empirical communications research]. Freiburg, Germany: Alber.

Three Contexts of the Spiral of Silence Theory

53

Noelle-Neumann, E. (2006). Die Erinnerungen [The Reminiscenses]. München, Germany: F. A. Herbig. Noelle, E., & Neumann, E. P. (Eds.). (1956), Jahrbuch der öffentlichen Meinung 1947–1955. [Yearbook of Public Opinion 1947-1955]. Allensbach, Germany: Verlag für Demoskopie. Noelle-Neumann, E., & Petersen, T. (2005). Alle, nicht jeder. Einführung in die Methoden der Demoskopie [Everybody, Not Everyone. Introduction into the methods of opinion polling] (4th ed.). Berlin, Germany: Springer. Petersen, T. (2002). Das Feldexperiment in der Umfrageforschung [Field Experiments in Opinion Polls]. Frankfurt, Germany: Campus. Weimann, G. (1991). The influentials. Back to the concept of public opinion leaders. Public Opinion Quarterly, 55, 267–279.

5 METHODOLOGICAL CONUNDRUMS IN SPIRAL OF SILENCE RESEARCH Jörg Matthes and Andrew F. Hayes

As a pollster and professor of communication at the University of Mainz, Elisabeth Noelle-Neumann was not only a public opinion and communication scholar; she was also a methodologist at heart. She provided the field with a specified research methodology to test the spiral of silence, was known for her intuition and hardheadedness when constructing items and designing questionnaires, and was never reluctant to criticize other spiral of silence studies she perceived as inappropriate in methodological terms. Needless to say, Elisabeth Noelle-Neumann engendered a rich legacy of ideas that sparked many interesting debates and infused the fields of public opinion and political communication with theoretical and methodological vitality. Since the publication of her 1974 article in the Journal of Communication, Noelle-Neumann had clear and strict ideas about how one should and should not test the spiral of silence. Yet ironically, and despite her own studies and hundreds of others conducted across the globe, the full complexity of spiral of silence theory (including all of its aspects and assumptions) has never been subjected to a comprehensive empirical test—and it probably never will be. Instead, spiral of silence researchers tend to separate the model into discrete, manageable chunks or assumptions, implying that other aspects of the theory remain constant, are irrelevant, or can be ignored. Yet, after almost 40 years of scholarship with an abundance of designs, operationalizions, and measures, there is still no agreement on how to design a test of these assumptions, posing some knotty methodological problems to spiral of silence scholarship. In this chapter, we revisit and tackle these issues by discussing the main pillars of the theory from a methodological vantage point. To put this chapter in context, we begin with a critical discussion about the measurement of the key independent and dependent variables in spiral of silence research: fear of social

Methodological Conundrums in Spiral of Silence Research

55

isolation, perception of the opinion climate, and willingness to express one’s views. We then turn to two largely neglected but still unresolved methodological challenges: testing opinion spirals over time and designing a test of the spiral of silence across countries. The Methodological Toolkit of Spiral of Silence Scholarship The Theory of Public Opinion, usually referred to as the spiral of silence (NoelleNeumann, 1974, 2001), can be described as a complex pattern of interlinked propositions explaining how people observe (and react to) what other people think about current social and political topics. As people have an innate fear of becoming socially isolated and ostracized, they are constantly observing their surroundings for indicators of the opinion climate. Primarily based on what the media conveys about public opinion and what people experience in dayto-day discussions with others, they develop a sense about the positions that are shared by most other people and those that are less popular. When individuals perceive their own opinions to be in a minority or losing strength in public sentiment, then they are less likely to publicly express their views compared to those who feel their opinions are shared by most others. As a result, when some people self-censor their views in public, the climate of opinion further shifts in the direction of the perceived majority, resulting in a downward spiraling of the size of the minority until only a small group of “hard-core” people remain that still express their views irrespective of their minority status. Such a spiral of silence is contingent on at least two additional assumptions. First, as a theory of media effects, spiral of silence theory posits that only consonant media coverage can convey the climate of opinion. Second, Noelle-Neumann argued (e.g., Noelle-Neumann, 2001) that in order for a spiral of silence to develop the issue at stake must be one imbued with controversy and questions of morality or ethics (for a discussion, see Hayes & Matthes, in press). Fear of Social Isolation Noelle-Neumann assumed that “most people are afraid of becoming isolated” (1977, p. 144), and given this assumption is not bound to specific situations (such as holding minority views), fear of social isolation (FSI) can be understood as a general tendency of individuals (see Hayes, Matthes, & Eveland, 2013). However, FSI—while mentioned in almost every single spiral of silence study—has rather been treated as a factum or constant not necessitating further empirical analysis. Others have construed FSI as a state explaining opinion expression in a specific situation (e.g., Neuwirth, Frederick, & Mayo, 2007). Glynn and McLeod (1985) took a different stand and suggested to treat FSI as a continuum along which people differ rather than as a social constant or a state. In line with this, some recent research has advocated that fear of social isolation

56

Jörg Matthes and Andrew F. Hayes

is more pronounced in some people than in others, either as an antecedent to (or intervening variable between) perceptions of the opinion climate and public expression of opinions (Petrič & Pinter, 2002; Shoemaker, Breen, & Stamper, 2000). From a methodological perspective, however, most of these operationalizations can be criticized for their lack of validity and reliability. Some studies only worked with single items making reliability assessments impossible (Huang, 2005; Moy, Domke, & Stamm, 2001). Other measures (e.g., Ho & McLeod, 2008; Scheufele, Shanahan, & Lee, 2001) can be criticized for criterion contamination because they also tap the criterion that fear of social isolation aims to predict (i.e., fearing to become isolated by public speaking a minority viewpoint). Yet other studies have used more valid items but fail to convincingly demonstrate (or even attempt to demonstrate) their measurement instrument’s psychometric properties such as reliability and validity (e.g., Scheufele et al., 2001). As should be apparent, spiral of silence scholarship lacks a standardized, reliable, and valid measure of FSI. Certainly, one possible reason is that FSI was not the main independent variable in these studies. Another reason could be, as Willnat, Lee, and Detenber (2002) put it, that studies often lack a clear conceptualization and operational definition of FSI. In response to this, Hayes et al. (2013) have recently proposed a five-item measure of FSI (e.g., “It is important to me to fit into the group I am with”), demonstrating that people who differ in their level of FSI also differ in the attention they pay to information in the media environment about what others are thinking—one of the theory’s fundamental but hitherto unexplored predictions. Although Hayes et al. reported strong evidence for the psychometric properties of their measure—using data from eight countries across the globe along the way—further evidence is needed to fully validate this scale, especially when exploring other roles that fear of social isolation might play in the dynamics of public opinion evolution, such as moderating the relationship between opinion climate and opinion expression. Opinion Climate As a second key aspect of the theory, it is assumed that individuals possess a quasi-statistical sense (or organ) that helps them “form a picture of the distribution of opinion in their social environment and of the trend of opinion” (Noelle-Neumann, 1974, p. 45). This quasi-statistical sense is stimulated mainly by the mass media, especially in circumstances of consonant media messages regarding the opinion climate. According to the original conception of the theory, the quasi-statistical sense of people remains inaccessible to conscious thought. Therefore, Noelle-Neumann (2001, p. 60) repeatedly stressed that investigative approaches that focus solely on whether respondents feel they share the majority opinion or not—regardless of which position they

Methodological Conundrums in Spiral of Silence Research

57

actually take on a concrete issue—generally do not yield any particularly valuable results. The question of which view is in fact gaining or losing ground among the population is decisive here, not whether the individual also consciously perceives this process. Most studies, in contrast, have assessed people’s subjective sense of the opinion climate. This reflects the assumption that the perceptions of the opinion climate, rather than the true distribution of opinions, shape people’s willingness to express their views. Shamir (1993), for instance, asked respondents to estimate the share of the population favoring or opposing a political position. Others took perceived reference groups’ opinions as predictors of opinion expression (Moy et al., 2001). Yet others measured perceived opinion climate as conveyed through the mass media (Matera & Salwen, 1992), what “the public” believes broadly construed (Lasorsa, 1991), or from recent conversational partners (Scheufele et al., 2001). Frequently, investigators measure self-other congruence with respect to several reference groups simultaneously (e.g., Moy et al., 2001). Answers to these questions are then usually combined with the respondents’ own opinions as an operationalization of whether or not their views match with what they believe is in the majority or minority opinion. But are such questions that tap people’s beliefs about what the general public thinks methodologically sound indicators for belonging to a majority or minority? Besides a lack of reliability information that comes with the use of single items, scholars end up with rather crude ordinal variables when they combine two single items (i.e., perceived majority opinion and own opinion). Furthermore, these operationalizations tend to ignore the key assumption of spiral of silence theory that the quasi-statistical sense is fueled by the mass media. As originally conceived in spiral of silence theory, the mass media are the primary source of information about the climate of opinion. From this information, people make decisions as to how comfortable they are publicly voicing an opinion that runs counter to what the consonant and ubiquitous mass media report. However, only a very few studies have gathered content analytical data to compare measures of media exposure with data on the opinion climate (e.g., Eveland, McLeod, & Signorielli, 1995). Noelle-Neumann herself collected content analytical data, but at the aggregate—not individual— level. Yet others have correlated climate perceptions with media use (Neuwirth & Frederick, 2004). Taken together, such (mostly cross-sectional) designs are unable to establish that there is a connection between news content that people were exposed to and their quasi-statistical sense about the opinion climate. Public Opinion Expression There is also a considerable debate about how to operationalize people’s willingness to express their views. Noelle-Neumann originally used a conversation with passengers on a train, a procedure which has come to be known as

58

Jörg Matthes and Andrew F. Hayes

the “train-test.” As train travel is not common in all countries and cultures, the test has been adapted to other public situations that differ in the size of the hypothetical public and the potential audience—such as a bus ride (Shamir, 1997), an airplane fl ight (Lasorsa, 1991), a social gathering (Moy et al., 2001; Scheufele et al., 2001), an interview with a reporter (Shamir, 1997), a doctor’s waiting room (Petrič & Pinter, 2002), a wedding (Willnat et al., 2002), or a public meeting (Gonzenbach & Stevenson, 1994). As elegant and intuitive as these approaches are, they have all been criticized for their hypothetical nature (Glynn, Hayes, & Shanahan, 1997; Hayes, Uldall, & Glynn, 2010; Scheufele et al., 2001). That is, if spiral of silence theory predicts that the climate of opinion exerts an influence on opinion expression— because of a realistic threat to become socially isolated—then it is more than likely that hypothetical and real conversational settings differ in their susceptibility to this process (Scheufele et al., 2001). With the current state of the literature, it remains to be demonstrated that hypothetical measures can capture how people actually behave in real public situations. Therefore, a more rigorous test would require that people are given the possibility to express their views in a true conversational setting where the climate of opinion is either varied experimentally or measured post-hoc. Such settings could also vary the “publicness” of an act of opinion expression, the anonymity of subjects, the size of the audience that is present, and the ability of the audience to demonstrate open disapproval. Even after 40 years of scholarship, such designs are urgently needed to test this aspect of the theory (see Hayes et al., 2010, for a recent exception). Besides, more efforts are needed to establish the reliability and construct validity of these various measures of willingness to speak out. Observing the “Spiral” in the Spiral of Silence According to the notion of a silence spiral, it is the dominant camp that will become more dominant and louder over time, while the losing camp will become increasingly silent. Such a time factor, as Scheufele and Moy (2000, p. 11) called it, refers to a “process of formation, change, and reinforcement of public opinion,” so that, “[o]ver time changing perceptions of the opinion climate influence people’s willingness to express minority opinions” (p. 10). Time is therefore of utmost importance when designing studies to test the spiral of silence. Yet time has been shamefully neglected in previous research. From a theoretical perspective, four distinct but intertwined predictions about the relationship between opinion climate and opinion expression can be made. First, perceived opinion climate and opinion expression should be correlated; that is, individuals who feel they are in a minority at time-point 1 should be less likely to speak out at time-point 1 when compared to people who perceive themselves to be in a majority position. This rather static idea can be called the social conformity hypothesis. Second, because this process unfolds over

Methodological Conundrums in Spiral of Silence Research

59

time, changes in the opinion climate should lead to changes in willingness to speak out. That is, if the opinion climate gets more hostile over time, then an individual’s willingness to publicly express his or her opinion should decrease over time as well. We call this change hypothesis. This hypothesis basically says that social conformity unfolds over time. Third, and at the heart of the theory, “the tendency of the one to speak up and the other to be silent starts off a spiraling process which increasingly establishes one opinion as the prevailing one” (Noelle-Neumann, 1974, p. 44). This process is triggered by ongoing exposure to the mass media that reflect the opinion climate (i.e., through opinion polls), and by observing one’s close surroundings. “Supporters of one side […] are willing to stand up for their opinion. […] Thus, they appear stronger to the observer than they really are and this, again, induces other people to take a stand also, while the opposition hides more and more and retires into silence” (Noelle-Neumann, 1977, p. 149). Rephrased, the willingness of one person to speak out actually changes the subsequent opinion climate so that, in turn, the dominant opinion (climate) will become even stronger. We call this the reinforcing spiral hypothesis. Fourth, and the most challenging assumption, Noelle-Neumann theorized that this process unfolds over time until it leads to a kind of dead end or paralysis, where only the hardcore remain and no further dynamic changes can be observed. We call this, for lack of a better term, the paralysis hypothesis. Noelle-Neumann (1974, 2001) approached these various ideas using trendsurvey data, demonstrating that the number of people supporting a certain position increased or decreased depending on the opinion climate. Such aggregate designs allow an intuitive understanding of the ups and downs of key variables. However, due to the ecological fallacy problem, we are unable to draw any conclusions about a spiraling process at the individual level. As a consequence, such data are ill-suited to address the dynamic aspects of the theory. Even more alarming, the vast majority of studies on the spiral of silence have employed cross-sectional survey data. Such data can be used only to test the static social conformity hypothesis, not the other three hypotheses. Thus, as Neuwirth et al. (2007) stated, there “is a clear need for panel designs that will allow researchers to explore how shifts in fear [of isolation] and judgments of majority opinion influence changes in open discussion and conformity-based deception” (p. 466; similarly Gonzenbach & Stephenson, 1994). However, the few panel studies on the spiral of silence also fall short in testing the opinion dynamics the theory proposes. McDonald, Glynn, Kim, and Ostman (2001), for instance, conducted a secondary analysis of a four-wave panel survey conducted in 1948 and estimated the effect of opinion climate on candidate preference separately for each wave. However, like NoelleNeumann’s aggregate level change model, such results say nothing about the individual change of climate perceptions and opinion expression. Besides, it is a very crude and imprecise procedure that builds on many fishy assumptions and

60

Jörg Matthes and Andrew F. Hayes

usually leads to erroneous conclusions. Using a change score model, Shamir (1997) regressed a two-wave change score of opinion expression on climate perceptions, explaining some variance of the change score. As intuitive as change scores may be as a measure of change, correlates of change are typically hard to interpret, as change scores are contaminated by regression to the mean, and tend to be negatively correlated with time 1 state (see, e.g., Hertzog & Nesselroade, 2003). Taken together, it appears that the designs employed in spiral of silence scholarship are unable to test the spiral process over time; that is, the change, the reinforcing spiral, and the paralysis hypotheses. Further shortcomings include, again, the use of single items in many studies and the failure to address the invariance of measures over time (i.e., measurement invariance, see Pitts, West, & Tein, 1996). The theoretically and methodically more intriguing and challenging dynamic aspects—the reinforcing spiral and paralysis hypotheses—require data that involve at least three panel waves. With such data, latent growth modeling (see e.g., Selig & Preacher, 2009) constitutes a method of choice. Still, a serious pitfall is that the choice of the time interval exerts a substantial impact on the likelihood of observing an opinion spiral. As discussed by Selig and Preacher (2009), for instance, one might be unable to observe an effect because the interval was either too short for the dynamic process to start, or too long so that any potentially observable effect has long faded. This creates a challenge to spiral of silence researchers because often there is no available theoretical or empirical basis for choosing a time-lag for a given specific issue of interest. For the very few panel studies available in spiral of silence research, intervals vary from several weeks to several years, which points to fundamentally different processes evolving over time. One solution to this problem might be to conduct studies employing varying time-lags for subsets of a sample in a multi-wave design. Examination of these dynamics in multi-wave experimental settings would additionally allow for a more fi ne-grained analysis of variation over time, between and within individuals. Testing the Universal Nature of the Theory As another blind spot in spiral of silence research, the cross-cultural applicability of the theory “has remained virtually unanswered” (Scheufele & Moy, 2000, p. 17; see Matthes, Hayes, Rojas, Shen, Min, and Dylko, 2012). The few studies that have attempted to assess the cross-cultural generalizability of the theory’s predictions are mainly based on comparisons between Asian and Western countries. Based on the assumption that collectivist cultures like those in Asia value the collective work of groups while Western cultures tend to emphasize the strength of the individual, it is commonly hypothesized that spiral of silence processes should be at work more prominently in collectivist rather than individualist cultures. Asking respondents about the very same issue, Lee,

Methodological Conundrums in Spiral of Silence Research

61

Detenber, Willnat, Aday, and Graf (2004) found a spiral of silence effect in Singapore but not in the United States. These fi ndings were validated by Huang’s (2005) study using different issues in American (i.e., American involvement in Somalia) and Taiwanese samples (i.e., the timing of a presidential election). From a methodological perspective, however, the failure to fi nd effects for individualist countries can be the result of using different issues in different countries—simply because those issues cannot be compared or do not meet the requirements for testing the theory in a single country (Matthes et al., 2012). Alternatively, one could fail to replicate fi ndings across countries because the very same issue was used—due to a substantially different meaning and fundamentally different contexts in those countries. Testing the theory across countries, therefore, requires that the key variables are reasonably similar with respect to their conceptualization and operationalization. This strategy, however, can result in a research paradox (Matthes et al., 2012). On the one hand, it might seem sensible to use different indicators in different countries because of country-level idiosyncrasies. On the other hand, using different indicators makes it hard—if not impossible—to compare fi ndings across studies. Similarly, what might be a morally-laden or controversial issue at one place can be completely uncontroversial at another. In short, as long as spiral of silence scholars work with context-specific measures, any significant or non-significant result can be attributed to the characteristics or appropriateness of that particular context. This might lead scholars to rationalize their decisions posthoc, depending on a study’s fi ndings. One solution to this challenge has recently been proposed by Matthes et al. (2012). They suggest working with situation-independent measures, such as personality traits. The major advantage of this strategy is that it aggregates across contexts within a specific country or culture. Traits are understood as situation-independent, as are the general tendencies of individuals. Thus, issueor situation-specific idiosyncrasies have less impact on the relationships behind a spiral of silence compared to issue-specific measures. Matthes et al. posit that two personality traits reflect the basic mechanisms behind a spiral of silence: (a) fear of social isolation, as the driving force that sets the spiral into motion (Hayes et al., 2013), and (b) willingness to self-censor (WTSC; Hayes et al., 2005)—how willing a person is to publicly express minority viewpoints. If some people generally and across situations fear social isolation more than others, then these people should also generally and across situations self-censor their opinions in the face of hostile opinion surroundings. Hayes et al. (2013) and Matthes et al. (2012) have developed Spanish, Russian, Korean, German, French, and Chinese versions of both measures. They administered these scales to sample from nine countries using the same sampling company, sample size, and data collection procedure. The positive correlation between FSI and WTSC in eight of the nine countries suggests substantial cross-cultural generality of basic spiral of silence processes. Of course, such a dispositional approach

62

Jörg Matthes and Andrew F. Hayes

cannot be considered as an ideal and full-fledged test of spiral of silence theory in a single country. It is also not suited to model the dynamic aspect of the theory. But it is a meaningful way of testing a basic component of the theory across countries. Yet more data are needed to understand whether the relationship between fear of social isolation and willingness to self-censor is contingent on cultural, political, and societal factors. Conclusion There seems to be as many operationalizations of spiral of silence theory’s key independent and dependent variables as there are empirical studies, which reflects a lack of consensus on how to design methods and measures to test central predictions of the theory. Furthermore, when carefully looking at NoelleNeumann’s complex predictions, we must conclude that the accumulated empirical evidence is inconsistent with the postulated mechanisms. At best the evidence is inconclusive. And fi nally, there seems to be a methodological gap between Noelle-Neumann’s own designs and measures and what other scholars have suggested following up on her 1974 Journal of Communication piece. We believe, if the theory is taken literally, that there is a strong need for alternative and complementary methodologies sensitive to the theory’s predictions. In particular, longitudinal studies (combined with content analytical data at the individual level) and observational experiments would provide important insights into the spiraling process, as well as the opportunity to causally test the main predictions of the theory. Scholars need to invest more efforts in establishing the reliability and validity of their measurement procedures, and they need to be more receptive to the state of the art of social science methodology, especially with regard to data analysis. But should we take the theory literally? As evident throughout this book, spiral of silence theory consists of a set of related predictions and contingent conditions. It has never been shown comprehensively, and it probably never will be, that those who fear social isolation are more attentive to the opinion climate about a moralized issue, which in turn, makes them more susceptible to majority cues in constant media coverage, so that precisely these people tend to suppress their opinions in public situations, starting a spiraling process that changes future opinion climate again pushing up this whole process until a small group of die-hards remain. Without question, ignoring or falsifying only one single piece of this puzzle may already distort the whole picture. What is more, how can one foresee if media coverage will be consonant, if people expose themselves to this coverage given a multitude of news sources online and offl ine, if there will be a majority climate, if the issue will be emotionalized, and if there is an optimal point in time to observe all of this when designing one’s instruments for a three-wave panel study? With this cornucopia of conditions and assumptions, it is almost impossible to truly test the theory

Methodological Conundrums in Spiral of Silence Research

63

of spiral of silence. Whatever the disconfi rming fi nding is, one always could explain that some condition of the theory was not met, that something was not right, or that just something else happened which doesn’t necessarily contradict the theory. This makes the theory impossible to falsify. We argue, therefore, that spiral of silence theory should not be treated as a literally applicable, word-for-word set of predictions that must be translated to a research design in line with Noelle-Neumann’s original ideas. Instead, we believe spiral of silence is a handy theoretical tool that is best suited to describe some fascinating aspects of public opinion dynamics, stressing the importance of observing other people’s opinions. Testing (or exploring) some selected aspects of the theory has been nourishing new ideas in public opinion research across the globe, increasing our sensitivity to methodological issues when studying public opinion, and helping us to understand the role of the opinion climate for people’s willingness to speak out, their attitudes, or voting decisions. To understand the phenomenon behind the spiral of silence, and to move public opinion scholarship forward, there is therefore no need for a benchmark design or a single one-and-only operationalization of the theory’s classic independent and dependent variables. There is also no need to slavishly adopt every aspect and prediction of the theory. There is, however, a more general need for carefully designed, multi-item measurement instruments, internally and externally valid experimental procedures, and increased sensitivity to the dynamics and processes that unfold over time at the individual level. References Eveland, W. P., Jr., McLeod, D. M., & Signorielli, N. (1995). Actual and perceived U.S. public opinion: The spiral of silence during the Persian Gulf War. International Journal of Public Opinion Research, 7, 91–109. Glynn, C. J., & McLeod, J. M. (1985). Implications of the spiral of silence theory for communication and public opinion research. In K. R. Sanders, L. L. Kaid, & D. Nimmo (Eds.), Political Communication Yearbook 1984 (pp. 43–65). Carbondale: Southern Illinois University Press. Glynn, C. J., Hayes, A. F., & Shanahan, J. E. (1997). Perceived support for one’s opinions and willingness to speak out: A meta-analysis of survey studies on the “spiral of silence.” Public Opinion Quarterly, 61, 452–461. Gonzenbach, W. J., & Stevenson, R. L. (1994). Children with AIDS attending public school: An analysis of the spiral of silence. Political Communication, 11, 3–18. Hayes, A. F., Glynn, C. J., & Shanahan, J. (2005). Willingness to self-censor: A construct and measurement tool for public opinion research. International Journal of Public Opinion Research, 17, 298–323. Hayes, A. F., & Matthes, J. (in press). Self-censorship, the spiral of silence, and contemporary political communication.  In K. H. Jamieson & K. Kenski (Eds,), Oxford handbook on political communication. Oxford, England: Oxford University Press.  Hayes, A. F., Matthes, J., & Eveland, W. P. (2013). Stimulating the quasi-statistical organ: Fear of social isolation motivates the quest for knowledge of the opinion climate. Communication Research, 40, 439‒462.  Hayes, A. F., Uldall, B., & Glynn, C. J. (2010). Validating the willingness to self-censor scale II: Inhibition of opinion expression in a real conversational setting. Communication Methods and Measures, 4, 256–272.

64 Jörg Matthes and Andrew F. Hayes

Hertzog, C., & Nesselroade, J. R. (2003). Assessing psychological change in adulthood: An overview of methodological issues. Psychology and Aging, 18, 639–657. Ho, S. S., & McLeod, D. M. (2008). Social-psychological influences on opinion expression in face-to face and computer-mediated communication. Communication Research, 35, 190–207. Huang, H. (2005). A cross-cultural test of the spiral of silence. International Journal of Public Opinion Research, 17, 324–345. Lasorsa, D. L. (1991). Political outspokenness: Factors working against the spiral of silence. Journalism Quarterly, 68, 131–140. Lee, W., Detenber, B. H., Willnat, L., Aday, S., & Graf, J. (2004). Public outspokenness in two cultures: A test of the spiral of silence theory in Singapore and the US. Asian Journal of Communication, 14, 205–226. Matera, F. R., & Salwen, M. B. (1992). Support for Radio Marti among Miami’s Cubans and non-Cubans. International Journal of Intercultural Relations, 16, 135–144. Matthes, J., Hayes, A. F., Rojas, H., Shen, F., Min, S. J., & Dylko, I. (2012). Exemplifying a dispositional approach to cross-cultural spiral of silence research: Fear of social isolation and the inclination of self-censor. International Journal of Public Opinion Research, 24, 287–305. McDonald, D. G., Glynn, C. J., Kim, S.-h., & Ostman, R. E. (2001). The spiral of silence in the 1948 presidential election. Communication Research, 28, 139–155. Moy, P., Domke, D., & Stamm, K. (2001). The spiral of silence and public opinion on affi rmative action. Journalism & Mass Communication Quarterly, 78, 7–25. Neuwirth, K., & Frederick, E. (2004). Peer and social influence on opinion expression: Combining the theories of planned behavior and the spiral of silence. Communication Research, 31, 669–703. Neuwirth, K., Frederick, E., & Mayo, C. (2007). The spiral of silence and fear of isolation. Journal of Communication, 57, 450–468. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24, 43–51. Noelle-Neumann, E. (1977). Turbulences in the climate of opinion: Methodological applications of the spiral of silence theory. Public Opinion Quarterly, 41, 143–158. Noelle-Neumann, E. (2001). The spiral of silence: Public opinion — Our social skin (3rd ed.). Chicago, IL: University of Chicago Press. Petrič, G., & Pinter, A. (2002). From social perception to public expression of opinion: A structural equation modeling approach to the spiral of silence. International Journal of Public Opinion Research, 14, 37–53. Pitts, S. C., West, S. G., & Tein, J.-Y. (1996). Longitudinal measurement models in evaluation research: Examining stability and change. Evaluation and Program Planning, 19, 333–350. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12, 3–28. Scheufele, D. A., Shanahan, J., & Lee, E. (2001). Real talk: Manipulating the dependent variable in spiral of silence research. Communication Research, 28, 304–324. Selig, J. P., & Preacher, K. J. (2009). Mediation models for longitudinal data in developmental research. Research in Human Development, 6, 144–164. Shamir, J. (1993). Pluralistic ignorance revisited: Perception of opinion distributions in Israel. In: International Journal of Public Opinion Research, 5, 22–39. Shamir, J. (1997). Speaking up and silencing out in face of a changing climate of opinion. Journalism & Mass Communication Quarterly, 74, 602–614. Shoemaker, P. J., Breen, M., & Stamper, M. (2000). Fear of social isolation: Testing an assumption of spiral of silence theory. Irish Communications Review, 8, 65–78. Willnat, L., Lee, W., & Detenber, B. H. (2002). Individual-level predictors of public outspokenness: A test of the spiral of silence theory in Singapore. International Journal of Public Opinion Research, 14, 391–412.

6 SPEAKING IN SPIRALS An Updated Meta-Analysis of the Spiral of Silence Carroll J. Glynn and Michael E. Huge

The spiral of silence was developed as a theory of a silent majority by German pollster and political scientist Elisabeth Noelle-Neumann (1974, 1984, 1993). By proposing separate behavioral tendencies for those in the majority and those in the minority, the spiral of silence addresses the theoretical formation and maintenance of public opinion, as well as its subsequent impacts on different societal factions. Though the theory is often credited as being one of the more influential models of how public opinion is formed and maintained, Glynn, Hayes, and Shanahan (1997) demonstrated that the spiral’s effect—at least according to the extant published and unpublished literature at that time—was quite small, though the correlation between perceived opinion support and willingness to speak out was both positive and significant. In this chapter, we update the previously reported meta-analysis to include both published and non-published articles that have appeared since Glynn et al.’s original 1997 publication. The spiral of silence remains one of the few theoretical attempts to understand public opinion as a multi-level process. Noelle-Neumann argued that those who perceive that they are in the minority—even if they are actually in the majority—will be more hesitant to express their viewpoint to an individual or group who is thought to hold a different or opposite viewpoint. Those in the majority, on the other hand, feel increasingly confident in expressing their views. The “spiraling” process is activated when these expressions of perceived majority status lead to further withholding by those who believe they are in the minority. Majority views are thus overrepresented, while minority viewpoints become correspondingly underrepresented.

66

Carroll J. Glynn and Michael E. Huge

Spirals in Action It is perhaps the dynamic and iterative nature of the spiral of silence process that has garnered the theory so much attention. Rather than proposing a snapshot view of one point in time, the spiral of silence proposes an over-time model that is continuously affected and impacted by individuals and groups and their perceptions of the current opinion climate. Noelle-Neumann (1984, 1993) based part of her theoretical argument on the work of de Tocqueville in noting that individuals often feel more comfortable going along with a seemingly incorrect majority instead of risking social alienation from the group or larger segment of society (i.e., fear of isolation). However, not all members of a group or public feel this same need to belong. As noted in research on the spiral of silence, both the hardcore and avant-garde are willing to express unpopular opinions (Noelle-Neumann, 1993). The hardcore are willing to maintain and express unpopular views even after they have gone out of favor, while the avant-garde are ahead of the curve in expressing opinions that are not yet popular. Yet it is not only interpersonal interaction that can impact the public opinion process. Media outlets—which Noelle-Neumann saw as “consonant” and “ubiquitous”—present views on issues that can impact how opinion evolves over time. Individuals scan their environment in an effort to perceive how others feel about a given political or social issue. Through a “quasi-statistical sense” (Noelle-Neumann, 1993), individuals are able to gauge the opinions of others and can then use that information when deciding how and if they will present their own views. In this fashion, the spiral of silence incorporates both the interpersonal and mass media components of opinion formation. Though the operationalization of the dependent variable in spiral of silence research has varied widely, a respondent’s willingness to express his or her opinion is most often used. Early research focused on creating a hypothetical—and yet realistic—public situation in which the respondent could choose to share their opinion or remain silent (e.g., Noelle-Neumann, 1984). Studies have also focused on measuring voting intention, willingness to talk about the issue with close friends and family, monetary contributions to political candidates, and the expression of views on a news broadcast. Meta-Analysis Meta-analysis allows for a quantitative comparison of multiple studies across time. By combining results from separate studies, meta-analytic techniques allow researchers to obtain an overall view of the extant data and results for a given theory. Glynn et al. (1997) ran a meta-analysis on spiral of silence studies that tested the relationship between the perceived climate of opinion and respondents’ willingness to express an opinion, finding a small but significant and positive relationship (Pearson r = 0.054, median Pearson r = .063, 95% confidence interval from .028 to .080). Further analysis in their study revealed

Speaking in Spirals

67

that this result was indeed significant (Stouffer’s Z = 5.21, p < .0001, fail-safe N of 221).1 Glynn et al. also tested the impact of moderators on willingness to express an opinion, but found no significant relationships between the moderators tested. We used similar methods to Glynn et al. (1997) and collected relevant data since the publication of the original meta-analysis. We used the ISI Web of Knowledge online database to search for articles from 1995 to 2011 containing “spiral” and “silence” in the title, keywords, or abstract. We also scanned references of found articles to identify additional studies and contacted over 20 researchers with a history of spiral of silence research and publication. Glynn et al.’s restriction on the inclusion of articles was followed, limiting our research “to studies where an individual was presented with a hypothetical situation and was then asked whether he or she would be willing to enter a conversation or express his or her opinion” (p. 454). Studies were further limited to approaches “where the investigator explicitly operationalized perceptions of the opinion distribution or ‘climate’ (current or future) and then contrasted them with the participant’s own opinion” (p. 455). Results The updated analysis reveals a nearly-identical pattern of results when compared to Glynn et al. (1997). Twelve new studies with 19 separate tests were included for the present analysis. Fourteen reported a positive correlation between a supportive opinion context and respondents’ willingness to express their opinion. Three studies reported a negative relationship, and two reported a correlation of zero between the perceived opinion climate and willingness to express an opinion. All effect sizes were converted to Pearson’s correlation coefficients (r), and those correlations were transformed for analysis using Fisher’s r-to-Z transformation (Rosenthal, 1991, pp. 73–74). From the 12 obtained studies, we computed correlations based on almost 9,800 responses (see Table 6.1). The overall correlation between perceived support and willingness to express an opinion was very small (Pearson r = .050, SD = 0.51, 95% CI = .031 to .069). Even though the average correlation was quite small, the combined probability of the new results is statistically significant (Stouffer’s Z = 4.80, p < .0001, failsafe N = 57). When the 12 new studies were included in an overall analysis of the spiral of silence effect, we again obtained a small, but positive and significant result (overall Pearson r = .053, SD = 0.58, 95% CI = .042 to .064; Stouffer’s Z = 7.40, p < .0001, fail-safe N = 584). In other words, though the effect itself was small, the combined probability of obtaining such a result from the 12 new studies is significantly different from chance. Given that the updated correlation is nearly identical to the previous result, it appears that the perceived opinion climate does have a slight impact on opinion expression, and this impact has been quite consistent over the past three decades of research.

68

Carroll J. Glynn and Michael E. Huge

TABLE 6.1 Correlations and Z-scores for Studies Contributing Data to the Meta-Analysis Study

Authors

Year

Time

N

Pearson r

Z score

1

Noelle-Neumann

1973

Current

1191

0.10

3.46

2

Mohn

1983

Current

219

0.07

1.03

3

Mohn

1983

Future

167

0.06

0.88

4

Donsbach & Stevenson

1984

Current

341

-0.02

-0.37

5

Donsbach & Stevenson

1984

Future

335

0.09

1.66

6

Glynn & McLeod

1984

Future

336

0.14

2.57

7

Lashin

1984

Current

352

0.08

1.50

8

Lashin

1984

Future

347

0.08

1.49

9

Lee

1989

Current

287

0.07

1.18

10

Matera & Salwen

1989

Current

347

0.10

1.86

11

Mutz

1989

Current

240

0.00

0.00

12

Mutz

1989

Future

240

0.00

0.00

13

Tokinoya

1989

Current

294

0.09

1.54

14

Tokinoya

1989

Current

298

-0.09

-1.55

15

Tokinoya

1989

Current

310

0.03

0.53

16

Tokinoya

1989

Current

352

0.05

0.94

17

Tokinoya

1989

Current

300

-0.04

-0.69

18

Salmon & Neuwirth

1990

Current

432

0.07

1.58

19

Salmon & Oshgan

1990

Current

478

0.00

0.00

20

Fuchs, Gerhards, & Friedhelm

1992

Current

453

0.10

2.13

21

Fuchs, Gerhards, & Friedhelm

1992

Future

377

-0.03

-0.58

22

Shamir & Shamir

1993

Current

1197

0.04

1.38

23

Shamir & Shamir

1993

Current

685

0.05

1.31

24

Shamir & Shamir

1993

Current

1197

0.01

0.35

25

Shamir & Shamir

1993

Future

685

0.05

1.31

26

Salwen, Lin, & Matera

1994

Current

377

0.03

0.58

27

Salwen, Lin, & Matera

1994

Current

213

0.08

1.16

28

Salwen, Lin, & Matera

1994

Current

335

0.00

0.00

29

Wedel

1994

Current

520

0.13

2.97

30

Wedel

1994

Future

520

0.17

3.90

31

Gonzenbach, Joblonski, & Joblonski

1995

Current

300

0.00

0.00

32

Gonzenbach, Joblonski, & Joblonski

1995

Future

302

0.13

2.26

Speaking in Spirals

Study

Authors

69

Year

Time

N

Pearson r

Z score

33

Shamir

1995

Current

685

0.05

1.31

34

Hayes & Glynn

1996

Current

94

0.23

2.23

35

Hayes & Glynn

1996

Current

84

0.25

2.29

36

Neuwirth

2000

Current

305

-0.10

-1.74

37

Neuwirth

2000

Future

305

-0.05

-0.87

38

Moy, Domke & Stamm

2001

Current

217

0.09

1.32

39

Moy, Domke & Stamm

2001

Future

217

0.01

0.15

40

Scheufele, Shanahan, & Lee

2001

Current

358

-0.03

-0.57

41

Scheufele, Shanahan, & Lee

2001

Future

358

0.08

1.51

42

Moreno-Riano

2002

Current

407

-0.01

-0.20

43

Willnat, Lee, & Detenber

2002

Current

627

0.08

2.00

44

Willnat, Lee, & Detenber

2002

Future

627

-0.03

-0.75

45

Kim, Han, Shanahan, & Berdayes

2004

Current

368

0.11

2.11

46

Kim, Han, Shanahan, & Berdayes

2004

Future

368

0.09

1.72

47

Neuwirth & Frederick

2004

Current

397

0.07

1.39

48

Priest

2006

Current

579

0.05

1.20

49

Hayes

2007

Current

815

0.08

2.28

50

Neuwirth, Frederick, & Mayo

2007

Current

1126

0.01

0.34

51

Neuwirth, Frederick, & Mayo

2007

Current

1126

0.00

0.00

52

Ho & McLeod

2008

Current

352

0.09

1.69

53

Ho & McLeod

2008

Future

352

0.10

1.87

54

Matthes, Rios Morrison, & Schemer

2010

Current

2599

0.02

1.02

55

Matthes, Rios Morrison, & Schemer

2010

Current

1201

0.00

0.00

56

Yun & Park

2011

Current

126

0.14

1.56

Discussion This chapter serves as an update to a previously-conducted meta-analysis (Glynn et al., 1997) of the overall effect of Noelle-Neumann’s spiral of silence theory (1974, 1984, 1993). Evidence from new data is strongly aligned with the previous research, indicating the presence of a very small but significant relationship between the perceived climate of opinion and respondents’ willingness to express an opinion. By combining new results with data from the original analysis, we were able to narrow the 95% confidence interval

70 Carroll J. Glynn and Michael E. Huge

for the overall effect while also increasing the “fail safe N,” indicating that more studies yielding a null result (i.e., zero correlation between the perceived opinion climate and willingness to express an opinion) would need to be added to the mix in order to raise the probability level of the overall effect size beyond the level of chance (i.e., p > .05) for the average correlation. It should be noted, however, that in several of the more recent studies included in our review (e.g., Matthes, Morrison, & Schemer, 2010; Neuwirth, Frederick, & Mayo, 2007) multivariate analysis reveals that the effect of the existing climate is quite small—if at all significant—when other key variables are also included as predictors. Work on predicting opinion expression at the individual level indicates that there is variability in how individuals respond in an opinion climate. Hayes and colleagues (Hayes, 2007; Hayes, Glynn, & Shanahan, 2005) found that shyness, communication apprehension, and willingness to selfcensor all impacted the degree to which opinions are expressed. Matthes et al. (2010) found that the climate of opinion only impacted opinion expression when opinions were held with low to moderate certainty. For individuals with high levels of certainty about their views, the opinion climate did not matter. Neuwirth et al. discovered that communication apprehension—measured as both a state and trait—and fear of isolation also impacted opinion expression within a given opinion climate. At a broader, more macro level, the notion that the media act in concert to present a unified view on a given topic seems at best dubious in today’s highly fragmented and specialized media environment. Those who are digitally connected to television, computers and smart phones are free to choose from an almost infi nite number of news sources, opinion content, and debate. Rather than presenting a consonant message—as Noelle-Neumann originally suggested—today’s environment allows and individual to obtain a plethora of perspectives. Some have pointed out, however, that the availability of such specialized content may encourage some to seek out only those news and opinion sources that support previously held opinions (e.g., Iyengar & Hahn, 2009; Slater, 2007). Even with these criticisms in mind, the spiral of silence has remained an important theoretical approach in the understanding of how public opinion is created, shaped, and solidified in the public sphere. By accounting for both macro- (e.g., media environments) and micro-level variables (e.g., interpersonal conversation), the theory attempts to explain public opinion as an iterative and over-time process. Note 1. Rosenthal’s (1979) fail-safe N quantifies the number of additional studies with a null result that would need to be added to the meta-analysis to raise the p value to a level of nonsignificance (e.g., p > .05).

Speaking in Spirals

71

References Donsbach, W., & Stevenson, R. L. (1984, May). Challenges, problems, and empirical evidence of the theory of the spiral of silence. Paper presented at the annual meeting of the International Communication Association, San Francisco. Fuchs, D., Gerhards, J., & Neidhardt, F. (1992). Offentliche Kommunikationsberetischaft: Ein test zentraler bestandteile der theorie der schweigespirale [Willingness to speak publicly: A test of central elements of the spiral of silence theory]. Zeitschrift fur Soziologie, 21, 284–295. Glynn, C. J., Hayes, A. F., & Shanahan, J. (1997). Perceived support for one’s opinions and willingness to speak out — A meta-analysis of survey studies on the “spiral of silence.” Public Opinion Quarterly, 61(3), 452–463. Glynn, C. J., & McLeod, J. M. (1984). Public opinion du jour: An examination of the spiral of silence. Public Opinion Quarterly, 48(4), 731–740. Gonzenbach, W. J., King, C. M., & Jablonski, P. M.  (1999).  Homosexuals in the military:  An analysis of the spiral of silence. Howard Journal of Communication, 10, 281–296. Hayes, A. F. (2007). Exploring the forms of self-censorship: On the spiral of silence and the use of opinion expression avoidance strategies. Journal of Communication, 57(4), 785–U237. doi:10.1111/j.1460-2466.2007.00368.x Hayes, A. F., & Glynn, C. J. (1996). Unpublished manuscript. Hayes, A. F., Glynn, C. J., & Shanahan, J. (2005). Validating the willingness to self-censor scale: Individual differences in the effect of the climate of opinion on opinion expression. International Journal of Public Opinion Research, 17, 443–455. Ho, S. S., & McLeod, D. M. (2008). Socia l-psychological influences on opinion expression in face-to-face and computer-mediated communication. Communication Research, 35(2), 190– 207. doi: 10.1177/0093650207313159 Iyengar, S., & Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59(1), 19–39. Kim, S. H., Han, M., Shanahan, J., & Berdayes, V. (2004). Talking on ‘sunshine in North Korea’: A test of the spiral of silence as a theory of powerful mass media. International Journal of Public Opinion Research, 16(1), 39–62. Lashin, Y. M. (1984). Testing the spiral of silence hypothesis: Toward an integrated theory of public opinion (Unpublished doctoral dissertation). University of Illinois at Urbana-Champaign. Lee, H. (1989). Media credibility and the spiral of silence. Paper presented at the annual August meeting of the Association for Education in Journalism and Mass Communication, Washington, D.C. Matera, F. R., & Salwen, M. B. (1989). Speaking up for one’s views: Support for radio marti among Miami’s Cubans and non-Cubans. Paper presented at the annual August meeting of the Association for Education in Journalism and Mass Communication, Washington, D.C. Matthes, J., Morrison, K. R., & Schemer, C. (2010). A spiral of silence for some: Attitude certainty and the expression of political minority opinions. Communication Research, 37(6), 774–800. doi:10.1177/0093650210362685 Moreno-Riano, G. (2002). Experimental implications for the Spiral of Silence. Social Science Journal, 39(1), 65–81. Moy, P., Domke, D., & Stamm, K. (2001). The spiral of silence and public opinion on affi rmative action. Journalism & Mass Communication Quarterly, 78(1), 7–25. Mutz, D. (1989). The influence of perceptions of media Influence: Third person effects and the public expression of opinions. International Journal of Public Opinion Research, 1, 3–23. Neuwirth, K. (2000). Testing the spiral of silence model: The case of Mexico. International Journal of Public Opinion Research, 12(2), 138–159. Neuwirth, K., & Frederick, E. (2004). Peer and social influence on opinion expression: Combining the theories of planned behavior and the spiral of silence. Communication Research, 31(6), 669–703. doi:10.1177/0093650204269388 Neuwirth, K., Frederick, E., & Mayo, C. (2007). The spiral of silence and fear of isolation. Journal of Communication, 57(3), 450–468. doi:10.1111/j.1460-2466.2007.00352.x

72

Carroll J. Glynn and Michael E. Huge

Noelle-Neumann, E. (1974). Spiral of silence: A theory of public opinion. Journal of Communication, 24(2), 43–51. Noelle-Neumann, E. (1984). The spiral of silence: Public opinion, our social skin. Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1993). The spiral of silence: Public opinion, our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Priest, S. H. (2006). Public discourse and scientific controversy — A spiral-of-silence analysis of biotechnology opinion in the United States. Science Communication, 28(2), 195–215. doi:10.1177/1075547006293918 Rosenthal, R. (1979). The ‘fi le drawer problem’ and tolerance for null results. Psychological Bulletin, 86, 638–641. Rosenthal, R. (1991). Meta-analytic procedures for social research. Newbury Park, CA: Sage. Salmon, C. T. & Neuwirth, K. (1990). Perceptions of opinion “climates” and willingness to discuss the issue of abortion. Journalism Quarterly, 67(3), 567–577. doi:10.1177/107769909006700312 Salmon, C. T., & Oshagan, H. (1990). Community size, perceptions of majority opinion, and opinion expression.. In L. A. Grunig & J. E. Grunig (Eds.), Public relations research annual (vol. 2, pp. 157–171). Hillsdale, NJ: Erlbaum. Salwen, M. B., Lin, C., & Matera, F. R. (1994). Willingness to discuss “Official English”: A test of three communities. Journalism Quarterly, 71, 282–290. Scheufele, D. A., Shanahan, J., & Lee, E. (2001). Real talk — Manipulating the dependent variable in spiral of silence research. Communication Research, 28(3), 304–324. Shamir, J. (1995). Information cues and indicators of the climate of opinion — the spiral of silence theory in the Intifada. Communication Research, 22(1), 24–53. Shamir, J., & Shamir, M. (1993). The dynamics of Israeli public opinion on peace and the territories. Tel Aviv, Israel: Tel Aviv University. Slater, M. (2007). Reinforcing spirals: The mutual influence of media selectivity and media effects and their impact on individual behavior and social identity. Communication Theory, 17(3), 281–303. Snyder, L. B., Hamilton, M. A., Mitchell, E. W., Kiwanuka-Tondo, J., Fleming-Milici, F., & Proctor, D. (2004). A meta-analysis of the effect of mediated health communication campaigns on behavior change in the United States. Journal of Health Communication, 9, 71–96. doi: 10.1080/10810730490271548 Tokinoya, H. (1989). Testing the spiral of silence in East Asia. KEIO Communication Review, 10, 35–49. Wedel, T. (1994). The spiral of silence in popular culture: Applying a public opinion theory to radio station popularity (unpublished Master’s thesis). California State University, Fullerton. Willnat, L., Lee, W. P., & Detenber, B. H. (2002). Individual-level predictors of public outspokenness: A test of the spiral of silence theory in Singapore. International Journal of Public Opinion Research, 14(4), 391–412. Yun, G. W., & Park, S.-Y. (2011). Selective posting: Willingness to post a message online. Journal of Computer-Mediated Communication, 16(2), 201–227. doi: 10.1111/j.1083-6101.2010.01533.x

PART II

Media Fragmentation, Internet, and Public Opinion

This page intentionally left blank

7 NORMATIVE IMPLICATIONS OF THE SPIRAL OF SILENCE Hearing the Other Side in the 21st Century Diana C. Mutz and Laura Silver

This book serves as testimony to the unusually generative nature of the spiral of silence theory, which has spawned huge amounts of scholarship focused on the social dimension of opinion formation. Our focus in this chapter will be on furthering development of the normative implications of opinion climates. We do this by comparing some of the central ideas in The Spiral of Silence (NoelleNeumann, 1993) to those in Hearing the Other Side (Mutz, 2006). For those unfamiliar with the ideas in both books, a brief overview is in order. Both Hearing the Other Side and The Spiral of Silence focus on the importance of social influence in conditioning political attitudes and behaviors. In The Spiral of Silence, emphasis is on how mass mediated forms of social influence operate to suppress opinion expression and thus silence minority viewpoints, or at least those views perceived to be in the minority. When people find very little evidence of public support for their views, they become less likely to express those viewpoints, thus further encouraging silence among those of that viewpoint. Without contradicting these ideas, Hearing the Other Side (Mutz, 2006) adds an acknowledgement of the positive side of these same phenomena, arguing that for those who are surrounded by like-minded others, social influence actively facilitates and encourages political participation of many kinds. A supportive micro-level opinion climate both reassures people of the correctness of their own views, and lowers the social costs of actively participating in support of these views. If one is surrounded by like-minded others, then there is little chance of disturbing social harmony by openly espousing political views; after all, they are widely shared by others within one’s environment. Thus opinion climates have both positive and negative influences. At their core, both books are concerned with how opinion climates affect political attitudes and behaviors. The Spiral of Silence (Noelle-Neumann, 1993)

76

Diana C. Mutz and Laura Silver

indirectly acknowledges, yet also largely ignores, the facilitative role that perceptions of the opinion climate can play in encouraging mass political participation. Based on the underlying logic of the theory, those who are not suppressed into voluntary silence should have been spurred on to increased levels of speaking out and participating. Expressing one’s political views is, after all, another form of political participation. But the climate of opinion takes on a largely repressive role in The Spiral of Silence (Noelle-Neumann, 1993), whereas the positive effects of reinforcing opinions on participation receive greater emphasis in Hearing the Other Side (Mutz, 2006). In combination, what these two theories suggest is that social influence, whether from interpersonal networks or mediated impressions of mass opinion, is a double-edged sword, neither purely a force for good nor evil. Moreover, the two books in combination serve to remind us that both reinforcing and contradictory viewpoints have their place in a healthy democracy. In the sections that follow, we begin by updating several of the underlying premises of the spiral of silence given today’s media environment. We bring current evidence to bear on what kind of opinion climate is likely to predominate in face to face and mass media contexts, particularly in the United States. Most importantly, we also elaborate on the multitude of both positive and negative effects that may flow from people’s perceptions of a supportive or hostile opinion climate. Based on this discussion, we take a big picture approach and provide a theoretical discussion addressing the central unresolved question underlying not only research on the spiral of silence, but much of contemporary research on political communication: what kinds of media and interpersonal networks are best for democracies? I. Updating the Underlying Premises Some of the underlying premises of the spiral of silence strike today’s student as anachronistic at best, with respect to the media and politics of the contemporary United States. Others premises may not seem self-apparently wrong so much as in need of revision. The three premises we address in this section include: 1. Mass media are “consonant,” that is, they portray mass opinion as all of one cloth. 2. People’s perceptions of the opinion climate are often inaccurate, with those perceiving themselves in minority positions often underestimating the actual extent of support for their views. 3. Opinion climates affect people primarily by encouraging them to drop out of public debate. Media Consonance Media consonance has long been a controversial assertion in the literature on the spiral of silence. In the United States, there was a sense in which this assertion

Normative Implications of the Spiral of Silence

77

rang true for a previous generation of scholars. Political content then consisted largely of the TV evening news, and the three networks carried roughly the same content, with controversial issues portrayed in a homogeneous fashion (Scheufele, 2007). Evidence of inter-media agenda setting suggested that what was covered did not differ much from television to radio to newspapers (McCombs, 2004), and a small number of elite media tended to lead others in their news emphases (Fishman, 1978). Still, this is not quite the same as suggesting U.S. media is consonant in terms of a specific political leaning or bias, as the spiral of silence suggests. If anything, the media of this generation was maligned for focusing exclusively on middle-of-the road views and “indexing” the voices of elites in power (Bennett, 1990). The indexing hypothesis is, at its root, derivative of the spiral of silence. Indexing implies that were it not for the fact that certain viewpoints do not get much coverage, those viewpoints would have more adherents than they currently do. This is largely the same as saying that some views will be silenced because they are not well represented in the media. Today’s media environment would seem to alleviate some of these concerns. As reviewed extensively elsewhere (see, e.g., Prior, 2007), the number of media outlets has proliferated, and outlets with more clearly partisan points of view have been a part of this expansion. Few scholars today launch complaints about the restricted range of viewpoints available in U.S. political media; many more seem concerned that the media over-represent extreme political positions, encouraging ideological segregation (Sunstein, 2002a). A need for dramatic tension to hold viewer attention fuels journalists’ desire to include confl icting viewpoints, regardless of whether they are deemed reasonable or legitimate. Journalists are particularly likely to feature more unusual disputes, such as internal strife among party members, and these less predictable disagreements have disproportionate impact (Groeling, 2010). The rise of the Internet is similarly linked to a proliferation in the availability of political viewpoints. Lesser-known groups and less popular viewpoints are able to have a more prominent presence online (Lawrence, 2010; Papacharissi, 2004) because previous limitations on access—printing presses, TV stations and the like—are no longer such formidable obstacles. Moreover, the 24-hour news cycle produces much heavier demands for content of all kinds, thus making even highly unusual perspectives part of the ongoing supply. Overall, one might argue that potential spirals of silence seem less likely than they once did, because these days virtually any viewpoint can fi nd support somewhere in the U.S. news media. And indeed, scholars do seem less concerned about this potential mechanism than they once were, with one possible exception. Because of steep competition for relatively small television audiences, political programs tend to play up confl ict, by choosing as illustrative viewpoints those people from opposing extremes, who can then duke it out in an entertaining fashion. Concern about this practice has led some to question

78 Diana C. Mutz and Laura Silver

whether it is now the middle-of-the-road view that is essentially “left out” of public discussion (Fiorina, Abrams, & Pope, 2004). Thus people who are neither extremely conservative nor liberal may be the ones who feel silenced by this particular format. On the other hand, even with the proliferation of partisan sources, more mainstream sources are still present and widely used in today’s media environment. Accuracy of Perceptions Studies of perceived mass opinion, as well as those of perceived network opinion, suggest that perceptions of others’ political views are not always accurate. In both cases, evidence supports the idea that inaccuracies are most often in the direction of overestimating support for one’s own views (Fields & Schuman, 1976; Krueger & Clement, 1994; Mutz & Martin, 2001; Wojcieszak & Price, 2009). In terms of personal networks, not only are people already likely to have friends and family who are of like mind, but those whose views are ambiguous will be assumed to agree with them even when they do not. In the case of mediated representations of mass opinion, it is less clear whether people actively distort what they see in the news to interpret it as consistent with their own views (Bartels, 2002; Gerber & Green, 1999; Kunda, 1990; Taber & Lodge, 2006). But even if skewed perceptions do not encourage impressions of likeminded exposure via media, the increased opportunities for selective exposure in today’s media environment may lead toward the same outcome: impressions of a highly supportive opinion climate. Based on the logic of the spiral of silence, these distortions would appear to be highly functional; to the extent that social pressures operate to influence individual decisions about whether to speak out or participate politically, this perception of mass support should encourage participation rather than silencing it. On the other hand, if people do not recognize that some of their friends are of opposing political perspectives, then it will likewise be difficult for them to potentially benefit from this cross-cutting exposure. Both levels of political tolerance as well as willingness to consider others’ opinions when formulating their own views may be inhibited as a result. Our suggestion that people today have negligible reasons to fear social isolation in the climate of opinion surrounding them may strike some as inconsistent with the widely discussed “hostile media” phenomenon. While we agree that when asked about “the media” or “the press,” people tend to be overwhelmingly negative (Ladd, 2009), most Americans are quite positive when asked about the news sources that they personally use (see Goldman & Mutz, 2011). Partisan media tailored to individual viewpoints should, if anything, increase the extent to which people’s own sources are viewed as congenial.

Normative Implications of the Spiral of Silence

79

Responses to the Opinion Climate Both The Spiral of Silence (Noelle-Neumann, 1993) and Hearing the Other Side (Mutz, 2006) suggest that people respond in important ways to the climate of opinion surrounding them. This emphasis on politically relevant consequences of opinion climates is likely to be the most important and enduring contribution of the spiral of silence. And although speaking out publicly is certainly one important potential consequence of perceiving social support for opinions or a lack thereof, it is not the only possibility. We suggest four broad categories of politically relevant outcomes that might be affected by levels of perceived support for, or opposition to, one’s political views: (a) political participation, (b) preferences for a candidate or issue, (c) polarization of views and concomitant intolerance of oppositional perspectives, and (d) levels of social well-being outside the context of political controversy. In an effort to understand just how extensive and far-reaching the consequences of opinion climates may be, we briefly describe research relevant to these outcomes. Political Participation. Climates of opinion are likely to affect willingness to donate money to a political candidate or cause, to work on behalf of a candidate, to publicly display support through buttons and bumper stickers, and so on. Although studies of the spiral have involved some forms of participation as outcome measures, including petition-signing or wearing campaign buttons, we suggest that the emphasis within the spiral of silence on public expression has, in some ways, limited the expansion of this theory. Mobilizing and demobilizing effects from media use have been hypothesized many times in the past, with a variety of different theories in mind, and inconsistent fi ndings (Hofstetter, 1998; Nir & Druckman, 2008). Some theories proposed withdrawal from participation as a logical consequence of the negativity in political media (Norris, 2003; Robinson, 1976), whereas others suggested that media provided citizens with political information that would lead to greater efficacy and participation (Kenski & Stroud, 2006). Still others argued that media encouraged participation only when it explicitly included “mobilizing information,” that is, specifics on how to get involved, where to attend a meeting, whom to contact, and so forth (Lemert & Ashman, 1983; Lemert, Mitzman, Seither, Cook, & Hackett, 1977). Studies of social networks and interpersonal communication have produced far more evidence bearing on the hypothesis that supportive interpersonal opinion climates lead to participation, with fairly consistent results. Since The People’s Choice (Lazarsfeld, Berelson, & Gaudet, 1948), and early Columbia studies (Berelson, Lazarsfeld, & McPhee, 1954), scholars have argued that people with like-minded political discussion networks were more likely to make up their minds early with respect to vote choice, to show interest in the campaign, and to participate in public ways (for more recent evidence in this vein, see Huckfeldt, Mendez, & Osborn, 2004; Mutz, 2002, 2006). While most of

80 Diana C. Mutz and Laura Silver

this is cross-sectional evidence, a multi-wave panel study from the 2008 presidential election provides evidence that partisan news exposure can alter levels of political participation. Using extensive information on the specific programs that individuals viewed, Dilliplane (2011) found substantial support for the idea that exposure to like-minded partisan news increased levels of campaign activity and encouraged an early vote decision. Unlike most previous research, this study demonstrates that increases in partisan news consumption go hand in hand with increases in these forms of campaign participation, over and above the general trend toward greater participation that naturally occurs during the course of a campaign. Moreover, exposure to confl icting news had precisely the opposite effect, dampening the prospects for participation. Thus partisan news has the potential to both mobilize and demobilize depending upon whether it is like-minded or hostile to the individuals’ own preferences. Beyond partisan media influence on willingness to speak out to support specific political views, the spiral of silence is also potentially applicable to a wide variety of consensually valued or consensually undesirable behaviors. For example, a burgeoning body of research has demonstrated the importance of social pressure for inducing one type of socially desirable behavior: voting. A field experiment by Gerber, Green, and Larimer (2008) demonstrated that making it clear to an individual that the neighbors would know whether s/ he voted resulted in substantially increased turnout (see also Davenport, 2010; Mann, 2010; Panagopoulos, 2010). Although social pressure in this case was not due to a desire to do what the majority was doing, it is the same basic mechanism as the spiral of silence: embarrassment and social isolation. Other studies in this vein also demonstrated that the influence of shame and social approbation was much greater than the influence of social approval (Gerber, Green, & Larimer, 2010). Political Preferences. The study of social influence on political preferences is perhaps the oldest area of research tied to the spiral of silence, with its roots in studies of face-to-face conformity pressures. The theory predicts that perceptions of an opinion climate lead to changes in individuals’ opinions and beliefs as a result of either normative or informational social influence. The literature addressing this hypothesis is sufficiently large that we will not attempt to review it here except to say that to the extent that most communication environments—whether mass or interpersonal—are supportive and likeminded, social influence pressures should not serve to change opinions very often. Indeed, the tenor of today’s social criticism is precisely a concern about lack of opinion change, even when circumstances might call for it. Instead of changing political preferences, most contemporary scholars studying the United States predict that perceived opinion climates should, if anything, strengthen existing opinions as discussed in greater depth below.

Normative Implications of the Spiral of Silence

81

Political Polarization and Intolerance. Given that interpersonal discussant networks are known to be highly homogeneous (Blakely & Snyder, 1999; Huckfeldt & Sprague, 1987; Mutz, 2006), and that media are thought to be increasingly reflective of their viewers’ and readers’ political preferences, crosscutting exposure could be increasingly unlikely. Under these conditions, polarization is the most likely outcome of social influence on aggregate opinions. Just as small group deliberations can produce more polarized views when there is no incentive to arrive at an agreement, portrayals of public opinion may have similar effects when conveyed by media. Within the United States, the most striking political change in recent decades has been the well-documented increase in the extremity of partisan views among political elites (Abramowitz & Saunders, 2008; Fleisher & Bond, 2004; Hetherington, 2001). There is considerably less consensus about the state of mass public opinion, however. The level of ideological consistency of those who identify with a political party has increased (Levendusky, 2009), with Republicans more predictably conservative now, and Democrats more consistently liberal. Yet it is not clear that people’s opinions are any more extreme than they ever were (see, e.g., Fiorina & Abrams, 2008). Nonetheless, debate over the extent of mass polarization in the United States has become a popular topic in both academic circles (Abramowitz, 2010; Fiorina, Abrams, & Pope, 2004) and in popular discourse (Brownstein, 2007). For those who see mass polarization, the concurrent rise of partisan media seems a likely cause (Iyengar & Hahn, 2009; Iyengar, Hahn, Krosnick, & Walker, 2008). Drawing on the spiral of silence, the underlying argument would be that social influence (operating through mass media portrayals of like-minded others) increases the extremity and certainty with which opinions are held. While some studies suggest that partisan sources can impact vote choice (Dalton, Beck, & Huckfeldt, 1998; DellaVigna & Kaplan, 2007; Dilliplane, 2011; Ladd & Lenz, 2009), empirical studies have not yet convincingly connected media exposure to increases in political polarization (see Stroud, 2010; Levendusky, 2013). The puzzling part of this state of affairs is that the media seem to be better sources of cross-cutting exposure than interpersonal networks. Although strict comparisons are problematic, Gentzkow and Shapiro (2011) suggest that even today, U.S. media expose their audiences to more cross-cutting perspectives than do their interpersonal networks. The potentially negative externalities of polarization naturally lead to consideration of what Hearing the Other Side (Mutz, 2006) promotes as positive consequences of perceiving support for the opposition: increased tolerance, respect for oppositional views, and the potential for compromise. To the extent that cross-cutting exposure accomplishes these ends, a “hostile” climate of opinion would seem to be a very beneficial thing indeed, conveying information critical to understanding the other side’s positions. Additionally, those who perceive themselves to be in the minority in one arena may be less likely to advocate trampling on the rights of others when they fi nd themselves in the majority.

82 Diana C. Mutz and Laura Silver

Even in competitive situations such as elections, cross-cutting exposure is valuable for establishing the legitimacy of the outcome. If citizens end up on the losing side of an election and do not hear the opinions of the many others who supported the winning candidate, then they will be less likely to view the election outcome as legitimate. Apolitical Social Well-Being. The impact of social influence on explicitly political behavior has consumed the bulk of attention in political communication research to date, but the relationship between political and nonpolitical life is also important to the overall quality of life citizens enjoy. It would be naïve to assume that differences of opinion, whether real or perceived, have no effect on personal relationships and life outside the political realm. Both The Spiral of Silence (Noelle-Neumann, 1993) and Hearing the Other Side (Mutz, 2006) are sensitive to the human side of confl ict; that is, they suggest that people will go to some lengths to avoid the unpleasantness of confrontation. Research outside the context of elections and public opinion suggests that diversity of all kinds can make people uncomfortable. At the interpersonal level, this idea has been studied most extensively in the context of social trust. As Putnam (2007) has noted, in ethnically diverse neighborhoods, “trust (even of one’s own race) is lower, altruism and community cooperation rarer, friends fewer” (p. 137). Greater ethnic diversity is clearly linked to lower social capital (see Costa & Kahn, 2003, for a review), but diversity of all kinds poses challenges to social solidarity and a sense of community. Conversely, people in homogeneous communities tend to be more satisfied and trusting. Although the logic of this state of affairs seems intuitive—most people fi nd life easier when surrounded by supportive others than by those who challenge their positions, political or otherwise—it is an uncomfortable truth at best. Confl ict also influences how people feel about their system of government. While academics are quick to point out the importance of highly competitive elections within political systems, individual citizens may not appreciate competition nearly as much as social scientists. For example, Brunell (2008) fi nds that voters in homogeneous districts are far more satisfied with government and politics than voters in highly competitive, heterogeneous districts. One obvious explanation for this is that most people in homogeneous districts get the political representation that they want; they are electoral “winners” more often than not. They enjoy solidarity with many of their friends and neighbors as well. What’s not to like? In highly competitive congressional districts, by contrast, a large number of voters may be bitterly disappointed with the outcome of the election, and they will not necessarily be near like-minded others who can commiserate with them in their loss. Indeed, in a diverse unbalkanized environment, they cannot even be sure which of their neighbors is on “their side.” One can see why a spiraling process in a district highly slanted toward one party might be both inevitable and rational for those in the minority. After all, if there’s really no hope of winning, what is the purpose of participation?

Normative Implications of the Spiral of Silence

83

Our larger point in reviewing these positive and negative consequences is to make it clear that the level of social support for one’s views—whether conveyed by media or by interpersonal networks—is neither purely an evil force operating to silence minority views, nor strictly a positive influence, spurring on those in the majority to greater participation. Instead, people’s perceptions of the extent to which they are supported or challenged by the opinion climate have a multitude of consequences that defy such simple characterizations. A supportive opinion climate can produce increased levels of participation and a greater likelihood of speaking out. But it can also give people the impression that their view is the only reasonable point of view on a given controversy. Thus perceiving a high level of reinforcement can result in polarization, intolerance of oppositional views, and even perceiving political confl ict itself as illegitimate. On the other hand, in addition to silencing oppositional views, perceiving one’s view to lack broad-based support can lead to decreased participation and a sense of political and social isolation. The fl ip side of that same coin is that greater tolerance for the opposition, moderation of extreme views, and an increased willingness to compromise—all valuable assets in a healthy democracy—are furthered by exposure to oppositional viewpoints. II. What Kind of Media Environment Is Best for Democracies? Media are central to the spiral of silence theory because they are thought to be the main source of perceptions of the climate of opinion. If we begin from the end goal that is implicit in the spiral of silence—equal comfort in expressing opinions, regardless of one’s political views—then what should the distribution of public opinion illustrated by the media look like? We will, for the sake of the current discussion, ignore the fact that not everyone is equally likely to see all opinions in the media environment, and assume each individual has equal access and motivation to see all available media. What then are the implications of different types of media environments for the likelihood of experiencing reinforcing and cross-cutting viewpoints? In Figure 7.1 we describe four ideal-type media environments, none of which should be interpreted as descriptive of any existing country’s current system. The hypothetical distributions shown in the figures represent the distribution of issue opinions made available by a given media system, with the vertical line representing the average opinion (though not necessarily the modal one) held by the public. Importantly, the news hole, that is the amount of space or time that exists to be fi lled, is assumed to be constant (the area under the line), and all that changes between graphs is how media time/space is used when illustrating the range of opinions. Consider each of the figures below in light of two criteria. First, what is the likelihood that media would squelch participation and systematically silence some viewpoints? Second, what is the likelihood that a given media environment would promote intolerance and extremism? To complicate things further,

y

y

x

x

D

B

y

y

x

x

FIGURE 7.1 Four Ideal Type Media Environments. Plots represent hypothetical distributions of issue opinions that are represented in the media across the x-axis, with the average opinion among the mass public indicated by the vertical line. The vertical axis represents the amount of coverage received by opinions on the horizontal axis.

C

A

Normative Implications of the Spiral of Silence

85

consider these questions in light of what is known about the media use habits and strength of opinions among people with more or less extreme views. Figure 7.1A describes a situation in which middle of the road or “mainstream” views are most widely available in the media, and more extreme views follow a pattern of progressively lower levels of availability—perhaps reminiscent for some of the pre-cable U.S. broadcast media. If one assumes that public opinions are also normally distributed, then this pattern of viewpoints perfectly reflects the distribution of mass opinion. Given the media emphasis on moderate viewpoints, this case is unlikely to promote extremism and intolerance, though those holding extreme political views are likely to perceive less support for their positions and perhaps be less likely to act on them as a result. Countering this problem is the fact that extreme issue opinions tend to be held with greater strength and passion than more moderate views (like Noelle-Neumann’s “hard core”), perhaps making this scenario less problematic than many others. In Figure 7.1B, we portray the availability of viewpoints in a media system biased toward either one side or the other of the political spectrum. This model, which seems to be what Noelle-Neumann had in mind when formulating her theory, offers either overwhelmingly liberal or overwhelmingly conservative viewpoints. As a result, the holder of an average opinion, and especially one with an opinion on the opposite side of the political spectrum, will not receive much reinforcement or evidence of shared viewpoints, and may feel isolated and unsupported. Even those in the middle of the road will underestimate support for their views. On the contrary, those at the same ideological extreme emphasized by media coverage will be emboldened and encouraged toward greater involvement. In Figure 7.1C, we illustrate a partisan duopoly, that is, a situation in which multiple media organizations with different partisan leanings concurrently make views on the left and right roughly equally available, with relatively little coverage of moderate views. This distribution appears to be what theorists like John Stuart Mill had in mind when advocating a marketplace of ideas. Advocates for each side make their views known, thus providing social reinforcement to those on both sides of the political spectrum for purposes of encouraging participation. However, if the views under the two nodes in this distribution came from non-overlapping media sources, such that any given individual was likely to be exposed to only one of the two, then the prospects for intolerance would increase. Finally, in Figure 7.1D, we illustrate a rectilinear model that satisfies the criterion of providing all possible viewpoints with equal amounts of social reinforcement. This model best satisfies the requirement of offering perfect competition in the marketplace of ideas, because all possible viewpoints are equally available and reinforced by media coverage. In many ways, Figure 7.1D seems like the ideal of fairness in media. But given a fi nite news hole, the extent

86 Diana C. Mutz and Laura Silver

to which any one part of the opinion spectrum receives reinforcing exposure is necessarily much lower (consider the dilemma of covering 10 presidential primary candidates equally and thus saying much less about each). Relative to some of the other models, there is less social reinforcement for most positions, but greater reinforcement of extremes relative to their absolute numbers. What are the implications of these varying distributions of opinions presented in the media environment? The answer to this question is neither simple nor straightforward. It requires taking into account and ultimately balancing the needs for enthusiastic political participation and a tolerant public sphere. Our own approach to this task begins with an observation that has withstood the test of time: interpersonal networks tend to be highly politically homogeneous. This was true at the dawn of empirical election research, and, given the large role that choice plays in who we spend time with as our friends and associates, we have little reason to think that this pattern is likely to change. For this reason, we suggest that perhaps interpersonal associations are, by default, the best means of providing the social reinforcement that citizens need to participate in the broader political world. Enclave communication, that is, interactions among those within homogeneous networks, is particularly important for translating minority positions into political action (examples include abolitionism and the gay rights movement, see Sunstein, 2000, 2002b). Given that it appears to be human nature that people are drawn to those similar to themselves on some dimension, normative theory may fi nd it easier not to fi ght this tide, but rather to harness it for the ends it appears to serve well already. If we relegate interpersonal contact to the reinforcement function, then mass media must, by default, serve the purpose of facilitating cross-cutting exposure. As indicated by the research we have reviewed, even in today’s fragmented mass media environment, most political media still provide a range of political perspectives, far more so than interpersonal networks. And even if they are not evenhanded in how they portray the other side, the impetus for news shows to focus on dramatic confl ict is often sufficient to motivate inclusion of some oppositional views. Partisans may choose reinforcing media, but because of their high levels of political exposure they also experience exposure to many mainstream ideas and positions. So although we concur that the structure of current media may not be ideally suited for the purpose of providing crosscutting exposure, it is, for the moment, still the best, most likely source around. Moreover, because media more easily transcend geography and its accompanying homogeneity, it is better suited to this task than many other forms of communication. If we were to pick one of the four media systems in Figure 7.1 for purposes of maximizing overall exposure to cross-cutting views (assuming that people have little problem fi nding support and reinforcement for their opinions in other venues), it probably would be Figure 7.1C, The Partisan Media Duopoly, or Figure 7.1D, The Perfect Competition Polyopoly. In these two cases we see

Normative Implications of the Spiral of Silence

87

the greatest availability of cross-cutting views for the largest portion of the population. The duopoly also enhances the prospects for selective exposure if the viewpoints on one side of the spectrum only come from certain sources, and opposing viewpoints from other sources. But that need not necessarily be the case. Current estimates suggest that only around 2% of Americans expose themselves to strictly like-minded programming (Dilliplane, 2012; Mutz & Young, 2011). Figure 7.1A, The Mainstream Media Monopoly, also may do an acceptable job of cross-cutting exposure, but it will provide a far narrower range of cross-cutting views to the public. Of course, the observations we offer above are predicated on a roughly normal distribution of mass opinion, which may not always be accurate. Normative proscriptions would probably be different if opinions among the mass public were heavily skewed. For example, if 95% of the public was against the death penalty, should the opinions covered by the media still include large percentages of oppositional viewpoints? Should the ideal distribution of viewpoints in the media mirror the distribution of opinion among the public, or make all political views equally accessible even if it means under-representing some and over-representing others? And do citizens need to be exposed to views that differ a great deal from their own in order to recognize benefits, or is only a minor difference of opinion still beneficial? Are many different viewpoints necessary in order to properly appreciate a diversity of opinion, or are only a few sufficient? Our larger purpose in exploring the examples in Figure 7.1 is to point out that we have little if any agreement in the field of political communication on what kind of climate of opinion the media should ideally portray. As a result, our conclusions about whether a given fi nding is normatively desirable are all over the map. Is it media’s role to promote participation? To discourage extremism? To foster tolerance or the spirit of compromise? Most admonitions about what media should or should not do to maximize healthy democratic functioning concern the kinds of information that need to be conveyed to the mass public in order to ensure accountability; they seldom have much to say about the kinds or amounts of opinions to which the public should be exposed. Instead, theory in this area has been limited to concerns about bias in media sources, with the underlying concern focused on the potential for persuasion due to an imbalance in the opinions presented. Conclusion This chapter should make it clear that there is more at stake in the spiral of silence than persuasion alone. The opinions that are (or are not) presented by media are directly relevant to the vibrancy of debate in a public sphere. There is a tendency to study representations of opinion in the media strictly when the concern is about persuading the public, and to study information in the media

88 Diana C. Mutz and Laura Silver

when the focus is on the quality of political debate or the state of the public sphere more generally. But the range of issue opinions offered is surely a meaningful indicator of the quality of the public sphere, even if no one is persuaded by those opinions. Whether and how people are influenced by their perceptions of the distribution of mass opinion remains virgin research territory. We see exciting unexplored prospects for empirical research in this area. For example, are individuals sensitive to the actual distribution of opinions or purely the majority versus minority? Does the median view matter at all if it is not attractive to large numbers of people? And do people judge distributions of issue opinions in media coverage differently based on their preexisting assumptions about the distribution of mass opinion? In addition to suggesting new avenues for research, our chapter is also intended to promote broadening the realm of research to which the spiral of silence is applicable. Toward that end, it is essential that the spiral of silence relinquish the notion that social influence is necessarily a silencing force. Social pressure is neither inherently a force for normatively negative nor for normatively positive outcomes, and both kinds of consequences are worthy of study. Additionally, if the theory is to continue to enjoy widespread research attention, it must be applied to a wider range of outcomes important to democratic political processes—not just speaking out. Finally, we urge scholars to elaborate upon how micro-climates should work in a contemporary communication environment. In an era in which people both hear about others’ opinions and behaviors from many different potential sources, and also have great control over their face-to-face social environments and interpersonal networks, how will these forces interact? As mass and interpersonal communication become increasingly melded via the use of online interpersonal networks for distributing online mediated information, will these two climates of opinion become perceived as increasingly similar? The spiral of silence continues to have much to offer in informing important research questions in the 21st century. References Abramowitz, A. (2010). The disappearing center: engaged citizens, polarization, and American democracy. New Haven, CT: Yale University Press. Abramowitz, A. I., & Saunders, K. L. (2008). Is polarization a myth? The Journal of Politics, 70(2), 542–555. Bartels, L. M. (2002). Beyond the running tally: Partisan bias in political perceptions. Political Behavior, 24, 117–150. Bennett, W. L. (1990). Toward a theory of press‐state relations in the United States. Journal of Communication, 40(2), 103–127. Berelson, B. R., Lazarsfeld, P. F., & McPhee, W. N. (1954). Voting: A study of opinion formation in a presidential campaign. Chicago, IL: University Of Chicago Press. Blakely, E. J., & Snyder, M. G. (1999). Fortress America: Gated communities in the United States. Washington, DC: Brookings Institution Press.

Normative Implications of the Spiral of Silence

89

Brownstein, R. (2007). The second civil war: How extreme partisanship has paralyzed Washington and polarized America. New York, NY: Penguin Press. Brunell, T. (2008). Redistricting and representation: Why competitive elections are bad for America. New York, NY: Routledge. Costa, D., & Kahn, M. (2003). Understanding the American decline in social capital, 1952–1998. Kyklos, 56(1), 17–46. Dalton, R. J., Beck, P. A., & Huckfeldt, R. (1998). Partisan cues and the media: Information flows in the 1992 presidential election. The American Political Science Review, 92(1), 111–126. Davenport, T. C. (2010). Public accountability and political participation: Effects of face-to-Face feedback intervention on voter turnout of public housing residents. Political Behavior, 32 (September), 337–368. DellaVigna, S., & Kaplan, E. (2007). The Fox News effect: Media bias and voting. The Quarterly Journal of Economics, 122(3), 1187–1234. Dilliplane, S. (2011). All the news you want to hear: The impact of partisan news exposure on political participation. Public Opinion Quarterly, 75(2), 287–316. Dilliplane, S. (2012). The impact of partisan news exposure on American political opinion and behavior: Evidence from the 2008 presidential election (Unpublished doctoral dissertation). University of Pennsylvania, Philadelphia. Fields, J. M., & Schuman, H. (1976). Public beliefs about the beliefs of the public. Public Opinion Quarterly, 40(4), 427–448. Fiorina, M. P., & Abrams, S. J. (2008). Political polarization in the American public. Annual Review of Political Science, 11, 563–588. Fiorina, M. P., Abrams, S. J., & Pope, J. C. (2004). Culture war? The myth of a polarized America. New York, NY: Longman. Fishman, M. (1978). Crime waves as ideology. Social Problems, 25(5), 531–543. Fleisher, R., & Bond, J. (2004). The shrinking middle in the US Congress. British Journal of Political Science, 34, 429–451. Gentzkow, M., & Shapiro, J. M. (2011). Ideological segregation online and offl ine. The Quarterly Journal of Economics, 126(4), 1799–1839. Gerber, A., & Green, D. (1999). Misperceptions about perceptual bias. Annual Review of Political Science, 2, 189–210. Gerber, A. S., Green, D. P., & Larimer, C. W. (2008). Social pressure and voter turnout: Evidence from a large-scale field experiment. American Political Science Review, 102(1), 33–48. Gerber, A., Green, D., & Larimer, C. (2010). An experiment testing the relative effectiveness of encouraging voter participation by inducing feelings of pride or shame. Political Behavior, 32, 409–422. Goldman, S. K., & Mutz, D. C. (2011). The friendly media phenomenon: A cross-national analysis of cross-cutting exposure. Political Communication, 28(1), 42–66. Groeling, T. (2010). When politicians attack: Party cohesion in the media. Cambridge, England: Cambridge University Press. Hetherington, M. J. (2001). Resurgent mass partisanship: The role of elite polarization. American Political Science Review, 95(3), 619–631. Hofstetter, C. R. (1998). Political talk radio, situational involvement, and political mobilization. Social Science Quarterly, 79(2), 273–286. Huckfeldt, R., Mendez, J. M., & Osborn, T. (2004). Disagreement, ambivalence, and engagement: The political consequences of heterogeneous networks. Political Psychology, 25(1), 65–95. Huckfeldt, R., & Sprague, J. (1987). Networks in context: The social flow of political information. The American Political Science Review, 81(4), 1197–1216. Iyengar, S., & Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59(1), 19–39. Iyengar, S., Hahn, K. S., Krosnick, J. A., & Walker, J. (2008). Selective exposure to campaign communication: The role of anticipated agreement and issue public membership. The Journal of Politics, 70(1), 186–200.

90

Diana C. Mutz and Laura Silver

Kenski, K., & Stroud, N. J. (2006). Connections between Internet use and political efficacy, knowledge, and participation. Journal of Broadcasting Electronic Media, 50(2), 173–192. Krueger, J., & Clement, R. W. (1994). The truly false consensus effect: An ineradicable and egocentric bias in social perception. Journal of Personality and Social Psychology, 67(4), 596–610. Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498. Ladd, J. M. (2009). The neglected power of elite opinion leadership to produce antipathy toward the news media: Evidence from a survey experiment. Political Behavior, 32(1), 29–50. Ladd, J. M., & Lenz, G. S. (2009). Exploiting a rare communication shift to document the persuasive power of the news media. American Journal of Political Science, 53(2), 394–410. Lawrence, E. (2010). Self-segregation or deliberation? Blog readership, participation, and polarization in American politics. Perspectives on Politics, 8(1), 141–157. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1948). The people’s choice: How the voter makes up his mind in a presidential campaign. New York, NY: Columbia University Press. Lemert, J. B., & Ashman, M. G. (1983). Extent of mobilizing information in opinion and news magazines. Journalism Quarterly, 60(Winter), 657–662. Lemert, J. B., Mitzman, B. N., Seither, M. A., Cook, R., & Hackett, R. (1977). Journalists and mobilizing information. Journalism Quarterly, 54, 721–726. Levendusky, M. (2009). The partisan sort: How liberals became Democrats and conservatives became Republicans. Chicago, IL: University of Chicago Press. Levendusky, M. (2013). How partisan media polarize America. Chicago, IL: University of Chicago Press. Mann, C. B. (2010). Is there backlash to social pressure? A large-scale field experiment on voter mobilization. Political Behavior, 32, 387–407. McCombs, M. (2004). Setting the agenda: The mass media and public opinion. Cambridge, England: Polity. Mutz, D. C. (2002). Cross-cutting social networks: Testing democratic theory in practice. American Political Science Review, 96(1), 111–126. Mutz, D. C. (2006). Hearing the other side: Deliberative versus participatory democracy. Cambridge, England: Cambridge University Press. Mutz, D., & Martin, P. (2001). Facilitating communication across lines of political difference: The role of mass media. American Political Science Review, 95(1), 97–114. Mutz, D., & Young, L. (2011). Communication and public opinion: Plus ᠒a change? Public Opinion Quarterly, 75(5), 1018–1044. Nir, L., & Druckman, J. N. (2008). Campaign mixed-message flows and timing of vote decision. International Journal of Public Opinion Research, 20(3), 326–346. Noelle-Neumann, E. (1993). The spiral of silence: Public opinion—our social skin. Chicago, IL: University of Chicago Press. Norris, P. (2003). A virtuous circle: Reinventing political activism. New York, NY: Cambridge University Press. Panagopoulos, C. (2010). Affect, social pressure and prosocial motivation: Field experimental evidence of the mobilizing effects of pride, shame, and publicizing voting behavior. Political Behavior, 32, 369–386. Papacharissi, Z. (2004). Democracy online: Civility, politeness, and the democratic potential of online political discussion groups. New Media & Society, 6(2), 259–283. Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. New York, NY: Cambridge University Press. Putnam, R. D. (2007). E pluribus unum: Diversity and community in the twenty‐fi rst century. Scandinavian Political Studies, 30(2), 137–174. Robinson, M. J. (1976). Public aff airs television and the growth of political malaise: The case of “the selling of the Pentagon.” The American Political Science Review, 70(2), 409–432. Scheufele, D. A. (2007). Spiral of silence theory. In W. Donsbach & M. W. Traugott (Eds.), Handbook of public opinion research (pp. 173–183). Thousand Oaks, CA: Sage. Stroud, N. J. (2010). Polarization and partisan selective exposure. Journal of Communication, 60(3), 556–576.

Normative Implications of the Spiral of Silence

91

Sunstein, C. R. (2000). Deliberative trouble? Why groups go to extremes. Yale Law Journal, 110 (October), 71–119. Sunstein, C. R. (2002a). Republic.com. Princeton, NJ: Princeton University Press. Sunstein, C. R. (2002b). The law of group polarization. Journal of Political Philosophy, 10(2), 175–195. Taber, C. S., & Lodge, M. (2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50(3), 755–769. Wojcieszak, M., & Price, V. (2009). What underlies the false consensus effect? How personal opinion and disagreement affect perception of public opinion. International Journal of Public Opinion Research, 21(1), 25–46.

8 MEDIA AND PUBLIC OPINION IN A FRAGMENTED SOCIETY Patricia Moy and Muzammil M. Hussain

“Newscasters and politicians were right in the living room of many Americans in the 1970s on a routine basis, but they left at seven o’clock. For Americans in 2005, they stand by at every hour of the day, ready to drop a mountain of information at the click of a mouse or the push of a remote” (Prior, 2007, pp. 2–3). When Elisabeth Noelle-Neumann conceived of and developed her spiral of silence theory in the 1960s and early 1970s, she formulated it in an era when scholars perceived a return to all-powerful media—namely, that the media were able to exert strong effects because audience members actively turned to newspapers, television, radio, and magazines to help defi ne social reality. Indeed, the spiral of silence emerged as a theory not only of public opinion, but also of media effects. With their ubiquity, consonance, and cumulativeness, the mass media of decades ago allowed individuals to gauge the climate of public opinion and speak out or not, depending on whether they perceived themselves to be in the minority or majority opinion. In this process of effects, the media served what Noelle-Neumann (1993) termed an “articulation function,” providing audience members with arguments used to back up their opinions. After all, “if the mass media fail to provide them, there will be no words” (p. 172). However, it is a truism to note that the relatively centralized media map of the 1970s bears little resemblance to what exists today. Prior’s (2007) description above of the new-millennium terrain precedes the boom of social media in 2006, the decentralization of broadcast television networks, and the fragmentation of audiences. The political information environment in the 1970s was mostly dominated by organized mass media, including newspapers, broadcast television, and radio—and aptly described as a media landscape. In contrast, the political information environment of the 2000s is marked by a combination of expert-curated information and mass amateurization (e.g., citizen

Media and Public Opinion in a Fragmented Society

93

bloggers)—characterized as a “new media ecosystem” (Rainie, 2010), one that is highly interactive and evolving. Today’s highly fragmented society is characterized, to varying degrees within the United States and other advanced democratic states, by a host of cable television outlets, an infi nite number of online sources, talk radio shows, and partisan media. This proliferation of outlets has emerged with respect to both the national and local media in numerous countries (see Lang, 2004, for an early review) and has had great implications for changes in the reporting of news. With this increase in the number of non-institutional media outlets, some institutional outlets fought to keep and maintain their audiences by turning to reporting “softer” news (Ladd, 2012). Of course, discussions of media institutions and outlets often get confounded as “sooner or later it will all be on the internet,” as Neuman, Bimber, and Hindman (2011, p. 23) proclaimed, and citizens change how they seek information. For instance, mobile technologies, personalized news streams, and exposure to myriad online environments all offer citizens a vast array of opportunities to engage with public information. The days of yore, in which Americans received most of their news from newspapers or television, have transitioned to a period when individuals are turning to the Internet for news (Pew Research Center, 2008). This new media terrain—with countless outlets, a heavy reliance on technology, and greatly popularized content—also carries implications for democratic engagement. As Stroud (2011) notes, citizens’ use of “niche news” can profoundly shape what they know about their social world, how they feel about issues, and potentially mobilize them to action. To what extent does this fragmented society force a revisiting of how the spiral of silence operates? This chapter examines broadly how two key assumptions of Noelle-Neumann’s theory—individuals’ fear of isolation from society and their scanning their environment to gauge the opinion climate—hold up in this new age. Fear of Isolation from Society Noelle-Neumann’s (1993) view of public opinion situates it as a tool of social control. As such, the two key assumptions undergirding the spiral of silence include one top-down phenomenon—society threatening deviant individuals with isolation—and its bottom-up counterpart—individuals fearing isolation from society. According to the theory, these two work together to influence the public expression of opinions. Specifically, in any social collective, only a sufficient level of agreement on values and goals can ensure cohesion, or public opinion (Noelle-Neumann, 1991). Consequently, discussions of public opinion in the spiral of silence literature typically revolve around “opinions on controversial issues that one can express in public without isolating oneself ” (NoelleNeumann, 1993, pp. 62–63). Early critics of the theory have questioned the extent to which it is society at

94

Patricia Moy and Muzammil M. Hussain

large from which individuals fear isolation. After all, studies in social psychology (e.g., Newcomb, 1948) as well as the spiral of silence (e.g., Glynn & Park, 1997; Moy, Domke, & Stamm, 2001) have shown that individuals’ reference groups wield great power in shaping attitudes and behaviors. Technological advancements, however, have decimated the notion of a mass society, much less one bound by location. As the Internet allows individuals to interact across geographically disparate spaces, citizens are no longer relegated to membership in reference groups that reside in their neighborhood or at their place of business. Indeed, some scholars show that in the United States, the organization of individuals into groups has evolved in tandem with the evolution of media systems: discussion networks in 2004 were significantly smaller and more fragmented when compared to 1985 (McPherson, SmithLovin, & Brashears, 2006). In addition, more people today seek to associate with others like themselves, and while societal networks are now more racially heterogeneous, they tend to remain very homogenous in terms of general socioeconomic variables like education (Paxton, 2007). These trends suggest a contemporary social organization that reflects “neo-tribalism,” whereby social organizing has moved beyond the mass society metaphor of the 20th century to one where individuals are increasingly part of “little masses” of heterogeneous fragments organized around affect-based collectives (Rokka, 2010, p. 382). These new tribes are not constrained by the longevity or physical boundaries of more traditional groups, and their members belong to overlapping groups (Rokka, 2010; see also Cova & Cova, 2002). Naturally, digital media are prime locations for the cultivation of these new groups and their expression of opinion (Robards & Bennett, 2011). These new patterns of aggregation have significant implications for how fear of isolation and the spiral of silence theory may function in today’s society. On the one hand, there is cautious optimism for a reduced role of fear of isolation in determining whether people speak out. As Wang and Wellman (2010) discuss in their study of social connectivity, the Web allows individuals to forge new relationships, remain in touch with existing friends, and meet newer and different individuals whom they might not normally do so in offl ine spheres. Applied to the spiral of silence, a preponderance of group memberships might mean individuals do not worry about being ostracized from a particular group as they have other communities in which to engage. In other words, they might feel free to speak their minds in a certain group at the risk of isolation if they know other groups (or microcosms of their public) might share their views. As well, the anonymity that Web-based communication affords facilitates the expression of views, regardless of whether they concur with the majority view or are in the minority. On the other hand, collectives that are more homogeneous may tend to be less tolerant of deviance and therefore exert greater pressure on their members and induce a greater fear of isolation. Inherent homogeneity notwithstanding, interactive aspects of new media technologies tend to

Media and Public Opinion in a Fragmented Society

95

complement, not replace, offl ine relationships, and therefore can strengthen community and social capital (Pruijt, 2002). This would argue for a stronger role of fear of isolation in the expression of one’s views as group members find themselves in a more tight-knit collective, one driven by a concern or issue. The facilitation of communication and opinion expression by media technologies thus is a dual-edged sword at the societal level. Individuals in selfselected and like-minded collectives today might be more willing to speak out in “safe discussions” (Eveland & Hively, 2009) and feel politically efficacious, drawing upon their sense of strength in numbers. However, while discourse and deliberation may thrive—albeit in a less than fully diverse community of thought—when those in homogeneous communities realize the existence of other, differing communities, cleavages around an issue become more prominent, factions become more polarized, extreme actions are taken to gain support for one’s views, and enclaves form that do not speak to other groups. Certainly Papacharissi (2002, pp. 16–17) was prescient in warning that political expression online might leave citizens with a false sense of empowerment, citing how the reach of virtual opinions may be more diverse, but may also decrease as Internet fragmentation continues. Scanning the Climate of Opinion According to the spiral of silence theory, fear of isolation drives individuals to gauge the climate of opinion so that they can behave appropriately, express “proper” opinions, and avoid sanctions. The mass media play a critical role in how individuals come to understand the world around them, particularly as they are exposed to the concerns, attitudes, and actions of others outside their immediate life space (Mutz, 1998). In the mid-20th century, the mass media landscape generally was characterized by uniformity of content (Key, 1961), a consonance that meant audience members typically received relatively similar views on issues and events. Today’s mass media stand in marked contrast to their counterparts of a half-century ago. With increased political partisanship, the balkanization of mass media has created outlets for citizens of all stripes. In the United States, ideologically coherent, conservative opinion media (e.g., Fox News, talk radio shows, and the editorial page of the Wall Street Journal) exist alongside mainstream media (e.g., network news) and outlets (e.g., New York Times and MSNBC) often deemed “liberal” by conservative media. As Jamieson and Cappella (2008) illustrated, the most prominent conservative media reinforce each other’s messages, creating an “echo chamber” that ultimately undercuts their opponents. Although the emergence of explicitly partisan media should assuage those who believe a partisan press is preferable to reporting by journalists who are “bland” (Diamond, 1986, p. 18), it has generated much concern and controversy

96

Patricia Moy and Muzammil M. Hussain

in academic and popular discourse. Indeed, the presence of such media outlets introduces greater accountability to the political system. The highly emotive content often seen in these outlets invites moral outrage and emotion, which in turn motivate citizens to become engaged. This content also arms audience members with arguments to deploy in subsequent conversations ( Jamieson & Cappella, 2008). At the same time, political incivility can undermine trust in the political system (Mutz & Reeves, 2005), though this fi nding has not always been replicated (Brooks & Geer, 2007). Partisan media are only one indicator of the degree to which the previously singular mass media have fractured. Today, individuals who crave nothing but political news can easily fi nd content to satisfy their needs just as easily as those who eschew information about public affairs and current events. Nowhere is this easier than on the Internet, home to an infi nite number of Web sites (partisan or otherwise) and an unlimited repository of information. This increased access to information, however, does not necessarily translate into a richer public sphere. Research indicates that citizens who receive their news online develop different issue agendas than those who receive their news from more traditional sources, like print newspapers (Althaus & Tewksbury, 2002; Schönbach, de Waal, & Lauf, 2005). Also, as Mutz and Young (2011) speculate, the prevalence of automated information fi ltering might very well create information environments that reduce dissonant voices in discussion circles, inherently changing the nature of the public sphere. Today’s fragmented media environment then provides citizens with great control over what information they consume, but has significant ramifications for the study of public opinion processes, including theorizing about the spiral of silence theory. Bennett and Iyengar (2008) have called for a revisiting of media effects, arguing that the fragmentation of the national audience and proliferation of content-specific (and often partisan) outlets make it only more likely for media messages to reinforce audience members’ predispositions. This relationship likely stems from how citizens exercise selective exposure and select information that is consonant with their beliefs and attitudes (Klapper, 1960). In her review of the literature, Stroud (2011) summarized four reasons that explain why citizens gravitate toward like-minded information. First, as cognitive misers, individuals wish to reduce the level of mental effort they use to process information and, as consonant information requires less effort, they opt for this content over dissonant information (Edwards & Smith, 1996). Second, individuals who experience cognitive dissonance—or confl icting stances that result in psychological discomfort (Festinger, 1957)—might select consistent information not to dispel existing dissonance, but to prevent future dissonance. Third, people might seek like-minded information to fulfi ll their need for closure, or the desire for a solution to a problem or issue (Kruglanski, 1989). Stroud’s (2011) fourth reason for selective exposure links directly to the

Media and Public Opinion in a Fragmented Society

97

fragmentation of mass media and concerns the varying levels of trust that individuals hold in specific sources. Studies of persuasion and attitude change have long focused on the effects of source credibility, but decades of declining trust in media around the world (Donsbach, Rentsch, Schielicke, & Degen, 2009; Pew Research Center, 1997) have garnered interest in news media credibility. Clearly, audience members who trust news media are more likely to adopt the climate of opinion presented (Tsfati, 2003) and are more likely to concur with their presentation of the most social problems (Wanta & Hu, 1994). Moreover, because the news media are no longer a singular phenomenon, and given increasing political partisanship, recent attention has turned to the study of the hostile media phenomenon—the perception of partisans to perceive media content as incongruent with their own views (Vallone, Ross, & Lepper, 1985). Studies in this area acknowledge that credibility is not a static trait within any source, but that individuals’ perceptions of media bias are held relative to their own views, regardless of whether the content is actually biased (Ho et al., 2011). The body of research on selective exposure includes a textured set of fi ndings, hinging on individual-level factors such as motivation, the specific medium used, and the issue being examined (see Brosius & Peter, 2011, for an overview). The extent to which various selective exposure processes are at play differs as well; for example, individuals can attempt to increase their exposure to like-minded information, but without systematically avoiding incongruent information (Garrett, 2009). Naturally then, shifts in the media, social, and political environments bring into question the extent to which individuals scan their environment to gauge the climate of opinion, as the spiral of silence theory postulates. A theoretically more useful question to ask concerns the extent to which individuals, often suffering from information overload brought on by media technologies can scan their environment successfully. As content choices, news recommendation engines, and automated fi ltering technologies proliferate, information is tailored in a way that very well might be deleterious. As Mutz and Young (2011, p. 130) noted, “[t]he fear is that people’s information environments will, without their knowledge or consent, become structured so as to make likeminded views more readily available than dissenting voices. What bothers people most is not being presented with opportunities to fi lter so much as having it done for them without their awareness.” Looking Forward As the information environment continues to evolve, and as citizens change how they process information, any theorizing about the spiral of silence and how it operates should strive to articulate the larger processes by which public opinion is formed. Donsbach and Stevenson (1986) noted how the chain of causal relations in the theory begins with the social-psychological variable of fear of isolation and ends at the macro-sociological level with integration into

98 Patricia Moy and Muzammil M. Hussain

society. This chapter has spoken to how one central tenet, fear of isolation from society, may not hold in this new media environment. As well, individuals’ formerly active scanning of the environment to gauge the climate of opinion may not hold as selective exposure processes work in tandem with technological advancements. These theoretical shifts need to be considered alongside conceptualizations of what it means to express an opinion in contemporary society. As individuals nowadays can blog, upload videos to YouTube, and “like” a candidate or forward posts on Facebook, the Internet offers a new domain for the tribunal of public opinion, one peppered with multiple communities or neo-tribes. As scholars disentangle related concepts such as specialization, segmentation, fragmentation, and polarization—terms that defi ne today’s mediated world (Tewksbury & Rittenberg, 2012)—they also need to consider the individual- and grouplevel processes by which opinion expression and other evolving forms of political action take place. One should not presume that interactive technologies and the plethora of available content make it easy for citizens to fi nd resonant opinion climates and action opportunities. Future theorizing and empirical assessments need to scrutinize the conditions under which citizens choose to simply engage with the public, whether this entity is defi ned for or by them. References Althaus, S. L., & Tewksbury, D. (2002). Agenda setting and the “new” news: Patterns of issue importance among readers of the paper and online versions of the New York Times. Communication Research, 29(2), 180–207. Bennett, W. L., & Iyengar, S. (2008). A new era of minimal effects? The changing foundation of political communication. Journal of Communication, 58(4), 707–731. Brooks, D. J., & Geer, J. G. (2007). Beyond negativity: The effects of incivility on the electorate. American Journal of Political Science, 51(1), 1–16. Brosius, H.-B., & Peter, C. (2011). Selective exposure. In P. Moy (Ed.), Oxford bibliographies: Communication. Retrieved April 15, 2012, from http://www.oxfordbibliographies.com Cova, B., & Cova, V. (2002). Tribal marketing: The tribalisation of society and its impact on the conduct of marketing. European Journal of Marketing, 36(5/6), 595–620. Diamond, E. (1986, March 31). The best and the blandest. New York Magazine, p. 18. Donsbach, W., Rentsch, M., Schielicke, A.-M., & Degen, S. (2009): Entzauberung eines Berufs. Was die Deutschen vom Journalismus erwarten und wie sie enttäuscht werden [Disentchantment of a profession. What Germans expect of journalism and how they get disappointed]. Konstanz: UVK Medien. Donsbach, W., & Stevenson, R. L. (1986). Herausforderungen, probleme und empirische evidenzen der theorie der schweigespirale. [Challenges, problems and empirical evidences of the spiral of silence theory]. Publizistik, 31, 7–34. Edwards, K., & Smith, E. E. (1996). A disconfi rmation bias in the evaluation of arguments. Journal of Personality and Social Psychology, 71(1), 5–24. Eveland, W. P., Jr., & Hively, M. H. (2009). Political discussion frequency, network size, and “heterogeneity” of discussion as predictors of political knowledge and participation. Journal of Communication, 59(2), 205–224. Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press.

Media and Public Opinion in a Fragmented Society

99

Garrett, R. K. (2009). Politically motivated reinforcement seeking: Reframing the selective exposure debate. Journal of Communication, 59, 676–699. Glynn, C. J., & Park, E. (1997). Reference groups, opinion intensity, and public opinion expression. International Journal of Public Opinion Research, 9, 213–232. Ho, S. S., Binder, A. R., Becker, A. B., Moy, P., Scheufele, D. A., Brossard, D., & Gunther, A. C. (2011). The role of perceptions of media bias in general and issue-specific political participation. Mass Communication & Society, 14(3), 343–374. Jamieson, K. H., & Cappella, J. N. (2008). Echo chamber: Rush Limbaugh and the conservative media establishment. New York, NY: Oxford University Press. Klapper, J. T. (1960). The effects of mass communication. New York, NY: The Free Press. Key, V. O. (1961). Public opinion and American democracy. New York, NY: Knopf. Kruglanski, A. (1989). Lay epistemics and human knowledge: Cognitive and motivational bases. New York, NY: Plenum Press. Ladd, J. M. (2012). Why American hate the media and how it matters. Princeton, NJ: Princeton University Press. Lang, S. (2004). Local political communication: Media and local publics in the age of globalization. In F. Esser & B. Pfetsch (Eds.), Comparing political communication: Theories, cases, and challenges (pp. 151–183). New York, NY: Cambridge University Press. McPherson, M., Smith-Lovin, L., & Brashears, M. E. (2006). Social isolation in America: Changes in core discussion networks over two decades. American Sociological Review, 71(3), 353–375. Moy, P., Domke, D. S., & Stamm, K. (2001). The spiral of silence and public opinion on affi rmative action. Journalism & Mass Communication Quarterly, 78, 7–25. Mutz, D. C. (1998). Impersonal influence: How perceptions of mass collectives affect political attitudes. New York, NY: Cambridge University Press. Mutz, D. C., & Reeves, B. (2005). The new videomalaise: Effects of televised incivility on political trust. American Political Science Review, 99(1), 1–15. Mutz, D. C., & Young, L. (2011). Communication and public opinion: Plus ça change? Public Opinion Quarterly, 75(5), 1018–1044. Neuman, W. R., Bimber, B., & Hindman, M. (2011). The internet and four dimensions of citizenship. In R. Y. Shapiro & L. R. Jacobs (Eds.), The Oxford handbook of American public opinion and the media (pp. 22–42). New York, NY: Oxford University Press. Newcomb, T. M. (1948). Attitude development as a function of reference groups: The Bennington study. In M. Sherif (Ed.), An outline of social psychology (pp. 139–155). New York, NY: Harper & Brothers. Noelle-Neumann, E. (1991). The theory of public opinion: The concept of the spiral of silence. In J. A. Anderson (Ed.), Communication yearbook 14 (pp. 256–287). Newbury Park, CA: Sage. Noelle-Neumann, E. (1993). The spiral of silence. Public opinion — Our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Papacharissi, Z. (2002). The virtual sphere: the internet as a public sphere. New Media & Society, 4(1), 9–27. Paxton, P. (2007). Association memberships and generalized trust: A multilevel model across 31 countries. Social Forces, 86(1), 47–76. Pew Research Center for the People and the Press. (1997, March 21). Fewer favor media scrutiny of political leaders: Press “unfair, inaccurate and pushy.” Retrieved January 21, 2010, from http://people-press.org/reports/display.php3?ReportID=112 Pew Research Center for the People and the Press. (2008, August 17). Key news audiences now blend online and traditional sources. Retrieved January 17, 2012. from http://www.peoplepress.org/2008/08/17/key-news-audiences-now-blend-online-and-traditional-sources/ Postmes, T., & Brunsting, S. (2002). Collective action in the age of the Internet — Mass communication and online mobilization RID A-9004-2010. Social Science Computer Review, 20(3), 290–301. Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. New York, NY: Cambridge University Press.

100

Patricia Moy and Muzammil M. Hussain

Pruijt, H. (2002). Social capital and the equalizing potential of the Internet. Social Science Computer Review, 20(2), 109–115. Rainie, L. (2010). How libraries can survive in the new media ecosystem. Retrieved from http:// www.pewinternet.org/Presentations/2010/May/Spain.aspx Robards, B., & Bennett, A. (2011). MyTribe: Post-subcultural manifestations of belonging on social network sites. Sociology—the Journal of the British Sociological Association, 45(2), 303–317. Rokka, J. (2010). Netnographic inquiry and new translocal sites of the social. International Journal of Consumer Studies, 34(4), 381–387. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12(1), 3–28. Schönbach, K., de Waal, E., & Lauf, E. (2005). Online and print newspapers: Their impact on the extent of the perceived public agenda. European Journal of Communication, 20(2), 245–258. Stroud, N. J. (2011). Niche news: The politics of news choice. New York, NY: Oxford University Press. Tewksbury, D., & Rittenberg, J. (2012). News on the internet: Information and citizenship in the 21st century. New York, NY: Oxford University Press. Tsfati, Y. (2003). Media skepticism and climate of opinion perception. International Journal of Public Opinion Research, 15(1), 65–82. Vallone, R. P., Ross, L., & Lepper, M. R. (1985). The hostile media phenomenon: Biased perceptions and perceptions of media bias in coverage of the Beirut massacre. Journal of Personality and Social Psychology, 49, 577–585. Wang, H., & Wellman, B. (2010). Social connectivity in America: Changes in adult friendship network size from 2002 to 2007. American Behavioral Scientist, 53(8), 1148–1169. Wanta, W., & Hu, Y-W. (1994). The effects of credibility, reliance, and exposure on media agenda setting: A path analysis model. Journalism and Mass Communication Quarterly, 71, 90–98.

9 PUBLIC OPINION EXPRESSION IN ONLINE ENVIRONMENTS Patrick Rössler and Anne Schulz

Online media offer Internet users an increasing amount of information about what others think, how they respond to media content, and what they believe to be relevant to people in their personal environment. It is safe to say that all these perceptions, complementing traditional sources, probably shape how online users assess the climate of opinion. At the same time, users themselves have more opportunities than ever before to express their own point of view, which in turn may shape how their readers and viewers perceive the climate of opinion. But beyond its rapid growth in volume, these online environments also display particular characteristics that distinguish them from communication in the offl ine world: Asynchronous interaction, the anonymity of others and the possibility to completely hide one’s own identity are only three of these qualities which might influence individual behavior (namely the willingness to express an opinion) in a way that questions basic assumptions of the spiral of silence theory (SoST). When Noelle-Neumann outlined her theory in the early 1970s, the Internet as we know it today was inconceivable. But its breathtaking diff usion within the past 20 years made it a part of the everyday life of billions of people around the world. As media scholars, we generally face the question of whether theories of media influence need to be revised due to these developments—or if they are still able to explain phenomena we observe in our society today (e.g., Holbert, Garrett, & Gleason, 2010; Rössler, 2007, p. 91). According to Yun and Park (2011), the latter might prove to be a fruitful endeavor: “In reality […] it is not very clear where offl ine communication ends and online communication begins. […] The merging of traditional media and online media suggests that the theories of traditional mass communication are still relevant for many types of CMCs [computer-mediated communications], if not all” (p. 202).

102

Patrick Rössler and Anne Schulz

Scholars so far have only occasionally verified basic SoST assumptions in CMC, e.g., by testing the willingness to speak out in online chat rooms using a real-life-setting (Wanta & Dimitrova, 2000), or by analyzing reactions in a face-to-face discussion compared to CMC debates (Ho & McLeod, 2008; McDevitt, Kiousis, & Wahl-Jorgensen, 2003). Others have experimentally manipulated anonymity in online-only settings (Yun & Park, 2011). In contrast to these empirical studies, our chapter argues from a theoretical point of view: We will use evidence from a broad range of fields, including particularly current research in media psychology, to discuss the robustness of central SoST assumptions against the backdrop of recent changes in people’s use of new media environments. We discuss whether conditions which are especially found in computer-mediated environments (e.g., anonymity or respectively non-identifiability) contradict or support Noelle-Neumann’s theory.1 The Multi-Level Logic as a Theoretical Framework for SoST As a theory that relates both to individual and societal processes NoelleNeumann’s SoST needs to be conceptualized as a multi-level theory, so it can be described with a multi-level logic outlined by German sociologist Hartmut Esser (1999).2 Following his argument, we distinguish between four types of influences described previously by Pan and McLeod (1991): macro to macro, macro to micro, micro to micro, and micro to macro. Accordingly we will propose to model basic elements of the SoST as a modification of Esser’s approach (see Figure 9.1).3 This application results in a more detailed perspective on some processes within SoST that might have changed due to Internet communication. As probably the most substantial change, we propose to introduce a mesolevel of analysis which addresses group influences on how people perceive the climate of opinion. Esser’s (1999) model of sociological explanation starts with a so-called collective phenomenon I located on the upper left side. In our case, the tone of the mass media exerts an influence on how individuals in a society perceive the climate of opinion, but also on the perceptions within the social entities to which individuals can amalgamate (Noelle-Neumann, 1993). What Esser calls the logic of situation (1; macro to micro) comes close to the quasi-statistical sense in SoST: Individuals permanently assess what the majority might think, and which opinion, on the other hand, is supported by a minority of people only. For this purpose they observe mass media messages and personal communication in their immediate social environment (Scheufele, 2008, p. 348). Because of their consonant, cumulative, and ubiquitous coverage, Noelle-Neumann argued that the mass media have always played a key role within this perceptual process (for further considerations on the logic of the SoST within the Internetage see Schulz & Roessler, 2012).

Public Opinion Expression in Online Environments

103

macro-l8'lal macro-lavel t~.

of ... the

m".

ma. mn. media

collective phenomenon

ooIleetiva phenomenon II coIl&cliYe

-,

mno-Ievel me.o-Ievel &Oclal toel,l

anvlronmen1 envlronmeni

sodal entity I

entity II

public opinion opinion

, .apl plral ralof of

.,lienee ilenee

micro-lavel perception pereeptionof of opInionclimate opInioncilmate (qual4tati.tical (quasl4tati$tical monitoring)

action II

indivic1Jalli indMdual 1t

action I

rindividual uivlduallI

2

own~lnl0r.

fur fear of Iioial laolat on

1 logical ai'Jlllion

2

lc9col~

wlilingnell wlillngnHl

toapellkout to lpeak out

;1

lagi:d8ggMgalUl

FIGURE 9.1 Basic elements of the SoST, adapted to Esser’s model of sociological explanation (1999). Source: diagram by the authors, based on Esser (1999, pp. 100–107).

Individuals compare their own opinion with what they perceive as being the opinion of a majority in society, and the behavioral outcome of this comparison may in Esser’s terms be described by the logic of selection (2; micro to micro): People who believe their opinion to be consonant with the majority will take action and express this opinion in public. Other individuals who experience dissonance between these opinions have to face sanctions by their social environment exposing their own views in public. According to Noelle-Neumann, human beings have a social nature that makes them fear isolation from society. Being threatened by exclusion when expressing the wrong idea, they chose to remain silent. Over time, the consequences are explained by the logic of aggregation (3; micro to macro) in Esser’s model: The assumed majority opinion gets stronger and stronger, while—as indicated by the term spiral of silence—the opposite views may become almost invisible in public. The outcome of this process leads to Esser’s collective phenomenon II, the public opinion in terms of NoelleNeumann and also to opinion formation processes on the meso-level of social groups which might follow the dynamics of SoST as well. The dashed line between collective phenomena I and II refers to a macro-to-macro relationship between tone in the mass media and public opinion. In this chapter we will not elaborate on the vast number of developments within the media landscape during the past 40 years, and their impact on the SoST. Rather, our theoretical considerations will try to understand the

104

Patrick Rössler and Anne Schulz

circumstances under which individuals expect sanctions from others in online environments which did not exist in Noelle-Neumann’s era and how fear of isolation arises hereinafter. Also, we want to take a brief look at the consequences which might occur for the processes described by the logic of selection and the logic of aggregation (see Moy and Hussain’s Chapter 8, this volume). Expecting Sanctions in Offline and Online Environments In SoST the social environment of an individual emerges from his or her actions in public. Noelle-Neumann (1993, p. 19) is less interested in private settings such as families and circles of friends which constitute an important part of people’s social life but, by and large, do gain little relevance in public opinion formation. Her focus refers to a social environment of mostly anonymous participants emerging by encounters in public space where speaking out may potentially affect others, and the speaker is identifiable and confronted immediately by the reactions of these others (p. 81). Neither Noelle-Neumann nor her critics have had reason to propose that this limited but anonymous social environment could—beyond its influence on the perceived opinion climate—not directly impact an individual’s fear of isolation. Looking at SoST in non-virtual contexts, this makes perfect sense. If the individual expresses a non-conforming opinion in an offline public sphere, he or she always has to expect sanctions inevitably and of necessity, because he or she has no opportunity to escape from the situation. Accordingly, the influence of the social environment on an individual’s expectation of sanctions and subsequent fear of isolation is invariant as it only depends on the individual’s perceived climate of opinion. No matter how the social environment in a particular case is compounded, it exerts the same influence on each individual belonging to the minority: The person will have to expect sanctions, while the act of sanctioning needs not even to occur. The mere threat of potential social isolation forces the individual to remain silent (p. 77). The fundamental changes in our everyday communication environment include that in cyberspace, the social environment no longer consists of random encounters during a train ride representing the anonymous public. The Internet and particularly social media have facilitated the establishment of new social environments which are entities outside one’s own private settings, but may remain stable although usual mechanisms of personal friendship, face-toface meetings and a shared physical space of living do not apply. We assume that the expectations of sanctions (which seem to be unavoidable in offl ine situations) are likely to vanish at least partly in these online environments. As discussed previously by McDevitt et al. (2003, p. 458), we propose that the individual can easily estimate how likely it is to get sanctioned by the social environment. First, they assume that many types of sanctions depend on the physical presence of others. However, as Short, Williams, and Christie (1976, p. 65) have testified with their concept of social presence, physical presence (e.g.,

Public Opinion Expression in Online Environments

105

facial expression, posture, etc.) is with no doubt highly reduced in mediated settings. Second, McDevitt et al. (2003; referring to Burgoon, Buller, & Woodall, 1989) state that normative influence is often communicated non-verbally, e.g., by eye-contact, gestures, or similar. Mostly these signals cannot be conveyed adequately in CMC. Third, CMC is lacking social context cues which normally give hints about the status of the communication partners (see Kiesler, Siegel, & McGuire, 1984). If these cues are missing, group hierarchies are less obvious and thus lose their meaning as a source of social influence, as Ho and McLeod (2008) pointed out.4 Assuming that these conclusions hold true, central antecedents of the SoST are under question, challenging the SoST on the whole to a certain degree as well: If a person does not expect sanctions from his or her social environment, he or she has the opportunity to express his or her opinion freely, no matter if it corresponds to the perceived majority opinion of the general public or not. The person in question would, in other terms, exhibit disinhibited or anti-normative behavior and thus contradict basic assumptions of NoelleNeumann’s theory, which has taken this as a pathological exception. But before jumping to conclusions, the concept we have so far called expectations of sanctions needs further clarification. This additional factor shall describe the individual`s subjective estimation of the likelihood of sanctions which may vary between different communicative situations. Noelle-Neumann herself only spoke about a “threat of being excluded or losing one’s standing in society” (1993, p. 179), which was automatically experienced when the individual discovered that he or she is part of the minority. We argue that whether a person expects sanctions or not depends highly on situational factors such as the general characteristics or qualities of a communication space. One quality we expect to have a relevant impact on the extent of public opinion expression in online environments is the perceived anonymity. To investigate this idea, we want to make use of current fi ndings on the effects of anonymity in online and offl ine environments. So far, research conducted on the so-called SIDE model (Social Identity Model of Deindividuation; see below for a more detailed explication) has generated substantial insight in this regard. Yun and Park (2011, p. 216) have also mentioned possible connections between SoST and the SIDE model in passing: “… the results from the behavioral measures of this study [posting behavior in online forums] can be adequately accounted for by the SIDE model. (…) Further study on this model in the context of congruent and incongruent online forum messages will deepen our understanding of the psychological process of the message posting behavior in the context of public opinion formation.” We want to elaborate on this issue in greater detail and introduce a theoretical model from which we can possibly deduce how special characteristics of online environments—namely the anonymity and the identifiability (the two central dimensions in SIDE research)—influence the individual expectations of sanctions and thus a person’s willingness to speak out.5

106

Patrick Rössler and Anne Schulz

Anonymity and Identifiability in Offline and Online Environments According to the SIDE model, which will be explained below in greater detail, we have to distinguish between what the individual knows about the others (anonymity) and what the individual exposes from himself or herself to the others (identifiability; see, e.g., Reicher, Spears, & Postmes, 1995). These two concepts need to be understood as two dimensions independent from each other rather than the poles of the same dimension. While this is particularly relevant for online contexts, two accounts seem true for offl ine environments: First, we can assume that these independent dimensions almost always coincide (Mayer-Uellner, 2003, p. 42): Whenever two individuals unknown to each other meet on the street, anonymity and identifiability cancel each other out; while the one person makes himself or herself more and more identifiable, the other person also loses his or her anonymity (e.g., through face and body perception etc.). Second and related to the previous aspect, it is almost impossible to fi nd completely anonymous or unidentifiable persons in offl ine environments: Whenever two persons meet, visual markers immediately give hints about the identity of a person. Even if the person uses a mask to hide his or her face, the other will, for instance, be able to identify his or her gender by the person’s voice or body shapes. When it comes to online environments, both accounts no longer apply. First, according to the features of some virtual spaces, one person might be more identifiable than the other. Hence a particular individual is not identifiable for the others whereas the others are not anonymous at all (or vice versa). This is the one reason why we cannot assume a single dimension here—the one condition is not the opposite of the other, so if one person is identifi able in the online world this does not always mean that the others are, too. And second, we can also observe a complete anonymity or total non-identifi ability in online environments. Thus, as visualized in Figure 9.2, we distinguish between four different types of communicative spaces in online environments: high anonymity and high identifiability (QI), high anonymity and low identifiability (QII), low anonymity and low identifiability (QIII), and low anonymity and high identifiability (QIV). As these constellations are, of course, also conceivable in offl ine environments, we added examples for these situations, too.6 A SIDE Explanation for the Development of Expectations of Sanctions To formulate a theory-based hypothesis on the relevance of anonymity and identifiability for SoST, we will follow insights of research in social psychology about how these two dimensions affect the individual expectations of sanctions in online environments.

Public Opinion Expression in Online Environments

107

high anonymity nothing is known about the others

offline: train compartment

offline: holding a speech

online: unrestricted discussion forum

online: personal weblog

high identifiability

QII QI

low identifiability the individual hides his or her identity

the individual reveals information about his or her identity

QIII QIV offline: candid camera

offline: school class

online: faked profile in a social online network

online: social online network

low anonymity information about the others is available

FIGURE 9.2 Dimensions of anonymity and identifi ability in online and offl ine environments. Source: diagram by the authors.

Models of Deindividuation and the SIDE Model Evidence of how anonymous communication environments are connected to anti-normative behavior of and within groups was collected by research on aggression. On the one hand, traditional and modern models of deindividuation, represented, for example, by Zimbardo (1969) and Diener (1979), assume that factors as anonymity or immersion “lead to a loss of self hood and hence of control over behavior” in a group (Reicher et al., 1995, p. 161). Haney, Banks, and Zimbardo (1973) could show this impressively in their classical Stanford-Prison-Experiment: Due to uniforms for the guards and numbers for the prisoners, an anonymous environment was created, which later helped to explain the observed loss of self hood in terms of deindividuated and aggressive behavior. However, just as the Stanford-Prison-Experiment has been criticized for several reasons, the whole line of deindividuation research faced dissent (see, e.g., Reicher et al., 1995, pp. 167–175). Other studies have shown “that mediated interaction often includes friendly, non-combative discussion” (McDevitt et al., 2003, p. 458), although these are mostly anonymous as well. Therefore, in opposition to individualistic and anti-collectivistic approaches, a second line of research developed, which is “explicitly concerned with the social nature of self hood and its relationship to group behavior” (Reicher et al., 1995, p. 176). One important offspring of this research line is the Social Identity Model of Deindividuation (SIDE model), which basically proposes that anonymous circumstances lead to an increased behavioral orientation on group-norms rather

108 Patrick Rössler and Anne Schulz

than to a loss of self hood and subsequent aggressive behavior. Developed by Reicher et al. in 1995, it has since then been tested especially in computermediated environments (see also Postmes, Spears, Lea, & Reicher, 2000; Spears, Postmes, Lea, & Watt, 2001). The model is based on social identity theory (SIT; Tajfel, 1978) and the social-categorization theory (SCT; Turner, 1987) and consists of two elements: a cognitive dimension referring to the concept of anonymity and a strategic dimension concerning the effects of identifiability. The Cognitive Dimension of the SIDE Model The model starts from the assumption that the human self-concept consists of two major subsystems: the personal and the social identity. Both identities represent cognitive structures that provide a system of concepts for individual selfdefi nition (Gergen, 1971, p. 23). The personal identity refers to characteristics and typical behavior inherent in the individual such as “[…] feelings of competence, bodily attributes, ways of relating to others, psychological characteristics, intellectual concerns, personal tastes and so on“ (Turner, 1982, p. 18). The social identity is composed of all group memberships the individual is aware of. It is further assumed that in particular situations the personal identity switches to the social identity—this is most likely to happen in intergroup-contexts, but how this switch actually works is not fully understood yet (Turner, 1982, p. 21). If the social identity is salient, the individual orients his or her behavior towards the norms of a certain group—namely the norms of that group to which the individual feels to be part of in that particular situation. Prototypical group-members serve as behavioral examples (Brown, 2001, p. 557). This can also imply that the individual shows unusual behaviors or refrains from usual behaviors, both of which would not emerge if the personal identity was salient. The motivation for this behavior is explained by the social identity theory: “[…] self-categories tend to be evaluated positively and […] there are motivational pressures to maintain the state of affairs” (Turner, 1987, p. 57). The human being aims to achieve a self-concept as positive as possible, by establishing positive distinctiveness between groups (Tajfel, 1978, p. 83). This state can be reached by striving for memberships in groups which have a positive connotation. Due to permanent processes of group comparison, the individual can tell which groups are positively connotated and which are not; to maintain membership, the individual has to respect the norms that are valid in that group. With regard to SoST, the individual’s tendency to hide one’s own opinion when perceived as being part of the minority resembles the motivation to adopt the norms of a certain group according to the SIDE model: It is the fear of being excluded from certain circles. In Noelle-Neumann’s theory, this circle is the society as a whole; in the SIDE model the individual does not want to lose membership of a certain group. Although in the latter conformity is not ascertained normatively by a majority pressure but by simple role-taking (Price,

Public Opinion Expression in Online Environments

109

Nir, & Cappella, 2006, p. 51), the outcome may qualify a society as a particular type of group. The case of a salient personal identity on the other hand could be related to what Noelle-Neumann has described as the hard core or the avant-garde (Noelle-Neumann, 1993, 139‒142). The SIT terms of ingroup and outgroup can be equated with the minority and majority, and the process of comparing these groups with each other would in the SoST be described by the quasi-statistical sense. In a next step, the cognitive dimension of the SIDE assumes that anonymity increases the concept of identity (either personal or social identity), which has been salient before entering the anonymous situation (see Figure 9.3): If the social identity was salient, anonymity will lead to an increased orientation on group norms. But if group salience is low, anonymity increases the individual focus on very own needs and desires. This is explained by Reicher et al. (1995, p. 178): If the group level of identity is already emphasized, then anonymity is likely to enhance the salience of social identity still further by emphasizing the interchangeability of group members and obscuring interpersonal differences. However, anonymity is likely to have rather different effects if it is not accompanied by explicit group immersion. Where group salience is low and where group boundaries are indistinct, anonymity may undermine the process of depersonalization, by emphasizing one’s isolation from the group or by further obscuring group boundaries. Hence the salience of social identity should decrease. These assumptions were confi rmed empirically in a series of studies (see, e.g., Spears et al., 2001) and do not contradict basic assumptions of the SoST. Noelle-Neumann herself stated that fear of isolation—or the fear of being sanctioned and the subsequent orientation on the prevailing public opinion— is considerably higher in the considerably bigger anonymous public sphere (Noelle-Neumann, 1993, p. 215). To sum up, the cognitive dimension of the SIDE model indicates that individuals can experience the salience of either a social or a personal identity. One of each is always activated, and the switch between the two identities is led by situational factors (such as obvious intergroup contexts). Anonymity in a communicative space enhances the focus on the identity which has been salient before and accordingly the individual is oriented either towards his or her own norms (in the case of salient personal identity) or towards relevant group norms (in the case of salient social identity). In the next part of this chapter, we look at the strategic dimension of the SIDE model. The Strategic Dimension of the SIDE Model In his research, Ng (1982a, 1982b) showed that anonymity only enhances the individual’s focus on his or her social identity when facing other groups of

110

Patrick Rössler and Anne Schulz

s ocial identity salient

personal i dentity salient

anonymity and other deindividuating factors

increased group salience

reduced group salience

adherence to group norms and standards

adherence to personal norms and standards

FIGURE 9.3 The cognitive dimension of the SIDE model. Source: Spears, Lea, & Postmes (2000).

equal or lesser power but not when facing groups of greater power than the ingroup (Reicher et al., 1995, p. 186). So the salience of group membership is not a sufficient condition for the actual expression of group behavior. Actually, norm orientation towards the ingroup is decreased, when the individual perceives the presence of a stronger outgroup. The individual then decides strategically which range of behaviors it can display in public and which not. These power relations between groups are moderated by the identifiability of the individual: Whether an individual shows all behaviors which are typical for a salient ingroup depends on whether he or she is visible for ingroup and/or outgroup members, and on how strong the outgroup is perceived: [W]hen ingroup members are made more identifiable with respect to the outgroup, the relative power of outgroup over ingroup will be increased by increasing their ability to hold ingroup members to account for their actions. On the other hand, when ingroup members are made more identifiable with respect to ingroup members, the relative power of outgroup over ingroup will be decreased by increasing the ability of ingroup members to support each other in resisting the outgroup. (Reicher et al., 1995, p. 187)

Figure 9.4 shows four possible group-contexts in which a person can be located and possible consequences regarding the person’s norm orientation:

Public Opinion Expression in Online Environments

111

a. If the individual is identifiable for one (or more) weaker outgroup(s), following the SIDE model he or she will accommodate his or her behavior to suit the salient norms of the ingroup. The individual cannot be accounted for his or her actions by the outside.7 b. If the individual is identifiable for a stronger outgroup, the person will cut back the ingroup-adequate behaviors to an extent which is accepted by the outgroup. This does not mean that the individual automatically takes over all norms of the outgroup and forgets about the norms of his or her own (but in this situation absent) ingroup. The process of adapting behaviors will stop as soon as the individual cannot be sanctioned by the outgroup anymore. c. If the individual is only identifiable towards his or her own ingroup, he or she will act completely in line with the ingroup norms. With no stronger outgroup in sight distance, the person does not have to expect sanctions from outside. d. If the individual is identifiable for both the ingroup and a strong outgroup, the situation is more complex but well explained by SIDE scholars: Usually the individual would cut back behavior dissonant to the outgroup when appearing in a social setting. But the presence of the ingroup may provide a so-called social support (Spears, Lea, & Postmes, 2000, p. 4). Despite being identifiable for a stronger outgroup, the individual can transgress its norms because it is supported by his or her ingroup. When merging these assumptions of the SIDE model’s strategic dimension with the SoST, Noelle-Neumann’s concept does not include social support of the ingroup. Human beings always orient their behavior towards the big societal norm—public opinion. Comparing their position with public opinion, they adapt their behavior—either they speak out about their own point of view (which they perceive to be compatible with the majority) or they fall silent.

individual is identifiable towards:

weak outgroup

strong outgroup

ingroup

ingroup & strong outgroup

adherence to ingroup norms

limiting the adherence to ingroup norms to norms also accepted by the outgroup

adherence to ingroup norms

social support leads to adherence to ingroup norms even though members of stronger outgroups can watch the individual

b)

c)

d)

a)

FIGURE 9.4 Sources of norm orientation, depending on a person’s identifi ability. Source: diagram by the authors.

112

Patrick Rössler and Anne Schulz

In the latter case, when assuming dissonance, the public opinion represents the opinion of a stronger outgroup which is not questioned by the individual. For Noelle-Neumann’s (1993) approach, it is totally irrelevant for whom the individual is identifiable. Since its original formulation, this rigid idea has set off a persistent scientific debate, which discusses the role of reference groups within the opinion formation process (see, e.g., Price & Allen, 1990; Salmon & Kline, 1985). Online Environments: Bringing SIDE Assumptions and SoST Together According to the SIDE model, the point of reference for the behavioral orientation in a social setting depends on the group(s) towards which the individual is identifiable. However, the orientation towards group norms decreases in strength under conditions of anonymity. The four communicative situations we defi ned above will serve as scenarios to assess when the individual would, or would not expect sanctions. To proceed step-by-step, we will fi rst look at the dimension of anonymity alone. As shown in Figure 9.5, the norm adherence should become stronger the higher the anonymity is perceived in the particular communication space which applies for the quadrants QI and QII. This effect occurs because a group is perceived as more homogenous under conditions of anonymity—here group specific norms and subsequent behavior become more distinct (Reicher et al., 1995). Of course, the individual adheres to group norms also under nonanonymous conditions, but the strength of this adaption differs. Due to a high level of anonymity in virtual environments, these could qualify as spaces in high anonymity

low identifiability

QII QIII

QI QIV

high identifiability

low anonymity strength (high/low) of norm orientation depending on the factor of anonymity

FIGURE 9.5 Strength of norm orientation considering the dimension of anonymity. Source: diagram by the authors.

Public Opinion Expression in Online Environments

113

high anonymity

low identifiability

QII QIII

QI QIV

high identifiability

low anonymity strength (high/low) of norm orientation depending on the factor of anonymity strength (high/low) of norm orientation depending on the factor of identifiability

FIGURE 9.6 Strength of norm orientation considering both SIDE dimensions. Source: diagram by the authors.

which individuals focus stronger on salient (group) norms than in any comparable offl ine environment. The SoST would benefit from this indication, as it appears as if in online environments norm-conforming behavior becomes the rule rather than the exception. At the same time the second dimension of SIDE assumes that the higher a person’s identifiability, the higher the orientation towards particular group norms. Hence the norm orientation should increase in situations represented by the quadrants QI and QIV (see Figure 9.6). Apparently those situations might be of particular interest for SoST where the user is not identifiable (such as discussion forums, newsgroups, etc.) and individual behavior may contradict basic assumptions of the SoST. Suggesting that the effects of anonymity and identifi ability on norm orientation interact (they could cancel out or mutually intensify each other), we are prepared to propose rather detailed assumptions about how strongly individuals perceive expectations of sanctions. We conclude that expectations of sanctions should, roughly speaking, vary between three different levels depending on how anonymity and identifiability interact in a particular online environment (see Figure 9.7): A high expectation of sanctions should occur if the individual is confronted with a situation where knowledge about others is low but his or her own identity is recognizable for others. The individual faces both ingroup and outgroup norms and has to adapt his or her behavior (see context d of Figure 9.4). A private blogger, for example, would be in exactly this situation and would thus act in a permanent state of surveillance, assuming that any minute members of outgroups can observe his or her behavior, which therefore needs

114 Patrick Rössler and Anne Schulz

high anonymity nothing is known about the others

moderate expectations of sanctions

high expectations of sanctions

strong adherence to ingroup norms; outgroups have no influence

strong adherence to ingroup as well as to outgroup norms

high identifiability

QII QI

low identifiability the individual hides his or her identity

the individual reveals information about his or her identity

QIII QIV low expectations of sanctions

moderate expectations of sanctions

adherence to ingroup norms; outgroups have no influence

adherence to ingroup norms; outgroups are not present

low anonymity information about the others is available strength (high/low) of norm orientation depending on the factor of anonymity strength (high/low) of norm orientation depending on the factor of identifiability

FIGURE 9.7 Expectations of sanctions in online environments. Source: diagram by the authors.

to be oriented towards the outgroup norms.8 This is exactly the case where, according to the SoST, an individual perceives a dissonant climate of opinion and should fall silent or respond to the perceived majority opinion. In contrast, the opposite situation where the individual manages to hide his or her identity while others are well known induces little or no expectation of sanctions; the outgroup thus has no influence as one’s guiding principle is the ingroup norm. This case indeed contradicts the notions of SoST, but as explained above, this situation applies only for a few constellations in online or offline communication so the relevance of this outcome in reality should not be overestimated. If all participants are not identifiable or only poorly identifiable (see QII of Figure 9.7), behavior in general should follow the norms of the respective salient ingroup, particularly in anonymous environments. Here sanctions are expected on a moderate level; however it is worth questioning whether the effect of anonymity (leading to norm-conforming behavior) is strong enough to tip over the effect of the non-identifiable condition (leading to anti-normative behavior), or if in these situations, too, the orientation on group norms remains comparably low. Expectations of sanctions should also even out at an average level when all participants are identifiable for everybody involved in the discussion (see QIV of Figure 9.7); the ingroup norm is dominant. Obviously, these and other assumptions suggested here on the basis of theoretical considerations will have to be tested empirically in the future. According to the studies of Yun and Park (2011), Ho and McLeod (2008), Wanta and

Public Opinion Expression in Online Environments

115

Dimitrova (2000), and McDevitt et al. (2003) mentioned in the introduction of this chapter, we would also suggest to pursue experimental approaches. It would be a major progress in spiral of silence research to determine how online environments are actually perceived by the users; spaces which we have defi ned as anonymous might not necessarily appear anonymous for users acting there. Also, there is little knowledge that goes beyond anecdotal evidence whether norm-violating behavior is indeed a popular phenomenon in online discussions (we assume that it is to a lesser degree than expected). Subsequently, norm orientation could be manipulated in different settings serving as stimuli, and online environments in particular would offer the opportunity to register outspokenness over time. However, ethical constraints limit the options for researchers here as any field experiment may cause real consequences for the individuals outside the survey situation. Hence quasi-experimental designs seem to be appropriate for many purposes, using the archives of blogs and commentaries, supplemented by additional interviews with discussants. Conclusion Bringing together assumptions of the SoST and central ideas of SIDE research seems to be a fruitful approach when trying to explain the psychological processes relevant for public opinion formation in online environments. Our theoretical analysis suggests that central assumptions of the SoST, namely the perception of some sort of social control and the subsequent expectation of sanctions, are still valid in some online environments. As we argued with the help of current SIDE fi ndings, expectations of sanctions exist but may vary depending on the particular profi le of a communication environment. The more anonymous the environment, the more homogenous is the perception of the others—and the more important becomes norm-conforming behavior to not give a negative impression, which would bear the risk of exclusion from the group (i.e., the society). Correspondingly, the expectations of sanctions should rise. To which norms the individual adheres depends on the groups to which he or she is identifiable for: In most cases these should be at least the norms of his or her ingroup(s), assuming the salience of a social identity. However, this might be restricted when a stronger outgroup is present; if the individual is completely identifiably, high expectations of sanctions should lead his or her behavior to norm conformity. Depending on the level of expectations of sanctions, the perceived climate of opinion (which is the basic point of reference in Noelle-Neumann´s theory) becomes more or less important. In the case of high expectations of sanctions, the individual should behave in accordance to the perceived climate of opinion, so if a dissonant climate of opinion is perceived, the individual consequently falls silent. But if expectations of sanctions are low, it should be the rule that the perceived climate of opinion loses its importance and the individual is willing

116

Patrick Rössler and Anne Schulz

to express his or her opinion even in case of a perceived dissonance (which is basically ignored). Keeping in mind that anonymity and identifiability are not at all onlinespecific phenomena, our fi ndings should be applicable to offline contexts as well. But as explained above, high anonymous conditions that we can fi nd online are almost never met in offline environments. Online we are dealing with a new quality of interactions and thus effects should be observable which do not occur in offline settings (at least not in the same frequency and thus not with the same overall effect on the public opinion process in general). The SoST consequently has to face at least some problems that can only be validated by empirical research. Moreover, online and offline effects almost never occur separated from each other, as it seems reasonable individuals are usually involved in both types of communication at the same time and with regard to the same issues. While our theoretical analysis was unable to include interactions between online and offline processes, empirical research as well might face a serious problem separating out the respective processes properly. And to further enhance complexity in this realm we have to account for the fact that a substantial overlap between real-world groups and online groups exists: people in one’s personal environment often constitute the core of one’s social media network (see e.g., Boyd, 2008, 16‒17). Possibly the distinction between online and offl ine will prove to be dispensable for SoST in the future if media use patterns continue to merge. Notes 1. The authors wish to thank Kerstin Bienhaus, Julia Ebert, Madeleine Kling, Mareike Mewes, Kristin Ramcke, Johanna Schade, and Anja Schorr, constituting the research group “Meinungsraum” at the University of Erfurt, for their substantial contributions to our argument. We are deeply indebted to Seth Ashley for his thoughtful comments on the fi nal draft of this manuscript. The research was supported by the Stiftung Demoskopie Allensbach. 2. How this model was outlined is briefly explained by Esser (2004): “The model of sociological explanation is an advancement of the concept of situational logic proposed by Max Weber and Karl R. Popper. The concept was further developed and applied in sociology by, e.g., Robert K. Merton, James S. Coleman, Raymond Boudon, Siegwart Lindenberg and Reinhard Wippler, as well as more recently by John Goldthorpe. On the connection between this concept and the idea of ‘generative mechanisms,’ see Hedström and Swedberg, 1998. On the subsumption under details of general sociology, see Esser, 1993, 1999” (p. 1127). 3. That the SoST is easily applicable to that multi-level logic has been demonstrated earlier by Jäckel (2001) as well as Scheufele (2008), who were both discussing multi-level strategies by using Noelle-Neumann’s theory as an appropriate case. 4. However, in the spiral of silence process as outlined by Noelle-Neumann, status characteristics (e.g., age and gender; see Ho & McLeod, 2001, p. 192) are of no importance at all. Drawing parallels to Social Role Theory we could at the most ascribe a higher status to people holding the majority opinion (which is also supported by the mass media) whereas people holding the minority opinion would represent a lower status. 5. We do not want to rule out the possible influence of other features of CMC such as the degree of asynchronicity or the mere size of the communication space. Even individual factors (e.g., traits of character or personal involvement in the issue in question) could have an effect on how the user estimates the chance of getting sanctioned by his or her social

Public Opinion Expression in Online Environments

117

environment. But as other scholars have already looked at the effects of anonymity on fear of isolation, this seems to be a fruitful attempt to get started in SoST-online research. 6. Regarding these examples, the anonymity dimension seems related to audience size as it is more probable that when speaking, for instance, to a large crowd we may assume that this crowd is anonymous to the speaker. However, there are also situations where even members of a small audience might be unknown to the speaker, and on the other hand a larger crowd may be familiar for a person holding a keynote talk to his or her peers. 7. Another possibility, the identifi ability for a weak out-group and the in-group, is not indicated in our figure as the reaction of the individual in this situation should correspond to his or her reaction in a situation in which he or she is only identifi able to the ingroup: The person would orient his or her behavior totally on norms that are consonant with the ingroup. 8. Individuals who do not expect sanctions in these situations probably belong to the hard core or the avant-garde. We assume, for instance, that most bloggers belong to either the hard core or the avant-garde when writing provocative statements. The “normal individual” would accordingly not blog (speak up) against the perceived opinion climate.

References Boyd, d. (2008). Facebook’s privacy trainwreck. Exposure, invasion, and social convergence. Convergence, 14(1), 13–20. Brown, R. (2001). Beziehungen zwischen Gruppen [Relations between Groups]. In W. Stroebe, K. Jonas, & M. Hewstone (Eds.), Sozialpsychologie. Eine Einführung [Social Psychology. An Introduction] (pp. 537–576). Berlin, Germany: Springer. Diener, E. (1979). Deindividuation, Self-Awareness and Disinhibition. Journal of Personality and Social Psychology, 37(7), 1160–1171. Esser, H. (1999). Soziologie. Allgemeine Grundlagen [Sociology. General Basics]. Frankfurt, Germany: Campus. Esser, H. (2004). Does the “new” immigration require a “new” theory of intergenerational integration? International Migration Review, 38(3), 1126–1159. Gergen, K. J. (1971). The concept of self. New York, NY: Holt, Rinehart and Winston. Haney, C., Banks, C., & Zimbardo, P. (1973). A Study of Prisoners and Guards in a Stimulated Prison. Naval Research Reviews, 9, 1–17. Ho, S. S., & McLeod, D. M. (2008). Social-psychological influences on opinion expression in faceto-face and computer-mediated communication. Communication Research, 35(2), 190–207. Holbert, R. L., Garrett, R. K., & Gleason, L. S. (2010). A new era of minimal effects? A response to Bennett and Iyengar. Journal of Communication, 60(1), 15–34. Jäckel, M. (2001). Über welche Brücken muss man gehen? Die Mehr-Ebenen-Analyse und ihre Relevanz für die Rezeptionsforschung [Which Bridges to Cross? Multi-Level-Analysis and its Relevance for Research on Media Reception]. In P. Rössler, U. Hasebrink, & M. Jäckel (Eds.), Theoretische Perspektiven der Rezeptionsforschung [Theoretical Perspectives on Research on Media Reception] (p. 35–58). München, Germany: Reinhard Fischer. Kiesler, S., Siegel, J. & McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39(10), 1123–1134. Mayer-Uellner, R. (2003). Das Schweigen der Lurker. Politische Partizipation und soziale Kontrolle in Online-Diskussionsforen [The Silence of Lurkers. Political Participation and Social Control in Online Discussion Forums]. München, Germany: Reinhard Fischer. McDevitt, M., Kiousis, S., & Wahl-Jorgensen, K. (2003). Spiral of moderation: Opinion expression in computer-mediated communication. International Journal of Public Opinion Research, 15(4), 454–470. Ng, S. H. (1982a). Power and appeasement in intergroup discrimination. Australian Journal of Psychology, 34, 37–44. Ng, S. H. (1982b). Power and intergroup discrimination. In H. Tajfel (Ed.), Social identity and intergroup relations (pp. 179–206). Cambridge, England: Cambridge University Press.

118

Patrick Rössler and Anne Schulz

Noelle-Neumann, E. (1993). The spiral of silence. Public opinion — Our social skin. Chicago, IL: University of Chicago Press. Pan, Z., & McLeod, J. M. (1991). Multilevel analysis in mass communication research. Communication Research, 18(2), 140–173. Postmes, T., Spears, R., Lea, M., & Reicher, S. (Eds.). (2000). SIDE issues centre stage: Recent developments in studies of de-individuation in groups. Proceedings of the colloquium, Amsterdam. Price, V., & Allen, S. (1990). Opinion spirals, silent and otherwise. Applying small-group research to public opinion phenomena. Communication Research, 17, 369–392. Price, V., Nir, L., & Cappella, J. N. (2006). Normative and informational influences in online political discussions. Communication Research, 16, 47–74. Reicher, S. D., Spears, R., & Postmes, T. (1995). A social identity model of deindividuation phenomena. European Review of Social Psychology, 6, 161–198. Rössler, P. (2007). Wirkungsmodelle: die digitale Herausforderung revisited. Forschungsstand zu Wirkungen von Online-Kommunikation — ein rückblickender Essay [Effect Models: The Digital Challenge Revisited. Status Quo of Research on Effects of Online-Communication—A Retrospective Essay]. In S. Kimpeler, M. Mangold, & W. Schweiger (Eds.), Die digitale Herausforderung. Zehn Jahre Forschung zur computervermittelten Kommunikation [The Digital Challenge. Ten Years of Research on Computer-Mediated Communication] (pp. 91–106). Wiesbaden, Germany: Verlag für Sozialwissenschaften. Salmon, C. T., & Kline, F. G. (1985). The spiral of silence. Ten years later. An examination and evaluation. In K. R. Sanders, L. L. Kaid, & D. Nimmo (Eds.), Political communication yearbook 1984 (pp. 3–30). Carbondale: Southern Illinois University Press. Scheufele, B. (2008). Das Erklärungsdilemma der Medienwirkungsforschung. Eine Logik zur theoretischen und methodischen Modellierung von Medienwirkungen auf die Meso- und Makro-Ebene [The Dilemma of Media Effects Research. A Logic for Modeling Media Effects on Meso- and Macro-Level Units both in Theoretical und Methodical Respect]. Publizistik, 53(3), 339–361. Schulz, A., & Roessler, P. (2012). The spiral of silence and the Internet: Selection of online content and the perception of the public opinion climate in computer-mediated communication environments. International Journal of Public Opinion Research, 24(3), 346–367. Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunications. New York, NY: Wiley. Spears, R., Lea, M., & Postmes, T. (2000). On side: Purview, problems and prospects. In T. Postmes, R. Spears, M. Lea, & S. Reicher (Eds.), SIDE issues centre stage: Recent developments in studies of de-individuation in groups. Proceedings of the colloquium. Amsterdam: Royal Netherlands Academy of Arts and Sciences Spears, R., Postmes, T., Lea, M., & Watt, S. E. (2001). A SIDE view of social influence. In J. P. Forgas & K. D. Williams (Eds.), Social Influence: Direct and indirect processes (pp. 331–350). Philadelphia, PA: Psychology Press. Tajfel, H. (Ed.). (1978). Differentiation between social groups: Studies in the social psychology of intergroup relations. London, England: Academic Press. Turner, J. C. (1982). Towards a cognitive redefi nition of the social group. In H. Tajfel (Ed.), Social identity and intergroup relations (pp. 15–40). Cambridge, England: Cambridge University Press. Turner, J. C. (1987). A self-categorization theory. In J. C. Turner, M. A. Hoggs, P. J. Oakes, S. D. Reicher, & M. S. Wetherell (Eds.), Rediscovering the social group: A self-categorization theory (pp. 42–67). Oxford, England: Basil Blackwell. Wanta, W., & Dimitrova, D. (2000, June). Chatrooms and the spiral of silence: An examination of online-discussions during the final 1995 U.S. presidential debate. Paper presented to the International Communication Association, Acapulco, Mexico. Yun, G. W., & Park, S.-Y. (2011). Selective posting: Willingness to post a message online. Journal of Computer-Mediated Communication, 16(2), 201–227. Zimbardo, P. G. (1969). The human choice: Individuation, reason, and order versus deindividuation, impulse, and chaos. Nebraska Symposium on Motivation, 17, 237–307.

10 LINKING SOCIAL NETWORK ANALYSIS TO THE SPIRAL OF SILENCE, COORIENTATION, AND POLITICAL DISCUSSION The Intersection of Political Perceptions and Political Communication William P. Eveland, Jr. Scholars across several disciplines, and with varied theoretical orientations, have worked to understand the implication of others’ opinions for the likelihood of political conversation generally, and opinion expression specifically. Public opinion scholars have conducted scores of studies testing aspects of Noelle-Neumann’s (1993) spiral of silence theory. Others have taken a dyadic coorientation approach to opinion perception and interaction (McLeod & Chaffee, 1973). Over the past few decades, political communication scholars have examined the role of political agreement or homogeneity of discussion partners in encouraging or discouraging political discussion (e.g., Huckfeldt & Mendez, 2008; Mutz, 2006). In this chapter, I begin with a very brief summary of each of these areas of research. I follow this by synthesizing work in these diverse and mostly independent literatures, with the goal of theoretical advancement through integration. I pay special attention to at least five advantages of integrating social network concepts and methods as part of this integration: (a) making more explicit micro-to-macro linkages; (b) emphasizing reference groups; (c) observing social isolates; (d) moving beyond hypothetical situations; and (e) contrasting effects of perception and reality. Spiral of Silence The spiral of silence (SOS) considers communication in both mass mediated and interpersonal forms. The fundamental outcome of the SOS is individual-level interpersonal communication—specifically, the willingness to express opinions. Communication is also represented as one of the key causal forces, both in terms of representations of public opinion in the media and in social-level

120 William P. Eveland, Jr.

communication in support of or opposition to particular viewpoints. Thus, media and interpersonal communication trends are the fundamental sources of perceptions of the climate of opinion. These, in conjunction with the quasistatistical sense that facilitates acquisition of this information by individuals, and the fear of social isolation, affect individual willingness to speak out. The SOS as developed by Noelle-Neumann has some particular emphases. First, it addresses opinion expression specifically rather than political talk in general. Second, it considers both conversational as well as other symbolic expressions of opinion (e.g., signs, bumper stickers, buttons) that are often considered forms of political participation. Third, the opinion climate appears to be a national phenomenon rather than dyadic, reference group, or community-based. Fourth, the process of perception of opinion climates need not be observably accurate, nor even a conscious process. Although rarely has the SOS been tested as a whole, some of its key assertions regarding the impact of opinion climates on opinion expression have not been particularly well-supported (see Glynn, Hayes, & Shanahan, 1997). This suggests that some revisions of the SOS theory may be in order. Coorientation McLeod and Chaffee (1972, 1973) developed their coorientation model as an extension of Newcomb’s classic A-B-X model (Newcomb, 1953). Much like the SOS, McLeod and Chaffee summarized coorientation as a communicative approach that moved beyond the persuasion focus of many of the thencontemporary balance theories. The fundamental idea of coorientation is that two individuals—A and B—have some object (X) in a shared perceptual field and thus may be considered to “co-orient” toward it. Moreover, coorientation acknowledges that each individual holds a perception about how the other orients toward the object. Three important concepts for coorientation are: agreement (the degree to which orientations toward the object are shared by A and B), accuracy (the degree to which perceptions about the other match the actual orientation of the other), and congruency (the degree to which perceptions of the other match the orientations of the self ). After a brief flurry of research on coorientation, scholars appear to have— for the most part—set aside coorientation as a theoretical and methodological framework of study. However, the coorientation perspective still has much to offer political communication generally and the SOS specifically. As McLeod and Chaffee foreshadowed a year prior to the English publication of NoelleNeumann’s SOS theory in Journal of Communication in 1974, “Various forms of congruency … may very well determine the amount and forms of communication that occur between persons.” (1973, p. 491).

Linking Social Network Analysis to the Spiral of Silence 121

Political Discussion The literature on political discussion includes among its emphases the extent to which political discussion is broadly deliberative, and more specifically to what extent individuals are exposed to people with whom they disagree in their political conversations or networks (Eveland, Morey, & Hutchens, 2011). Although this research typically does not emphasize the notion of opinion expression as explicitly as does the SOS, nor does it require public discussion, there is clearly an important connection between the SOS and research positing a link between shared political preferences and the likelihood or frequency of political discussion. And, there is considerable evidence for some degree of structuring of political discussion networks on the basis of shared political preferences. Integrating Spiral of Silence, Coorientation, and Political Discussion Research It is our contention that the SOS, coorientation, and political discussion literatures have much to offer one another both theoretically and empirically, and that overlaying a social network approach onto them would further advance our understanding of these fundamental communication processes. But, there is almost no evidence of citation of coorientation research in the contemporary literature on political discussion or the SOS. Political discussion research has on occasion cited the SOS, but it tends to be more reliant on the literatures on homophily in social interaction and deliberative democracy (e.g., Huckfeldt, Johnson, & Sprague, 2004; Mutz, 2006). Therefore, below we compare and contrast these approaches on a few key issues. The coorientation approach emphasizes the difference between agreement and congruency. Scholars in the political communication literature, when they address it at all, have generally considered this in terms of the acceptability of a weaker but readily available perceptual measure (i.e., congruency) or an emphasis on the importance of perception for behavior compared to reality (i.e., agreement). In the SOS, most scholars also have emphasized the importance of perception over reality. However, the distinction is quite important in coorientation both terminologically and theoretically. The same is true for the SOS, about which Noelle-Neumann (2001, p. 60) explicitly argues: “the question of which view is in fact gaining or losing ground among the population is decisive here, not whether the individual also consciously perceives this process.” And, given that the vast literature comparing social reality perceptions with reality is foundational to the study of public opinion processes (see Eveland & Glynn, 2008), empirical efforts at addressing the implications of this inconsistency across the literatures deserves more attention. The distinction between congruency and agreement is intimately tied to the concept of accuracy. Although they theorize about the importance of

122 William P. Eveland, Jr.

congruency in affecting future communication, McLeod and Chaffee (1973, p. 487) emphasize that “Accuracy … seems an ideal criterion for communication in that it is (theoretically, at least) achievable through communication alone.” Given the fundamental recognition that the climate of opinion is determined by mass media and interpersonal communication (Noelle-Neumann, 1993), incorporating a model that explicitly considers these interrelated concepts could be an important theoretical advance in SOS research. Another key distinction is that the SOS emphasizes public opinion expression (including public displays such as signs and bumper stickers) and societal level opinion perceptions, whereas both coorientation and political discussion research emphasize perceptions within specific social relationships and communication in mostly private settings. Integrating both approaches would seem to be sensible, especially given research revealing the interplay between personal network and larger community-level opinion climates (Huckfeldt, Beck, Dalton, & Levine, 1995; McClurg, 2006). The Contribution of Social Network Analysis to the Spiral of Silence Social networks are, from our perspective, fundamentally communicative in both their structure as well as the process through which they produce effects, and they are highly relevant to work on the SOS, coorientation, and political discussion (see also, Eveland, Hutchens, & Morey, 2012). Nonetheless, to begin we must distinguish the typical studies of the SOS from dyadic coorientation and political discussion studies to illustrate the value of a social network data. A visual representation is presented in Figure 10.1. Spiral of Silence Survey Studies In their review of SOS research, Scheufele and Moy (2000, p. 15) argue that survey data are “the appropriate method for data collection for evaluating the willingness to speak out.” They base this claim on the weak external validity of experimental studies. However, they are not explicit regarding what sorts of surveys are to be favored. The vast majority of SOS survey studies are based on cross-sectional sample data and typically employ self-reports of willingness to speak out in hypothetical situations (see Scheufele, Shanahan, & Lee, 2001, for a review). As such, responses are to questions about future opinion expression in either a specific or possibly abstract social setting, typically with either a vague group (i.e., “friends”) or with a stranger. Therefore, SOS research tends to dissociate the opinion climate from the specifics of the interaction partners, such as the nature of the relationship (e.g., intimacy, length, future potential, power differentials). The political discussion literature suggests there are meaningful consequences of this lack of attention to relationship type when assessing avoidance of political disagreement (Morey, Eveland, & Hutchens, 2012). Although

Linking Social Network Analysis to the Spiral of Silence 123

Spiral o f Silence

Ovadic Coorienulion

Cli ma leo! Opinion

A

c

B

D

Eorocen l ric Network St udv B A

E

D

, ,

F

N

C

G

"

M

A

B

C

D

E

F

G

" Social Network Analvsls

D

,

FIGURE 10.1 Comparing data structures across research approaches.

there is variation in emphasis on perceptions of national, more local, and even reference group opinion climates (see Glynn & Park, 1997; Hayes, Shanahan, & Glynn, 2001; Oshagan, 1996; Scheufele & Eveland, 2001), the theory as developed by Noelle-Neumann emphasizes national climates of opinion. Dyadic Coorientation Studies Coorientation studies typically seek to identify pairs of individuals involved in some sort of existing dyadic relationship, such as parent-child, husband-wife, or friend-friend. Corresponding interviews—including asking dyads about their perception of their partner’s beliefs or opinions—are conducted with each member of the dyad. Measurement typically follows McLeod and Chaffee’s (1973, p. 484) coorientation measurement model. The four indicators—A and B’s cognitions about X, and A and B’s perceptions about each other’s cognitions about X—are employed to construct at least five important measures for coorientation: agreement, accuracy (separately for A and B), and congruency (separately for A and B). In coorientation studies, accuracy and agreement are in reference to a single dyad, not an overall climate of opinion in a community or nation. Political Discussion: Egocentric and Snowball Network Studies. Egocentric network data are relatively common in the study of political discussion. Using

124 William P. Eveland, Jr.

egocentric network data, respondents are asked to identify three to six specific members of their political networks through a “name generator.” Then, communication ties with these network members (e.g., how often the ego discusses politics with each alter) are assessed individually (e.g., Mutz, 2002; Nir, 2005; Richey, 2009), although it is uncommon to ask specifically about opinion expression on any particular political issue. Respondents are then asked to provide information about their named network alters, such as their party affi liation or candidate preferences, and so forth, through “name interpreters.” Egocentric network data are only suitable for the assessment of the congruency concept from coorientation because that concept is an intra-individual rather than interpersonal concept (McLeod & Chaffee, 1972). Therefore, references to terms such as agreement and disagreement in the literature using egocentric data typically correspond with the coorientation concept of congruency rather than the coorientation concept of agreement. Accuracy cannot be assessed with egocentric data. Snowball sample data are an extension of egocentric studies in which data are collected directly from the three to six discussion partners who are identified by the egocentric survey respondent as part of the name generator process. Data on the characteristics of network alters are then gathered directly from these alters (e.g., Huckfeldt & Sprague, 1995), and so full coorientation measurement (including agreement and accuracy) is possible with these data. Expanded Possibilities Using Whole Network Data. When I refer to the value of social network data throughout this chapter, I am referring specifically to “whole network” social network data. With whole network data, a social network is typically defined by a researcher using some existing (and sometimes relatively arbitrary) boundary specifications. For instance, researchers may use a list of employees in a small business, students in a given school, members of a volunteer organization, participants in an online discussion forum, or any other group with a clear boundary and a complete list of members. Using terminology from the SOS literature, these may be thought of as “reference groups” (Oshagan, 1996). Researchers gather data directly from each member of the network, including characteristics of the members and their communication ties (but usually not specifics about opinion expression) to all other members of the group. Thus, whole network data include individual-level data from all members of a predefi ned network or reference group, and each network member’s ties to all other network members, all collected directly from each member. In this way, researchers can see beyond the direct ties of a randomlyselected survey respondent (egocentric and snowball data) to the larger network structure. Whole network data can be restructured to provide information corresponding to that gathered via egocentric, snowball, and coorientation studies (see Marsden, 2002). So, how can the integration of SOS research with coorientation and political

Linking Social Network Analysis to the Spiral of Silence 125

discussion research be advanced through the use of whole network data and analysis principles? Making More Explicit Micro-to-Macro Linkages. The SOS is a multi-level theory of communication and public opinion, but many of the cross-level linkages need to be more clearly specified (Pan & McLeod, 1991). Social network analysis provides one means of explicitly testing the construction of perceived opinion climates by examining whether or not dyadic coorientation assessments can be aggregated to reflect overall opinion climate assessments (see Figure 10.2). That is, if interpersonal expressions of views are in fact helping to produce the opinion climate as Noelle-Neumann assumes, individuals must be privy to these conversations in some way, probably by being a part of them or at least observing them in social settings. Social network analysis that assesses coorientation can identify perceived opinion for each discussion partner, and can construct an aggregated measure across discussion partners that can be correlated with an overall perception of the opinion climate. Moreover, network characteristics such as density can be tested as moderators of reference group influence, such that the spiral process may differ by group depending on how cohesive it is or the role it plays in the lives of its members. Emphasizing Reference Groups. Since social network analysis must clearly defi ne the “network,” it seems to be a useful technique especially for the study of specific reference groups such as voluntary associations, workplaces, and family groups—all of which are environments for political opinion expression (see Eveland & Kleinman, 2013). By emphasizing communication and opinion perceptions at the level of reference group rather than the larger social system, it permits assessment of interpersonal and small group communication and influence processes (which serve as an important foundation of the SOS; see Price & Allen, 1990) and permits a direct comparison of the effects of reference groups with the effects of the overall climate of opinion. Moreover, dyadic data would encourage explicit consideration of variation in opinion expression

Grotpl

Group 2

Group 3

Individuals ABC

0

Group4

E

GroupS

Group 6

F

Aggregation Dyadic Kelationships

c,a

c,b

c,d

c,e

c,f

FIGURE 10.2 Multilevel Spiral of Silence Research Using Whole Network Data.

126 William P. Eveland, Jr.

across individuals depending on the viewpoints of shared network ties. For example, it is possible that, even if A and B share preferences, an expectation that opinions revealed in an interaction between them could be shared with C, with whom both individuals had social ties and who holds a different view, could scuttle any opinion expression between A and B. Observing Social Isolates. One of the advantages of social network analysis research is the ability to identify social isolates within a group setting. Given that a key concept of the SOS is the fear of social isolation (Hayes, Matthes, & Eveland, 2013), studying social isolates would seem to be an important component of SOS research. Are social isolates the hard core or avant-garde described by Noelle-Neumann? That is, are those who are willing to express opinions against the force of public opinion actually isolated? Do they have particular characteristics that distinguish them psychologically from others? These are interesting questions that have not been sufficiently addressed within the SOS literature, but could be using social network data. Moving Beyond Hypothetical Situations. Social network data move beyond hypothetical situations for opinion expression; they emphasize the “reality” of prior communication. However, gathering measures of the social network over time allows us to see changes in communication patterns that may form as network members learn about the consistency or inconsistency of viewpoints of specific others (e.g., Lazer, Rubineau, Chetkovich, Katz, & Neblo, 2010). The ability of panel data to help distinguish selectivity from influence processes—or to understand their mutual contributions—also would address recent discussions of how best to tap “reinforcing spiral” processes in communication effects research (Slater, 2007). Moreover, from a measurement perspective, social network data allow a certain amount of verification in that communication ties may be tested for reciprocality. That is, not only can we ask A if she expressed a given opinion to B, but we can also ask B if A had expressed that opinion. This can combat concerns that individuals will respond to our questions based at least in part on social desirability. Contrasting Effects of Congruency and Agreement. Social network data, like coorientation data and snowball sample data, can allow us to consider the relative impact of agreement (objective minority vs. majority status) compared to congruency (perceived minority vs. majority status) on speaking out. Given that this has been an issue of contention—Noelle-Neumann (2001) argues that it is not subjective but objective status that matters, whereas Mutz (2006, p. 25) argues that perceptions rule—a research design which permits a direct comparison would seem to have a significant advantage.

Linking Social Network Analysis to the Spiral of Silence 127

Conclusion Despite being a popular and intuitively appealing theory, the SOS has received at least its fair share of theoretical and methodological critiques over the years. Moreover, concerns about the predictive strength of the theory remain prominent (Glynn et al., 1997; Glynn & Huge, Chapter 6, this volume). I argue that a synthesis of the relatively distinct literatures on political discussion, coorientation, and the SOS using social network analysis methods and concepts as a foundation may offer one path forward for SOS proponents. It would also help to integrate conceptually closely related (but citationally quite distinct) literatures of relevance to political communication and public opinion scholars. Acknowledgment The author would like to thank Myiah J. Hutchens and Carroll J. Glynn for their input on an earlier version of this chapter. References Eveland, W. P., Jr., & Glynn, C. J. (2008). Theories on the perception of social reality. In W. Donsbach & M. Traugott (Eds.), The SAGE handbook of public opinion research (pp. 155–163). Thousand Oaks, CA: Sage. Eveland, W. P., Jr., Hutchens, M. J., & Morey, A. C. (2012). Social networks and political knowledge. In H. A. Semetko & M. Scammell (Eds.), The SAGE handbook of political communication (pp. 241–252). Thousand Oaks, CA: Sage. Eveland, W. P., Jr., & Kleinman, S. B. (2013). Comparing general and political discussion networks within voluntary organizations using social network analysis. Political Behavior, 35, 65–87. Eveland, W. P., Jr., Morey, A. C., & Hutchens, M. J. (2011). Beyond deliberation: New directions for the study of informal political conversation from a communication perspective. Journal of Communication, 61, 1082–1103. Glynn, C. J., Hayes, A. F., & Shanahan, J. (1997). Perceived support for one’s opinions and willingness to speak out: A meta-analysis of survey studies on the “spiral of silence.” Public Opinion Quarterly, 61, 452–463. Glynn, C. J., & Park, E. (1997). Reference groups, opinion intensity, and public opinion expression. International Journal of Public Opinion Research, 9, 213–232. Hayes, A. F., Matthes, J., & Eveland, W. P., Jr. (2013). Stimulating the quasi-statistical organ: Fear of social isolation motivates the quest for knowledge of the opinion climate. Communication Research, 40, 439–462. Hayes, A. F., Shanahan, J., & Glynn, C. J. (2001). Willingness to express one’s opinion in a realistic situation as a function of perceived support for that opinion. International Journal of Public Opinion Research, 13, 45–58. Huckfeldt, R., Beck, P. A., Dalton, R. J., & Levine, J. (1995). Political environments, cohesive social groups, and the communication of public opinion. American Journal of Political Science, 39, 1025–1054. Huckfeldt, R., Johnson, P. E., & Sprague, J. (2004). Political disagreement: The survival of diverse opinions within communication networks. New York, NY: Cambridge University Press. Huckfeldt, R., & Mendez, J. M. (2008). Moths, fl ames, and political engagement: Managing disagreement within communication networks. Journal of Politics, 70, 83–96.

128 William P. Eveland, Jr.

Huckfeldt, R., & Sprague, J. (1995). Citizens, politics, and social communication: Information and influence in an election campaign. New York, NY: Cambridge University Press. Lazer, D., Rubineau, B., Chetkovich, C., Katz, N., & Neblo, M. (2010). The coevolution of networks and political attitudes. Political Communication, 27, 248–274. Marsden, P. V. (2002). Egocentric and sociocentric measures of network centrality. Social Networks, 24, 407–422. McClurg, S. D. (2006). Political disagreement in context: The conditional effect of neighborhood context, disagreement and political talk on electoral participation. Political Behavior, 28, 349–366. McLeod, J. M., & Chaffee, S. R. (1972). The construction of social reality. In J. T. Tedeschi (Ed.), The social influence processes (pp. 50–99). Chicago, IL: Aldine. McLeod, J. M., & Chaffee, S. R. (1973). Interpersonal approaches to communication research. American Behavioral Scientist, 16, 469–499. Morey, A. C., Eveland, W. P., Jr., & Hutchens, M. J. (2012). The “who” matters: Types of interpersonal relationships and avoidance of political disagreement. Political Communication, 29, 86–103. Mutz, D. C. (2002). Cross-cutting social networks: Testing democratic theory in practice. American Political Science Review, 96, 111–126. Mutz, D. C. (2006). Hearing the other side: Deliberative versus participatory democracy. New York, NY: Cambridge University Press. Newcomb, T. M. (1953). An approach to the study of communicative acts. Psychological Review, 60, 393–404. Nir, L. (2005). Ambivalent social networks and their consequences for participation. International Journal of Public Opinion Research, 17, 422–442. Noelle-Neumann, E. (1993). The spiral of silence: Public opinion — Our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (2001). Commentary. International Journal of Public Opinion Research, 13, 59–60. Oshagan, H. (1996). Reference group influence on opinion expression. International Journal of Public Opinion Research, 8, 335–354. Pan, Z., & McLeod, J. M. (1991). Multilevel analysis in mass communication research. Communication Research, 18, 140–173. Price, V., & Allen, S. (1990). Opinion spirals, silent and otherwise: Applying small-group research to public opinion phenomena. Communication Research, 17, 369–392. Richey, S. (2009). Hierarchy in political discussion. Political Communication, 26, 137–152. Scheufele, D. A., & Eveland, W. P., Jr. (2001). Perceptions of ‘public opinion’ and ‘public’ opinion expression. International Journal of Public Opinion Research, 13, 25–44. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12, 3–28. Scheufele, D. A., Shanahan, J., & Lee, E. (2001). Real talk: Manipulating the dependent variable in spiral of silence research. Communication Research, 28, 304–324. Slater, M. D. (2007). Reinforcing spirals: The mutual influence of media selectivity and media effects and their impact on individual behavior and social identity. Communication Theory, 17, 281–303.

PART III

Allied Research Traditions

This page intentionally left blank

11 SOCIAL NORMS, SPIRALS OF SILENCE AND FRAMING THEORY An Argument for Considering Cross-Cultural Differences in Media Effects Research Timothy K. F. Fung and Dietram A. Scheufele

Some 40 years have passed since Noelle-Neumann fi rst articulated her spiral of silence theory (1974), which has inspired a considerable amount of research and is regarded as one of the most influential theories in public opinion and mass communication. Since then, the field of media effects has experienced significant changes. Bryant and Miron (2004) surveyed the state of mass communication theory and research using a content analysis of the major journals in the field. Their fi ndings identified framing theory as a model that is rapidly gaining ground and receiving more and more scholarly attention in the field. The growing body of research on framing signifies a paradigm shift in media effect research from a powerful effect model, which is exemplified by NoelleNeumann’s spiral of silence theory, to a negotiated effects model, which is represented—among other theories—by framing. Although the spiral of silence and framing effect theories fall under different research paradigms, both share the idea that socially negotiated norms or schemas are critical for explaining the influence of socio-cultural environments on individual responses to opinion disagreement and on mediated message interpretation. That is, we argue that the micro-level variations in opinion disagreement, in the spiral of silence, and in mediated messages interpretation, in framing effect theory, across groups are accounted by the same macro-level, social norm variable. In this sense, both models attribute an important role to social norms, as a contextual factor, in shaping our behavior at the aggregate level. In this chapter, we highlight the importance of social norms, a macro-level variable, across different paradigms and models in media effect research and their implications particularly for cross-cultural research. Because social norms vary from one culture to another, media effects, particularly in the case of the

132

Timothy K. F. Fung and Dietram A. Scheufele

spiral of silence and framing effects, are likely to vary across cultural boundaries. To this end, this chapter is divided into three parts. First, we very briefly review the evolution of media effect paradigms, from the inception of the spiral of silence theory to the more recent framing effect theory. Second, we discuss how, from a cross macro-micro theoretical perspective (Merton, 1968), the premises of the spiral of silence theory and the framing effect theory share the basis of social norms as contextual influence on individual behavior at group level, and how social norms, a macro-level variable, may modify the effects of the spiral of silence and framing. Finally, based on the fi ndings of cross-cultural comparisons on the spiral of silence and framing effect theories, we conclude by calling for the differentiation between universal and culturally specific effects in our field. From Return to the Concept of Powerful Mass Media to Negotiated Effects The quest for media effects has undergone a number of phases throughout the 20th century (for a review, see McQuail, 2010). Noelle-Neumann’s (1974) work on the spiral of silence revived the notion of powerful media effects in the field. Under the powerful media paradigm, mass media are considered to exert robust, direct, and long-term effects on audience’s cognition because media audiences are susceptible to pervasive and unified media content, over time (Noelle-Neumann, 1993). Since Noelle-Neumann’s call a return to the powerful effect model, media effects research has evolved and morphed into the so-called negotiated effect model (McQuail, 2010). The major distinctions of the negotiated effect model, which set it apart from the powerful effect model, are the recognition of the active role of the audience in the meaning-making process and the joint construction of (message) meaning between the media and the audience (Gamson, Croteau, Hoynes, & Sasson, 1992; McQuail, 2010; Potter, 2009). Because of selective representation, media can be influential in defi ning social phenomena. Audience members, however, can decide how to interpret the social phenomenon and whether or not to adopt the media representation. The effect of media, consequently, is contingent on the characteristics of audience members and their social milieu. Framing effects, as one example of a negotiated effects model, goes back to research in sociology and psychology over almost four decades (Tewksbury & Scheufele, 2009). More recent converging conceptualizations of framing connect both news production and news consumption processes (Scheufele & Tewksbury, 2007). As defi ned by Price and Tewksbury (1997), a framing effect is a process in which altering salient attributes of a media message, such as selection of information and thematic structure, renders thoughts applicable, resulting in their activation and use in evaluation.

Social Norms, Spirals of Silence and Framing Theory 133

Social Norms: The More Things Change, the More They Stay the Same As discussed earlier, the paradigm shift from the powerful effect model to the negotiated effect model reflects a major conceptual change of media effects in the discipline. Despite the changes in media effects research and the differences between framing and the spiral of silence, social norms as the contextual influence of opinion formation and behavior remains the theoretical basis underlying these two theories across micro and macro levels of analysis. That is, social norms either shape how we, in a spiral of silence, react to disagreement or how we, in a framing effect, react to mediated messages at the aggregate level. Both the spiral of silence and framing share the notion that individuals are situated in a social milieu and that social surroundings have an influential contextual impact on individual perception, opinion, and behavior. Social norms, at the collective level, can broadly be defi ned as shared customary rules or conventions, such as values, standard, traditions, customs, and rituals governing the behavior of a group or a society (Pepitone, 1976; Sherif, 1936). Social norms are critical for the functioning of a society, because they direct the behaviors of the members in the society to maintain social order (Scott & Marshall, 2009; Tittle, 1977). Some scholars (Bicchieri, 2006) refer to social norms as the “grammar of a society.” As social norms are rarely codified formally in a society, individuals tend to interpret social norms based on their observations of others’ behavior. Such a personal interpretation, accurate or not, forms an individual perception of social norms (i.e., perceived norms) (Lapinski & Rimal, 2005). In essence, the notion of social norms suggests that, particularly in novel or ambiguous situations, individuals tend to look at others in their social surroundings to make sense of the socially approved behaviors and the perceived prevalent behaviors so that they function as a guide to keep individual behavior in line with the majority (Cialdini & Trost, 1998). Social norms guide us how to behave as a group by specifying prescribed or proscribed conducts in a given situation (Lapinski & Rimal, 2005), i.e., they serve as a behavior-guiding force because of their informational (descriptive) and regulatory (injunctive) functions (Cialdini & Trost, 1998) and as a source of information about what most people do (or do not do) in a given context, particularly in unfamiliar and ambiguous situations (Cialdini & Trost, 1998). In this sense, social norms provide us with frames of references, in a social encounter (Newcomb, 1958; Sherif, 1936), to defi ne the meaning of a given situation and to understand specific events in that situation (Cialdini & Trost, 1998; Fazio, 1990). Research (e.g., Milgram, Leonard, & Berkowitz, 1969) provides empirical evidence on how the behavior of other people in our social surrounding influences our interpretation of and responses to ambiguous situations. Because social norms exemplify the common behaviors practiced and approved by most people in a group, the acts of the majority constitute a

134

Timothy K. F. Fung and Dietram A. Scheufele

pressure to regulate individual behavior. Such a pressure on individual behavior enforces the social norms and usually operates in the form of social sanctions, in case any group members violate the social norms (Lapinski & Rimal, 2005). The social sanctions form coercive social influence to drive individual members to conform to the majority. In this sense, social norms can mold the behavior of the members of a society through social sanctions. Social Norms and the Spiral of Silence Theory In the spiral of silence, the premises of the theory are primarily built upon the notion of social norms (Shulman & Levine, 2012). In explaining the public opinion formation process, Noelle-Neumann (1974) proposed two key hypotheses in her spiral of silence theory. One, namely the perception hypothesis, postulated that individuals are likely to scrutinize their social environment for the climate of opinion and its trend regarding a controversial issue. In this stage, media exert a strong influence on their perceptions of public opinion. The other hypothesis, namely the conformity hypothesis, posited that, driven by sanction fear, people are unlikely to express their personal views on the issue when they perceive their opinion unpopular after weighing it against the perceived opinion climate. The result of this tendency is a spiraling process in which minority views are perceived as increasingly weak and their supporters eventually fall silent, establishing the majority view as a social norm (Scheufele & Moy, 2000). The notion of social norms is therefore a recurring theme in the spiral of silence theory. In explaining the concept of public opinion, Noelle-Neumann (1983) borrowed from the work of Ross (1969) and Mead (1937) to claim that public opinion is an effective means of reaching a decision because it is flexible and cost-saving. This argument is consistent with the idea that social norms function as a “social proof ” heuristic (Cialdini & Trost, 1998; Lapinski & Rimal, 2005). Thibaut and Kelley (1959) argued that social norms provide normative information for individuals to follow, because the larger the crowd that engages in the same behavior in the same situation, the more “correct” we will perceive the behavior to be. In other words, social norms represent a consensus of social endorsement of a behavior in a given situation. As such, imitating the crowd aids our decision making by reducing our time and cognitive effort with a widely desirable outcome. In a similar vein, the regulatory nature of social norms is the basis of the conformity hypothesis in the spiral of silence theory. In particular, public opinion functions as a guardian of morality and tradition in a society (NoelleNeumann, 1983). Social norms specify what the members in the group “should” do (Cialdini, Kallgren, & Reno, 1991) and create expectations about potential sanctions on those who deviate from the group standard (Newcomb, 1943). The punishment that is enforced by the majority in the group behind the social norms motivates individuals to comply (Horne, 2001). Social norms, therefore,

Social Norms, Spirals of Silence and Framing Theory 135

are also a crucial tool for societies to build agreement on values and goals so as to reach collective cohesion (Noelle-Neumann, 1991). This is particularly crucial when individual and group interests are in confl ict such as opposing views in controversial issues (Neuwirth, Frederick, & Mayo, 2007). Social norms at the collective level, however, are culturally bounded (Berger & Luckmann, 1966; Hechter & Opp, 2001). One behavior, in a given situation, that is regarded as normal in one culture may be regarded as deviant in another culture (e.g., Hersh, 1970). Every culture, thus, has its own set of norms. As such, cross-cultural comparative research on the spiral of silence is important because it allows for examining the impact of the sociocultural environment on individual behavior such as reaction to opinion divergence and expression (Blumer, McLeod, & Rosengren, 1992). As argued by Scheufele and Moy (2000), confl ict management style is relevant for comparing the spiral of silence effect across cultures. Cultures, such as Japan, that place an emphasis on the value of respect for hierarchy and homogeneity, incline a more collaborative approach when facing opinion disagreement. However, cultures, such as the United States, that stress the value of individual freedom and achievement, favor a competitive approach. Cultures, who emphasize diversity, lean toward a more compromising approach when dealing with disagreement. In response to Scheufele and Moy’s (2000) call for a cross-cultural comparison of the spiral of silence effect, three studies were conducted over the past decade. Lee, Detenber, Willnat, Aday, and Graf (2004) argued that the spiral of silence effect would differ between Americans and Singaporeans because of cross-cultural differences in the independence-interdependence self-concept. As an individualistic society, the self-concept of Americans is independent. That is, Americans consider themselves as unique individuals and favor acting independently. On the contrary, in Singapore, as a collectivistic society, the self-concept of Singaporean is interdependent. They more strongly defi ne themselves in terms of their membership in social groups. As such, Singaporeans are more likely to experience fear of isolation than their American counterparts when they perceive their opinion is in the minority. The fi ndings showed that the spiral of silence effect was found in Singapore but not in the United States. Drawn upon the same individualistic-collectivistic differences, Huang (2005) tested the spiral of silence theory in the United States and Taiwan based on the argument of the self-other conceptions. As expected, the fi ndings provided support for the spiral of silence hypothesis in Taiwan, but not in the United States. However, as shown in her study, fear of isolation was not associated with a willingness to speak out. She argued that, rather than being afraid of ostracism, the conformity of the Taiwanese was motivated by maintaining social harmony, which is a prevailing norm in Chinese culture. The most recent study (Spencer & Croucher, 2008) examined the spiral of silence theory in Spain and France. Using Basque nationalism as the case study,

136

Timothy K. F. Fung and Dietram A. Scheufele

Spencer and Croucher argued that respondents who were geographically and culturally close to the Basque homeland would refrain to express their opinion when they perceived their opinion unpopular because of differences in their history of nationalism and their opinion about nationhood. As predicted, the spiral of silence effect varied across geographical and cultural boundaries. In sum, the three cross-cultural studies suggest that, rather than a universal effect across cultures, the spiral of silence effect is likely to be bounded by cultural norms, although more studies are needed to fully support such a proposition. Social Norms and Framing Effect Theory The framing effect is defined as “a dynamic, circumstantially-bound process of opinion formation in which the prevailing modes of presentation in elite rhetoric and news media coverage shape mass opinion” (Scheufele & Iyengar, forthcoming). The framing effect is resultant from a dynamic interaction between media messages and individual schema. Price and Tewksbury (1997) proposed the applicability model of framing effects to explain the underlying cognitive process of how media frames exert influence on individuals’ interpretation and understanding of an issue as well as on their attitude and judgment toward a social issue. In essence, by selecting, ignoring, emphasizing, and downplaying certain aspects of the social issue in the media message, the media frame renders relevant processing schemas from the mind of the audience for activation to interpret and form opinions about the message (Price & Tewksbury, 1997). Frequently, framing effects can refer to the process of frame-setting or the process of an “individual-level outcome of framing,” or both (Scheufele, 1999). As noted by Scheufele (2008), framing is a multilevel construct involving news production and news consumption. Therefore, it is impossible to explicate how the framing effect theory relates to the notion of social norms without an understanding of the full process model of framing. Below is a brief review of the model. According to Scheufele (2000), the concept of framing involves three distinct processes: frame-setting, frame-building, and an individual-level outcomes of framing. (a) Frame-building refers to the process of how news is being selected and produced. Previous research (e.g., Shoemaker & Reese, 1996) revealed at least five forces influencing how journalists frame a given issue: social norms and values, organizational constraints, pressure from political elites and interest groups, journalistic routines, and ideological and political orientation of journalists. (b) Frame-setting refers to the process of how a frame transfers from media outlets to audience, i.e., the process of how the media frame influences individual interpretive schema for processing. (c) Individual-level outcome of framing refers to the process of how individual interpretive schema, activated by the media frame, influences issue interpretation and opinion formation. Similar to the spiral of silence theory, the notion of social norms as a contextual influence on individual behavior at the meso or group level is also embedded

Social Norms, Spirals of Silence and Framing Theory 137

in framing effects theory. Specifically, the informational nature of social norms is evident in the form of “frame of reference” in the three processes of framing. At the frame-building stage, social norms, such as values, beliefs, political attitudes, ideology, and common practices of their profession, provide a frame of reference for journalists to make sense of a social event and construct the news story (Shoemaker & Reese, 1996). For example, in their comparative content analysis between the U.S. and Chinese national news, Chang, Wang, and Chen (1998) concluded: “The social construction of reality perspective suggests that the selection and presentation of news on the two networks [ABC News and CCTV] depend not so much on the properties of the event or issue itself, but rather on its position in the broader social structure relative to its external context” (p. 294). In addition, because journalists have to deal with a tremendous amount of information in a short period of time, on a daily basis, media routines, such as audience appeal, affords a frame of reference for journalists to make sense of unfamiliar and separate social events (Dunwoody, 1992). Through organizational constraints, such as its economic goals and the gatekeeping role of news editors, corporate ownership, and organizational structure have set a frame of reference for journalists to guide news production (Bantz, 1997; Breed, 1997; Shoemaker & Reese, 1996). Finally, external sources, such as political elites and officials and interest groups that provide and withhold information about a social event in press conferences, establish a frame of reference for journalists to direct their perception of the social event (Bennett, 1990; Bennett, Lawrence, & Livingstone, 2007; Dunwoody, 1997). At the frame-setting stage, a frame in a media message provides a frame of reference for audiences to make sense of a given issue. At the individual-level outcomes of framing stage, Goff man (1974) posited that individuals cannot fully understand the world because of its complexity; as a result, individuals actively organize and construe their daily experiences to make sense of the world. By doing so, individuals draw upon their interpretive schema (“primary frameworks” is Goff man’s term). This interpretive schema helps individuals to “locate, perceive, identify, and label” events and actions (Goff man, 1974, p. 21). The preexisting interpretive schema, which the individuals applied, serves as a frame of reference for the individuals to process the media message. It also serves as a heuristic cue for opinion formation (Fiske & Taylor, 1991). For instance, past studies on value-framing (e.g., Shen & Edwards, 2005) showed that people brought in their core values as the interpretive schema to make sense of the welfare policy. In Fung, Brossard, and Ng’s study (2011), individuals drew upon their ecological worldview to understand flood control policy. Core values represent what the mode of conduct or end-states of existence is considered desirable by most people in a group (Rokeach, 1973) and worldviews are the majority’s conception of how the world works and human’s place in the universe (Sire, 1976). Therefore, core values and worldviews function as frame of references for individuals to make sense of social issues.

138

Timothy K. F. Fung and Dietram A. Scheufele

The regulatory nature of social norms is also evident in framing effect theory. Social norms, in the form of media framing, instruct us the way in which we should interpret media messages. At the frame-building stage, political elites and interest groups acknowledge how framing an issue strategically may influence the public’s understanding and opinion on a social issue and take advantage of the power of framing for their own benefit simply by subtle changes in issue presentation (e.g., Luntz, 2008). Their primary goal is to manufacture consensus through strategic framing (e.g., Fung et al., 2011). At the frame-setting stage, as noted earlier, media frames set the boundary for audience members to conceive an issue by selecting, ignoring, emphasizing, and downplaying some aspects of the issue. That is, our interpretation of an issue is a function of which interpretative schema is being applied for processing (Scheufele, 2008). As such, media frames direct audiences to what should be considered (Druckman, 2001; Sniderman, Brody, & Tetlock, 1991; Sniderman & Theriault, 2004). At the individual-level outcome stage, the activated interpretive schema for processing governs what audience members’ opinions and actions should be. The interpretive schema being applied for processing promotes a future course of action (Benford & Snow, 2000). Past studies have showed, for example, that norms such as core values and worldviews, triggered by media frames, play a regulatory function in guiding individual attitudes and behaviors. As said earlier, core values denote what mode of conduct should be chosen or end-states of existence should be pursued in a group (Rokeach, 1973). Ecological worldviews specify how human beings should relate to nature (Kluckhohn & Strodtbeck, 1961). Therefore, core values and ecological worldviews direct individuals to support (oppose) particular policies (Shen & Edwards, 2005), or to choose one policy option over another (Fung et al., 2011). Framing effects are usually modeled as cognitive mechanisms (Cappella & Jamieson, 1997), that is, a news frame renders relevant cognitive schema activation for processing in the memory network of an individual. This begs the question if and how such cognitive conceptualizations are compatible with our proposed norm-based explanations of framing effects. The answer lies in a multilevel analytic framework of mass communication research (Pan & McLeod, 1991) that models communication as not only involving within-level relationships but also cross-level relationships, such as macro-micro-level effects. The spiral of silence, for instance, views individual opinion expression as a function of individual-level perceptions of opinion climate is specified at the micro-individual level. However, the theory also links those individual-level perceptions to majority perceptions at the macro level and ultimately to the establishment of social norms at the societal level. Similarly, framing theory involves micro-individual mechanisms as well as cross-level processes. More specifically, the premise of individual issue interpretation as a function of the resonance between the news frame and individual

Social Norms, Spirals of Silence and Framing Theory 139

schema is formulated at the micro level. How these schemas develop or are activated, however, shifts the focus to the macro level. In particular, social norms within particular groups or societies help shape the content and structure of individuals’ memory networks in those social groups. As a result, different social norms across groups also cause differential schema activation by the news frame for processing as cross-level macro-micro mechanism. Our proposed norm-based account therefore suggests that macro- and micro-level connections and their impacts on people’s individual-level schemas are increasingly crucial to understanding framing effect. Or, to put it differently, normbased explanations can strengthen our understanding of framing effect on how a macro-level variable (i.e., social norms) may account for micro-level variations across social groups (i.e. different schema activation across social groups). To further explicate the relationship between the two theoretical accounts of framing effect under the multilevel framework, we draw upon the Shen and Edwards’s (2005) value-framing study for illustration. In their study, a welfare policy was either framed in terms of the individualism value or humanitarianism value. At the micro-individual level, the activation of correspondent value schema for the welfare policy interpretation depends on the degree of applicability between the value schema and the value frame. At the cross-level macro-micro level, social norms may shape the content and structure of network memory where one aggregate group of individuals’ network memory has a more well-developed individualism value schema and that of the other group has a more elaborated humanitarianism value schema. As a result, social norms mold different networks of memory across groups, leading to different value schema activation across groups for the welfare policy interpretation. In sum, the informational and regulatory nature of social norms is embedded in the cross-level macro-micro process of framing. The way in which the notion of social norms aligns the framing effect theory with the spiral of silence theory is that it considers that our sociocultural environment governs the way we interpret mediated messages and form opinions about controversial issues at the meso level. Specifically, in the spiral of silence theory, norms, in the form of what most people think and do, direct us with regards to how to respond to disagreement. In the similar vein, core values are the specific mode of conduct preferable to most people in a group and ecological worldview is the specific mode of human-nature relationship desirable to most people in a group. These more specific sets of norms, in the framing effect theory, influence us with regards to how to respond to mediated messages. Cultural Norms and Interpretation of Mediated Messages Given the shared notion of social norms with the spiral of silence theory and the evidence, in the cross-cultural studies, that the effect of the spiral of silence is, to some extent, the product of cultural norms, it is reasonable to expect that

140 Timothy K. F. Fung and Dietram A. Scheufele

framing effects may vary across cultures. News production and news consumption do not take place in a vacuum, but rather in a sociocultural environment, which will, in turn, likely influence the news production and consumption processes. According to van Gorp (2007), for example, “a shared repertoire of frames in culture provides the linkage between news production and news consumption” (p. 61). Surprisingly though, relatively little research has been devoted to examine how cultural norms may account for framing effects. A more macroscopic focus in framing effect research is therefore important, not only because of its contribution to theory building but also to broaden its theoretical explanation by linking individual level behavior, such as opinion formation, to marco-level factors such as cultural norms. A recent study (Fung, 2010) attempted to address the gap by examining monocultural individuals in China and the United States and bicultural individuals in Hong Kong. Specifically, Fung conducted two experimental studies to explore how framing effects vary across and within cultures. In the fi rst experiment, Fung presented a news article regarding an elderly care policy to Beijing Chinese and Anglo-Americans. The elderly care policy was either emphasizing the fi lial piety value (value frame), a Confucian value regarding parent-child relationship, or was not emphasizing a particular value (control condition). The fi ndings revealed that Beijing Chinese were more likely to support the elderly care policy, generated greater thought complexity on the issue, and assigned greater responsibility of elderly care to adult children than the Anglo-Americans did. The underlying reason for accounting the framing effect across cultures is the differences in the availability of cultural value schema between Beijing Chinese and Anglo-Americans. In the second experiment, Fung tested the framing effect with Hong Kong Chinese, who incorporate both the Confucian value system and the American value system in their minds. To activate one of the value systems to become more accessible for processing, the respondents were either primed with Chinese or American cultural icons. After that, respondents were presented with a news article regarding the elderly care policy either emphasizing the fi lial piety value or not emphasizing any particular value. The fi ndings indicated that in taking care of aged parents, Hong Kong Chinese primed with Chinese cultural icons assigned more responsibility to adult children and less responsibility to parents themselves than those primed with American cultural icons did. The reason for such variation in the framing effect is the resonance between the media frame and the cultural value system being applied for processing. Conclusion: Call for Differentiating Universal Versus Culturally Specific Effects By examining the differences in conceptual changes in media effect paradigms and similarities in the theoretical foundation between the spiral of silence and

Social Norms, Spirals of Silence and Framing Theory 141

framing effect theories, this chapter argues that our sociocultural environment (i.e., social norms) exerts an influence on how we interpret mediated messages, form opinions on controversial issues, and publicly express our viewpoints on the issues. Social norms are cultural phenomena, and, as a result, individuals from different cultures responding to the same mediated messages may form different opinion and willingness to express their views about controversial issues. Although more research is needed to compare different cultural groups, the work on cross-cultural comparison in the spiral of silence research has shown systematic variations in willingness to speak out and other variables of interest across cultures. The empirical evidence in Fung’s (2010) study provides preliminary support that framing effects may likewise vary across cultures. The fi ndings in the cross-cultural comparison studies in the spiral of silence and framing effects indicate that media effects may be culturally malleable. As such, media effect scholars should consider differentiating between universal and culturally-specific effects. For example, the spiral of silence and framing effects may fall under cultural specificity, whereas agenda-setting and priming (Iyengar & Kinder, 1987; McCombs & Shaw, 1972), which are based on attitude accessibility to increase the issue’s saliency to retrieve easily from individual memory and subsequently make a political judgment, may fall under universality. Careful attention to the differences between universal and culturally specific processes can help the area of media effects develop more integrative, informed, and precise theories. A clear appreciation of the differences in media effects processes, across cultures, may also enable media effects scholars to identify, with more clarity, culturally appropriate outcome variables for investigation. References Bantz, C. (1997). News organizations: Confl ict as a crafted cultural norm. In. D. Berkowitz (Ed.), Social meanings of news: A text-reader (pp. 123–137). Thousand Oak, CA: Sage. Benford, R. D., & Snow, D. A. (2000). Framing Processes and Social Movements: An Overview and Assessment. Annual Review of Sociology, 26(1), 611–639 Bennett, W. L. (1990). Towards a theory of press-state relations in the US. Journal of Communication, 40(2), 103–125. Bennett, W. L., Lawrence, R. G., & Livingstone, S. (2007). When the press fails. Chicago, IL: Chicago University Press. Berger, P. L., & Luckmann, T. (1966). The social construction of reality. Garden City, NY: Doubleday. Bicchieri, C. (2006). The grammar of society: The nature and dynamics of social norms. New York, NY: Cambridge University Press. Blumer, J., McLeod, J., & Rosengren, K. (1992). An introduction to comparative communication research. In J. Blumer, J. McLeod, & K. Rosengren (Eds.), Comparatively speaking: Communication and culture across space and time (pp. 3–18), Newbury Park, CA: Sage. Breed, W. (1997). Social control in the newsroom: A functional analysis. In D. Berkowitz (Ed.), Social meanings of news: A text-reader (pp. 107–122). Thousand Oak, CA: Sage. Bryant, J., & Miron, D. (2004). Theory and research in mass communication. Journal of Communication, 54, 662–704.

142

Timothy K. F. Fung and Dietram A. Scheufele

Cappella, J., & Jamieson, K. (1997). Spiral of cynicism: The press and the public good. New York, NY: Oxford University Press. Chang, T., Wang, J., & Chen, C. (1998). The social construction of international imagery in the post-cold war era: A comparative analysis of U.S. and Chinese national TV news. Journal of Broadcasting & Electronic Media, 42(3), 277–296. Cialdini, R. B., Kallgren, C. A., & Reno, R. (1991). A focus theory of normative conduct: A theoretical refi nement and reevaluation of the role of norms in human behavior. In M. Zanna (Eds.), Advances in experimental social psychology (Vol. 24, pp. 202–235). San Diego, CA: Academic Press. Cialdini, R. B., & Trost, M. B. (1998). Social influence: Social norms, conformity, and compliance. In D. T. Gilbert, S. Fiske, & G. Lindzey (Ed.), The handbook of social psychology (4th ed., Vol. 2, pp. 151–192). New York, NY: Oxford University Press. Druckman, J. N. (2001). The implications of framing effects for citizen competence. Political Behavior, 23, 225–256. Dunwoody, S. (1992). The media and public perceptions of risk: How journalists frame risk stories. In W. Bromley & K. Segerson (Eds.), The social response to environmental risk: Policy formulation in an age of uncertainty (pp. 75–100). Boston, MA: Kluwer. Dunwoody, S. (1997). Science writers at work. In. D. Berkowitz (Ed.), Social meanings of news: A text-reader (pp. 155–168). Thousand Oaks, CA: Sage. Fazio, R. H. (1990). Multiple processes by which attitudes guide behaviors: The MODE model as an integrative framework. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. 75–109). San Diego, CA: Academic Press. Fiske, S., & Taylor, S. (1991). Social cognition (2nd ed.). Singapore: McGraw-Hill. Fung, K. F. (2010). Value framing, cultural cognitive systems and cultural cognition: Exploring how media framing effects vary across and within cultures. (Unpublished doctoral dissertation). University of Wisconsin-Madison, Madison, WI. Fung, T. K. F., Brossard, D., & Ng, I. (2011). There is water everywhere: How news framing amplifies the effect of ecological worldviews on preference for flooding protection policy. Mass Communication and Society, 14(5), 553–577. Gamson, W. A., Croteau, D., Hoynes, W., & Sasson, T. (1992). Media images and the social construction of reality. Annual Review of Sociology, 18, 373–393. Goff man, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA: Harvard University Press. Hechter, M., & Opp, K. (2001). Introduction. In M. Hechter & K. Opp (Eds), Social norms (pp. xi–xx). New York, NY: Russell Sage Foundation. Hersh, S. M. (1970, May). My Lai 4. Harper’s, 240(1440), 53–84. Horne, C. (2001). Sociological perspectives of the emergence of norms. In M. Hechter, & K. Opp (Eds.), Social norms (pp. 3–34). New York, NY: Russell Sage Foundation. Huang, H. (2005). A cross cultural test of the spiral of silence. International Journal of Public Opinion Research, 17(3), 324–345. Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion. Chicago, IL: University of Chicago Press. Kluckhohn, F. R., & Strodtbeck, F. L. (1961). Variations in value orientations. Evanston, IL: Row & Peterson. Lapinski, M. K., & Rimal, R. (2005). An explication of social norms. Communication Theory, 15(2), 127–147. Lee, W. P., Detenber, B., Willnat, L., Aday, S., & Graf, J. (2004). A cross-cultural test of the spiral of silence theory in Singapore and the United States. Asian Journal of Communication, 14(2), 205‒226. Luntz, F. (2008). Words that work: It’s not what you say, it’s what people hear. New York, NY: Hyperion. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176–187.

Social Norms, Spirals of Silence and Framing Theory 143

McQuail, D. (2010). Mass communication theory (6th ed.). London, England: Sage. Mead, M. (1937). Public opinion mechanisms among primitive peoples. Public Opinion Quarterly, 1, 5–16. Merton, R. K. (1968). Social theory and social structure. New York, NY: The Free Press. Milgram, S., Leonard, B., & Berkowitz, L. (1969). Note on the drawing power of crowds of different size. Journal of Personality and Social Psychology, 13(2), 79‒82. Newcomb, T. M. (1943). Personality and social change. New York, NY: Dryden. Newcomb, T. M. (1958). Social psychology. New York, NY: Henry Holt. Neuwirth, K., Frederick, E., & Mayo, C. (2007). The spiral of silence and fear of isolation. Journal of Communication, 57(3), 450–468. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24, 24–51. Noelle-Neumann, E. (1983). The spiral of silence: Public opinion — Our social skin. Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1991). The theory of public opinion: The concept of spiral of silence. In J. A. Anderson (Ed.), Communication yearbook 14 (pp. 256–287), Newbury Park, CA: Sage. Noelle-Neumann, E. (1993). The spiral of silence: Public opinion — Our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Pan, Z., & McLeod, J. (1991). Multilevel analysis in mass communication research. Communication Research, 18(2), 140–174. Pepitone, A. (1976). Toward a normative and comparative biocultural social psychology. Journal of Personality and Social Psychology, 34, 641–653. Potter, W. (2009). Conceptualizing the audience. In R. Nabi & M. B. Oliver (Eds.), The Sage handbook of media processes and effects (pp. 19–34). Thousand Oak, CA: Sage. Price, V., & Tewksbury, D. (1997). News values and public opinion: A theoretical account of media priming and framing. In G. Barnett & F. J. Boster (Eds.), Progress in communication science (Vol. 13, pp. 173–212). Greenwich, CT: Ablex. Rokeach, M. (1973). The nature of human values. New York: The Free Press. Ross, E. A. (1969). Social control: A survey of the foundation of order. Cleveland, OH: The press of Case Western Reserve University. Scheufele, D. A. (1999). Framing as a theory of media effects. The Journal of Communication, 49(1), 103–122. Scheufele, D. A. (2000). Agenda-setting, priming, and framing revisited: Another look at cognitive effects of political communication. Mass Communication & Society, 3(2&3), 297–316. Scheufele, D. A. (2008). Framing theory. In W. Donsbach (Ed.), The international encyclopedia of communication (pp. 1862–1868). New York, NY: Wiley-Blackwell. Scheufele, D. A., & Iyengar, S. (forthcoming). The state of fra ming research: A call for new directions. In K. Kenski & K. H. Jamieson (Eds.), The Oxford handbook of political communication theories. New York, NY: Oxford University Press. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12(1), 3–28. Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57(1), 9–20. Scott, J., & Marshall, G. (2009). A dictionary of sociology. New York, NY: Oxford University Press. Shen, F., & Edwards, H. H. (2005). Economic individualism, hmanitarianism, and welfare reform: A value-based account of framing effects. The Journal of Communication, 55(4), 795–809. Sherif, M. (1936). The psychology of social norms. New York, NY: Harper. Shulman, H. C., & Levine, T. R. (2012). Exploring social norms as a group-level phenomenon: Do political participation norms exist and influence political participation on college campuses? Journal of Communication, 63(2), 532–552. Shoemaker, P., & Reese, S. (1996). Mediating the message: Theories of influences on mass media content (2nd ed.). New York, NY: Longman. Sire, J. W. (1976). The universe next door. Downing Grove, IL: Intervarsity Press.

144 Timothy K. F. Fung and Dietram A. Scheufele

Sniderman, P. M., Brody, R., & Tetlock, P. E. (1991). Reasoning about politics: Explorations in political psychology. Cambridge, England: Cambridge University Press. Sniderman, P., & Theriault, S. M. (2004). The structure of political argument and the logic of issue framing. In P. M. S. Willem E. Saris (Ed.), Studies in public opinion: attitudes, nonattitudes, measurement error, and change (pp. 133–165). Princeton, NJ: Princeton University Press. Spencer, A., & Croucher, S. (2008). Basque nationalism and the spiral of silence: An analysis of public perception of ETA in Spain and France. International Communication Gazette, 70(2), 137–153. Tewksbury, D., & Scheufele, D. (2009). News framing theory and research. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (pp. 17–33). New York, NY: Routledge. Thibaut, J. W., & Kelley, H. H. (1959). The social psychology of groups. New York, NY: Wiley. Tittle, C. R. (1977). Sanction fear and the maintenance of social order. Social Forces, 55(3), 579–596. Van Gorp, B. (2007). The constructionist approach to framing: Bringing culture back in. Journal of Communication, 57, 60–78.  

12 THE INTERSECTION OF THIRD-PERSON EFFECT AND SPIRAL OF SILENCE Albert C. Gunther

When W. Phillips Davison developed his richly thoughtful, impressionistic piece about perceived communication effects on self and others, he saw immediately that his notion shared some of the real estate inhabited by Noelle-Neumann’s ideas about the dynamics of social perceptions. Davison acknowledged this resemblance in his discussion section, “The third-person effect is probably involved in the spiral of silence….” And, referring to election campaigning on the same page, he added: “Each person may reason: I haven’t been influenced by this widely publicized nonsense, but they probably have been” (Davison, 1983, p. 12). This last statement—about I and they—gets at the central similarity of these two theoretical ideas, but also one of the key differences. Both models propose that mass media play an important role in people’s perceptions of the climate of opinion, but beyond that point they take some different directions. Stage 1: Perceived Influence and the Climate of Opinion On a theoretical map, the place where third-person effect and spiral of silence come closest to one another is in the territory we might call perceived public opinion. But, conceptually at least, one has to think past the surface to see their proximity. This part of Davison’s model, what we now often refer to as the third-person perception, focused very explicitly on peoples’ perceptions of the influence of a message, and more specifically the differential in perceived influence on oneself and on others. To Davison, perceived influence was important because of the consequences that might ensue (more on this later). But for the spiral-of-silence model, it is important because of what it implies, namely that perceived influence would likely result in effects for perceived opinion. If

146

Albert C. Gunther

people are influenced, their opinions are likely to change (or to undergo more nuanced shifts, such as to solidify or intensify). At this place on the map, these two ideas work together in two ways. One, they both describe self-other differences, and thinking that others (and others’ opinions) may be different from one’s self and, two, they both invoke the sense of media influence and as a consequence that there is a perceived opinion trend, perhaps a trend away from one’s own position. So at this point, while it may appear that third-person effect and spiral of silence describe different concepts, they converge on a central notion that Noelle-Neumann called the perceived climate of opinion. But concepts are static and, in a sense, lifeless objects without theory. And if we survey the theories that might explain perceived climate of opinion, we see less resemblance. Noelle-Neumann, in fact, had little to say about the formation of perceived public opinion. She clearly believed that mass media loomed large in this landscape—referring to crowd reaction shots as cues, for example, and observing that “impressions about frequency distribution are conveyed by the media” (1993, p. 217). But in general, her picture of this part of the process is rather vague. She does suggest that “media imprint stereotypes through innumerable repetitions…” (Noelle-Neumann, 1984, p. 156), and she refers to the ever-popular “quasi-statistical organ” (Noelle-Neumann, 1974, p. 44) that people employ to survey the opinion climate. In that surveillance process, she says, “…we know that the media represent the most important source for the individual’s constant observation of his or her environment” (NoelleNeumann, 1993, p. 216). What’s more, she suggests very strongly that when people form misimpressions of the climate of opinion, “…we may suspect that media effects are involved” (p. 216). But how exactly media inspire those perceptions, or misperceptions, is not clearly or extensively explained. Subsequent research in other domains has added more detail to this part of the map. Exemplification studies, for example, tell us that anecdotal evidence is more influential than base-rate information in shaping opinion perceptions (e.g., Brosius & Bathelt, 1994) and studies of impersonal influence reveal that media content can affect perceptions of others more than any effect it may have on the self (Mutz, 1998). Along with Noelle-Neumann’s, these explanations appear to involve cultivation or representativeness processes in which people draw on cues about public opinion from media content. Here is where Davison points to an alternative route, for perceived communication influence, especially influence on others’ opinions, offers another intuitively appealing explanation for how individuals form an impression of what others are thinking. In his scenario, people (a) observe mass media content, (b) assume that content will influence others, and (c) thus infer the location, and perhaps the direction of movement, of popular sentiment on the public opinion landscape. A few early empirical studies tested third-person perceptions in tandem with

The Intersection of Third-Person Effect and Spiral of Silence

147

Noelle-Neumann’s model. One study showed that students believed media coverage of university divestment of fi nancial interests in South Africa had more influence on others than on themselves and that perceptions of a congruent trend in others’ opinions predicted respondents’ willingness to express their own opinions (Mutz, 1989). Similarly, Hong Kong residents who perceived news about the Sino-British dispute over Hong Kong’s political future to be influencing public sentiment away from their own were less likely to voice their own political point of view (Willnat, 1996). However, for the most part third-person-perception research did not dovetail exactly with Noelle-Neumann’s model because of one critical missing step: most early studies did not make a clear connection between media content and the climate of opinion. The destination outcome for most third-personperception studies stopped at perceived communication influence. Though presumably examining variables closely related to perceived public opinion, these studies did not make that one additional step to bridge the small divide between perceived influence on others and perceived opinions (or opinion trends) of others. The Persuasive Press Inference In the late 1990s, a series of experiments—the persuasive press inference studies—crossed this bridge by showing how presumptions about the persuasive influence of mass media could be directly related to changes in the perceived climate of opinion. In these experiments, respondents read news articles on current issues that were manipulated to present either a favorable or unfavorable slant. Results showed that the slant of media coverage significantly altered respondents’ perceptions of public opinion as well as recent opinion trends (Gunther, 1998; Gunther & Christen, 1999). (This effect remained robust even when accounting for projection biases—the tendency for people to think others’ opinions more closely resemble their own.) A related field experiment showed that readers of a local newspaper mistakenly believed news coverage was affecting community opinions even though they thought it was having no effect on their own (Mutz & Soss, 1997). Subsequent studies confi rmed this fi nding (Gunther, Christen, Liebhart, & Chia, 2001; Kim, Han, Shanahan, & Berdayes, 2004) and helped better establish the connection between perceived media influence and perceived climate of opinion. The reasoning behind presumed-influence inferences about the climate of opinion is built on several assumptions—assumptions salient in both thirdperson-effect and spiral-of-silence models. One, people attentive to a prominent issue can easily form an impression of the slant of media coverage of that issue (even though their sample of coverage may be far from representative because of selective exposure or other biases). Two, the law-of-small-numbers bias coupled with perceptions of broad media reach argue that these people will expect all

148 Albert C. Gunther

media content to closely resemble their sample and others to be exposed to that same content. Three, assuming that others are vulnerable (a fundamental notion in the third-person effect, especially when the message is seen as undesirable) will lead people to infer that other people are influenced by this media coverage and thus that public opinion will be moving in that direction. Inferring this relationship between media content and opinion climate would be, in effect, one (though probably not the only) function of Noelle-Neumann’s quasi-statistical organ. As noted above, undesirable messages—messages disagreeable to one’s own point of view—are perceived as especially influential on vulnerable others. Davison himself expressed this idea before the hostile media effect (Vallone, Ross, & Lepper, 1985) had a name. He noted anecdotal evidence that partisans on both sides of a contested issue saw media coverage of that issue as biased, speculating that “each observer assumes a disproportionate effect will be achieved by arguments or facts supporting the ‘wrong’ side of the issue” (Davison, 1983, p. 11). This notion about the undue influence of disagreeable information, known as the negative influence corollary, is important to both third-person effect and spiral-of-silence processes. (Positive or agreeable messages seem not to generate a third-person perception and can even work in the reverse direction, a phenomenon sometimes called the “first-person effect” in which people see themselves as more influenced than others.) The negative influence corollary is noteworthy also for spiral of silence theory because, insofar as “undesirable” messages are seen to be especially influential on the opinions of others, they are logically also going to create the sense that others opinions are diverging from one’s own. This divergence, in turn, will lead to the self-censorship and opinion suppression that Noelle-Neumann saw as an ultimate outcome of these social influence processes. Presumed influence is certainly not the only source of public opinion perceptions; other factors are surely at work, including reports (accurate or otherwise) of poll results, extrapolation from vivid anecdotes and exemplars, opinions of friends and acquaintances, sentiments on bumper stickers and yard signs, comments posted below news articles, and statements from politicians and opinion leaders. Most notable among these is the projection of personal opinion, a robust effect which tends to work against dysfunctional spiral of silence outcomes insofar as it causes people to see others as more similar to themselves (see, e.g., Shamir & Shamir, 2000). But the persuasive press inference appears to play a significant role. Stage 2: Third-Person Effects and the Consequences of Perceived Public Opinion Another striking resemblance between these two models is that they both incorporate a second stage or second leg in the causal sequence. In both cases,

The Intersection of Third-Person Effect and Spiral of Silence

149

perceived influence or the perceived opinion climate gives rise to a subsequent outcome. And again, there are important similarities and important differences. Noelle-Neumann’s spiral of silence theory focuses very clearly on one particular dependent variable—a form of self-censorship defi ned as a person’s willingness, or reluctance, to speak up in a social setting. The more people perceive the prevailing climate of opinion to be at odds with their own. the less likely they are to express those personal opinions. In Davison’s (1983) original article the consequence of perceived influence could take several forms, but the major outcome of interest was also related to censorship. The more people think a message may influence others in an undesirable way (especially relative to influence on the self ), the more they will support restrictions on that message. In an important sense both of these outcomes are rooted in communication censorship, but the objects of censorship are different. In the spiral of silence model, it’s censorship of the self; in the third-person effect it’s censorship of the presumably influential message. Theories for Stage Two The most frequently discussed and explored explanations for these censorship outcomes are also somewhat different. In the spiral of silence people are afraid to speak up because of a fear of social isolation, isolation that might follow the voicing of an unpopular point of view. It is a self-protective mechanism meant to avoid social sanctions and also shelter one’s own opinions, and ego, from the disapproval of others. Third-person effect explanations are also rooted in protection, but protection for others rather than from others. The most common explanation for message censorship is paternalism—the desire to protect other people, society at large, from the undesirable influences that are perceived to spring from misleading messages that may be found in advertising or propaganda, not to mention pornography, violence and other allegedly harmful media content (Gunther, 1995; McLeod, Eveland, & Nathanson, 1997; Rojas, Shah, & Faber, 1996). And interestingly, some third-person-effect research has found that the impulse to restrict such messages is stronger as the self-other difference grows larger. That is, whatever one thinks about influence on the self, it is how much more influence is perceived for others that determines how much one supports censorship of the offending message. (Note that this explanation also permits a self-serving motive; although censorship of such messages may afford a protection for society at large, it indirectly benefits the self insofar as it makes society in general a safer, less-threatening environment in which to live.) Divergent Outcomes Thus far, these two models have been remarkably parallel in structure. We see similarities and differences in concepts and theoretical links, but the general idea in both cases begins with perceived opinions of others and progresses

150

Albert C. Gunther

toward reactions—reactions that invoke restrictions of one sort or another—as a result of those initial perceptions. Clearly, reluctance to express one’s own opinion is a behavioral outcome that falls squarely within the third-person effect model. In fact, it’s not a stretch to call the spiral of silence notion a good example of the third-person effect. However, as the map has evolved and been redrawn through decades of subsequent research involving many hundreds of empirical studies, the second leg of the third-person effect model has significantly diverged. And the beginnings of this divergence were visible even in Davison’s seminal paper, for he saw outcomes beyond just censorship. Davison had a robust sense of humor. No student of this literature should miss his example of movie censors who, though avowedly unaffected by offensive cinematic scenes (except for episodes involving food and sex), felt they had to delete them to safeguard the welfare of the mass audience. But Davison could see further. Although he thought censorship offered the “most interesting field for speculation,” he noted other types of potential outcomes of perceived influence: that people might buy or sell stocks in anticipation of others’ reactions to relevant news reports, that reports of shortages might result in pre-emptive hoarding, that manufacturers might change a product line when news stories suggest consumers will shift their priorities. In a darker political mood, he speculated that “Exaggerated expectations about the effects of dissident communications have caused countless people to be incarcerated, tortured, and killed” by “priests and potentates” (1983, p. 14). And, indeed, as third-person-effect research traveled beyond the censorship territory explored in the mid-1990s a widening array of outcomes was discovered and described. Many of these fall under a somewhat broader variant of the third-person effect often called the “influence of presumed media influence” (Gunther & Storey, 2003). Perceived influence on others—especially peers— has been related to strategic voting (Cohen & Tsfati, 2009), adolescent smoking (Gunther, Bolt, Borzekowski, Liebhart, & Dillard, 2006), eating behaviors (Milkie, 1999), permissive sexual attitudes (Chia, 2006), thoughts of moving (Tsfati & Cohen, 2003), resistance to moving (Tsfati & Cohen, 2005), and a reluctance to dispense advertised prescription drugs (Huh & Langteau, 2007). Although these extensions of the third-person-effect and presumed-influence models embraced a diverse selection of so-called behavioral outcomes, many of them can be explained by social-influence, or conformity, or resistance to conformity (such as the hard-core individuals who refuse to keep quiet) theories. In this sense, they share one or more borders with Noelle-Neumann’s idea of public opinion as the glue that holds society together, and they are not so distant from the fear-of-isolation explanation that underlies the spiral of silence. An important practical element of both theories is the accuracy of perceptions, or the lack of it. People may sometimes be quite correct in their presumption of communication influences on the opinions and attitudes of others. Their reactions, for better or worse, may also be appropriate, and these scenarios

The Intersection of Third-Person Effect and Spiral of Silence

151

are consistent with the underlying theories. But both third-person-effect and spiral-of-silence processes take on more (perhaps ominous) significance when perceptions are faulty. And false perceptions are likely not uncommon because of various built-in biases, including the negative influence corollary and the persuasive press inference, that may skew perceptions of public opinion. In this context, politicians, policy makers and others with influence and power are especially interesting. These are people who may be highly sensitive (or oversensitive) to the potential influences of media coverage on others, particularly because their stature, or even their jobs, are often intimately intertwined with public sentiment. In addition, and just as importantly, their actions or reactions may be unusually consequential. Misjudgments and misperceptions are the charismatic examples, the ones that get attention because their outcomes are based on a false premise and may be dysfunctional. In these cases (e.g., disagreeable viewpoints, negative political advertising, pornography), people overestimate message influence on others and thus their support for restrictions, and the potential erosion in freedom of speech that follows, are based on a false perception. Similarly, individuals who mistakenly perceive a disagreeable trend in the climate of opinion will not voice their own thoughts and ideas, thus suppressing what may be an important and perhaps even popular point of view and diminishing the quality of public discourse. But it is important to remember that, in their many similarities, both third-person-effect and spiral-of-silence theories describe processes that occur whether the perceptions are accurate or not. Acknowledgment The author expresses his sincere gratitude to ByungGu Lee, Stephanie Tsang, JungHwan Yang, and Woohyun Yoo for editorial assistance and advice, and to Yariv Tsfati for helpful editing and comments on earlier drafts of this chapter. References Brosius H-B., & Bathelt, A. (1994). The utility of exemplars in persuasive communication. Communication Research, 21, 48–78. Chia, S. (2006). How peers mediate media influence on adolescents’ sexual attitudes and sexual behavior. Journal of Communication, 56, 585–606. Cohen, J., & Tsfati, Y. (2009). The influence of presumed media influence on strategic voting. Communication Research, 36, 359–378. Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47, 1–15. Gunther, A. C. (1995). Overrating the X-rating: The third-person perception and support for censorship of pornography. Journal of Communication, 45(1), 27–38. Gunther, A.C. (1998). The persuasive press inference: Effects of mass media on perceived public opinion. Communication Research, 25(2), 486–504. Gunther, A. C., Bolt, D., Borzekowski, D. L. B., Liebhart, J. L., & Dillard, J. P. (2006). Presumed

152

Albert C. Gunther

influence on peer norms: How mass media indirectly affect adolescent smoking. Journal of Communication, 56, 52–68. Gunther, A. C., & Christen, C. T. (1999). Effects of news slant and base-rate information on perceived public opinion. Journalism and Mass Communication Quarterly, 76, 277–292. Gunther, A. C., Christen, C. T., Liebhart, J. L., & Chia, S. C.-Y. (2001). Congenial public, contrary press and biased estimates of the climate of opinion. Public Opinion Quarterly, 65, 295–320. Gunther, A. C., & Storey, J. D. (2003). The influence of presumed influence. Journal of Communication, 53, 199–215. Huh, J., & Langteau, R. (2007). Presumed influence of DTC prescription drug advertising on patients: Physicians’ perspective. Journal of Advertising, 36(3), 151–172. Kim, S. H., Han, M., Shanahan, J., & Berdayes, V. (2004). Talking on ‘Sunshine in North Korea’: A test of the spiral of silence as a theory of powerful mass media. International Journal of Public Opinion Research, 16(1), 39–62. McLeod, D. M., Eveland, W. P., Jr., & Nathanson, A.I. (1997). Support for censorship of violent and misogynic Rap lyrics. Communication Research, 24, 153–174. Milkie, M. A. (1999). Social comparisons, reflected appraisals, and mass media: The impact of pervasive beauty images on black and white girls’ self-concepts. Social Psychology Quarterly, 62, 190–210. Mutz, D. C. (1989). The influence of perceptions of media influence: Third-person effects and public expression of opinions. International Journal of Public Opinion Research, 1, 3–23. Mutz, D. C. (1998). Impersonal influence: How perceptions of mass collectives affect political attitudes. Cambridge, England: Cambridge University Press. Mutz, D. C., & Soss, J. (1997). Reading public opinion: The influence of news coverage on perceptions of public sentiment. Public Opinion Quarterly, 61(3), 431–451. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24(2), 43‒51. Noelle-Neumann, E. (1984). The spiral of silence: Public opinion — Our social skin. Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1993). The spiral of silence: Public opinion — our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Rojas, H., Shah, D. V., & Faber, R. J. (1996). For the good of others: Censorship and the thirdperson effect. International Journal of Public Opinion Research, 8, 163–186. Shamir, J., & Shamir, M. (2000). The anatomy of public opinion. Ann Arbor: University of Michigan Press. Tsfati, Y., & Cohen, J. (2003). On the effect of the “third-person effect”: Perceived influence of media coverage and residential mobility intentions. Journal of Communication, 53, 711–727. Tsfati, Y., & Cohen, J. (2005). The influence of presumed media influence on democratic legitimacy: The case of Gaza settlers. Communication Research, 32, 794–821. Vallone, R. P., Ross, L., & Lepper, M. R. (1985). The hostile media phenomenon: Biased perception and perceptions of media bias in coverage of the Beirut massacre. Journal of Personality and Social Psychology, 49(3), 577–585. Willnat, L. (1996). Mass media and political outspokenness in Hong Kong: Linking the thirdperson effect and the spiral of silence. International Journal of Public Opinion Research, 8(2), 187–212.

13 PLURALISTIC IGNORANCE AND THE SPIRAL OF SILENCE MEET Mutual Lessons Jacob Shamir

Introduction Although the concept of pluralistic ignorance was fi rst introduced in the 1920s, renewed interest in it emerged 50 years later at about the same time that Noelle-Neumann’s writings on the spiral of silence were appearing in Englishlanguage journals. Both traditions focus on public opinion perceptions, that is, how people view other people’s opinions and values. The major empirical observation within the pluralistic ignorance tradition is that quite often people err in their assessment of public opinion. In other words, public opinion perceptions do not always parallel public opinion distributions. Public opinion perception is also key in Noelle-Neumann’s theory, in that people constantly observe their environment to see which opinion is prevalent and on the rise (the quasi-statistical sense). Moreover, the climate of opinion, which is the aggregate manifestation of these perceptions, is at the center of Noelle-Neumann’s theory of public opinion, not the frequency distribution of individuals’ opinions; and the spiral of silence is a reaction to changes in the climate of opinion. Both pluralistic ignorance and spiral of silence theories are concerned with the explanation of public opinion perceptions and their divergence from actual opinion distributions. Both traditions also combine the micro and macro levels of public opinion. Pluralistic ignorance has been studied from psychological and social perspectives, and the spiral of silence theory combines the two levels in its presumed mechanism of change (for an excellent early exposition in these terms, see Taylor, 1982; see also Scheufele, 2008). Furthermore, both traditions require dynamic formulation. The spiral of silence theory is an explicitly dynamic theory of public opinion, and the understanding of pluralistic

154

Jacob Shamir

ignorance is greatly enhanced by a dynamic perspective. Bringing together the two approaches’ dynamics broadens our outlook on public opinion, and highlights the role of both normative and informational sources of public opinion. Pluralistic Ignorance The social psychologist Floyd Allport introduced the concept of pluralistic ignorance in the 1920s. He saw it in psychological terms as an illusion, as unwarranted and mistaken impressions of how other people feel and think on various matters (Allport, 1924; Katz & Allport, 1931). From the 1970s on, renewed interest in these “patterns of false beliefs” emerged in diverse areas of research such as race relations, international politics, voting preferences, and bystanders’ reactions to people in distress (O’Gorman, 1986, pp. 335, 345). “‘At the extreme the phenomenon of ‘pluralistic ignorance’ … reflects a situation where people appear to operate within a ‘false’ social world, or at least one quite different from that observed by the presumably objective social scientist” (Fields & Schuman, 1976, p. 427). This misconstruction has been seen in the pluralistic ignorance tradition in public opinion research in terms of perceptual accuracy (Daschmann, 2008), as a misreading of the majority position. Most extreme and fascinating are those cases in which people misperceive the minority position on an issue to be the majority, and vice versa, and empirical studies have generally limited their analysis to such cases. In more moderate cases of pluralistic ignorance, people seriously misperceive the distribution of opinion but not to the point of misconstruing the mode. Thus pluralistic ignorance marks a continuum of bias and inaccuracy: it can be severe or mild. Pluralistic ignorance is perceived differently from an individual-level versus a social perspective. The individual-level perspective puts an emphasis on the “ignorance” element of “pluralistic ignorance,” whereas social perspectives highlight the “pluralistic” part of the phenomenon. In the social approach, “pluralistic ignorance is not ignorance in the ordinary sense of not knowing. On the contrary, it is knowledge of others that is mistakenly considered to be correct … socially accepted but false propositions about the social world” (O’Gorman, 1986, p. 333). The social nature of pluralistic ignorance makes the “ignorance” composite, shared and collective rather than personal distortions and individual ignorance. As such, the error is necessarily systematic rather than random, “invalid” rather than “unreliable” (Fields & Schuman, 1976, p. 427). Adopting the social perspective also leads one to realize that “there is a difference between situations involving the perceptual distortion of otherwise accurate information and those involving the veridical perception of false or misleading information” (O’Gorman, 1986, p. 334). According to O’Gorman, the second situation is that of pluralistic ignorance, and I follow his focus in this review. This is not to say that individual-level analysis and psychological processes are irrelevant to the understanding of pluralistic ignorance, but it is the

Pluralistic Ignorance and the Spiral of Silence Meet

155

social perspective that connects the pluralistic ignorance and spiral of silence traditions, and which is of most relevance to students of public opinion. Explaining Public Opinion Perceptions and Pluralistic Ignorance Attempts to understand public opinion perceptions, their accuracy, and pluralistic ignorance similarly intersect different levels of analysis and approaches. From the psychological perspective, pluralistic ignorance is due to people not being perfectly adequate information processors because of cognitive shortcomings and self-serving biases. The focus in explanation is on information processing, attribution, and internal motivations. From the social perspective, pluralistic ignorance is due to error-prone messages from the environment. Inadequate, misleading, or false information cues serve as invalid indicators for public opinion and produce pluralistic ignorance. Seeing pluralistic ignorance as a shared phenomenon and as one involving misinformation or even disinformation implies that it has social sources, and shifts the focus away from psychological processes toward the social environment and what in it operates so as to mislead people about it. The relevant concept becomes visibility, “the counterpart in social structure of what from the standpoint of psychological theory is social perception” (Merton, 1968, p. 404). Visibility is a property of society and of issues in social context, a function of various communication, social, and political factors, such as the nature of the media environment and of interpersonal networks, the prominence of an issue on the public agenda, resonance with cultural norms and values, and social and political structuredness (Shamir & Shamir, 2000). Visibility diminishes pluralistic ignorance. Noelle-Neumann’s spiral of silence theory provides two insights from this perspective. One is not explicitly related in her work to pluralistic ignorance, but is a necessary link in the spiral of silence theory, and that is expressed opinion of a specific kind. Dedicated and enthusiastic social and political groups who are in the minority but display their convictions and express themselves openly and self-confidently may produce a spiral of silence. “The climate of opinion depends on who talks and who keeps quiet” (Noelle-Neumann, 1993, pp. 4–5), and Noelle-Neumann’s prescription for a successful election campaign strategy is indeed to make the party visible in order to affect the climate of opinion (1993). Groups which make noise and appear strong and as gaining strength set in motion a dynamic spiraling process where expressive opinion and opinion perceptions nurture each other in a process which involves misleading people about the actual strength of groups and opinions, leading to mistaken opinion perceptions—in other words to pluralistic ignorance—as part of the process of change in the climate of opinion. Second, the mass media play a central role in Noelle-Neumann’s theorizing, in the construction of the climate of opinion and in the creation of pluralistic

156 Jacob Shamir

ignorance. In her view, journalists’ political leanings, generally liberal, are reflected in the media, and, at times, mold a climate that does not reflect the actual distribution of opinion in society. She even suggests that “…pluralistic ignorance, a condition whereby the people have a mistaken idea about how most people feel, will serve as a kind of guide for tracking down the media’s effects” (1993, p. 169). Noelle-Neumann thus sees the mass media as the source of error-prone messages that may lead to pluralistic ignorance. It should be noted that hegemonic or critical approaches to the media raise similar arguments, but of a different or opposite political flavor (e.g., Golding, 1981; Hall, 1977). Others point to general systematic biases in the media attributable to organizational constraints and routines, occupational practices, and technical limitations, which may introduce inaccuracy in the media’s portrayal of public opinion (Moy, 2008; Shamir & Shamir; 1997). Other mass media explanations for pluralistic ignorance that have been suggested include different journalistic approaches, the persuasive press inference and the cultivation effect (Eveland & Glynn, 2008). Noelle-Neumann developed the spiral of silence theory and the “return to the concept of powerful mass media” (1973) claim at the highpoint of the broadcast era, indeed basing it on the media’s ubiquity, cumulation, and consonance. This view was too simplistic then already (Salmon & Kline, 1985), but it defi nitely is inappropriate in present-day post-broadcast media environment. This new media environment, brought about by cable television, the Internet, and cell phone technologies, enhances partisan selective exposure among information seekers and gaps between citizens with political versus other content preferences (Bennett & Iyengar, 2008; Prior, 2007). The effect of this new media environment on opinion formation, climate construction, and pluralistic ignorance is only beginning to be explored (e.g., Debatin, 2008; Tsfati, Chotiner, & Stroud, 2012). The fragmentation and abundance of choice in this media environment seem to imply a smaller media effect, polarization, and lesser likelihood for spirals of silence and the construction of overriding climates of opinion. On the other hand, a new interdisciplinary paradigm termed Swarm Intelligence (Trianni, Tuci, & Passino, 2011) suggests that the new information and communication technologies may form a swarm-like entity of many individual actors with no centralized coordination which aggregate into a new intelligent entity that may be manifested in collective expressions of opinion (Garrett, 2006) or in a climate of opinion. We need to remember that public opinion perceptions, climates of opinion, and pluralistic ignorance build upon multiple indicators, stemming from various sources, not only from the mass media. Thus the impact of the media in broadcast as in post-broadcast media environments is mitigated, and our theoretical and empirical analyses should take the variety of processes and sources into account. Finally, both Noelle-Neumann and pluralistic ignorance scholars point to the role of salience and priming as important factors in determining

Pluralistic Ignorance and the Spiral of Silence Meet

157

which indicators will prevail and bring about (and also break) pluralistic ignorance, and these processes should be incorporated in future studies. A Diachronic View of Pluralistic Ignorance Pluralistic ignorance focuses on the gap between opinion distributions and perceptions, and most studies in this tradition are cross-sectional. Therefore, Glynn, Ostman, and McDonald’s review (1995) regards it as a product, not a process. Adopting a diachronic perspective to pluralistic ignorance advances our understanding of it, and the spiral of silence proposes one dynamic path to it. Noelle-Neumann’s theory describes a pattern of pluralistic ignorance where climate shifts occur without a corresponding change in private attitudes of opinion. The mass media or social agents—vocal party activists or social movements—affect public opinion perceptions in a way that swerves significantly from the actual distribution of opinion. Public opinion perceptions affect who speaks up and who keeps quiet, and these public expression patterns redefi ne opinion perceptions, producing a spiraling process which moves the climate of opinion further in favor of those who are visible. An opposite pattern would be one where attitudes move fi rst and climates lag behind. The climate of opinion, as Noelle-Neumann portrays so well, is the normative facet of public opinion, closer to the more lasting normative and structural characteristics of a society. Attitudes make up the evaluative facet of public opinion, which is also sensitive to events of the day and to political, socioeconomic, and international developments. Given this understanding of climates and private attitudes, we expect this pattern to be more prevalent. This process has been documented in the pluralistic ignorance literature and labeled “conservative bias.” Such divergence is most likely at times of social change, when attitudes shift with ongoing events and developments, while social structure norms and practice lag behind together with the climate. The concept was suggested in the context of racial relations in the United States (Fields & Schuman, 1976; Katz & Allport, 1931; O’Gorman, 1975, 1988), and refers to people’s tendency to consider others more conservative than themselves when attributing opinions other than their own. The term “conservative” in its ideological sense was intertwined with “conservative” in the preservative sense of being disposed toward existing views conditions or institutions. In the context of racial equality, the daily expressions of the existing social structure of racial inequality and segregation provided powerful (misleading) information cues in a conservative, not liberal direction, as to the prevailing norm long after attitudes changed. Similar conservative bias has also been demonstrated in societies engulfed in long-term confl ict. Longitudinal public opinion data among Israelis and Palestinians show that, when opinion perceptions exhibit pluralistic ignorance, it is pluralistic ignorance with a hawkish bias, since the

158 Jacob Shamir

overriding social norms, structures, and practices are oriented to the confl ict; the public discourse is nationalistic and ethnocentric; ingroup-oriented social and political actors and messages are more visible and culturally resonant; and the media tend to portray a hawkish climate (Shamir & Shamir, 2011; Shamir & Shikaki, 2010). Other, though fewer studies have documented an ideologically “liberal” bias in people’s assessments (e.g., Glynn, 1987; Taylor, 1982). It should be noted that the empirical predominance of ideologically conservative over liberal bias runs counter Noelle-Neumann’s claim that the media are the source of error about the opinion climate and that the media tend to be liberal. However, more fundamentally, conservative bias in the preservative sense, that is conservative bias toward prevailing social norms, conditions, and institutions, provides a stronger theoretical grounding to such pluralistic ignorance, be it conservative or liberal ideologically. Thus understood, conservative bias is consistent with the conception of pluralistic ignorance as a problem of visibility and information. In this rendering, conservative bias is a measurement error in terms of public opinion distributions due to the use of now invalid indicators. The same factors that usually serve as valid indicators of opinion distributions and contribute to visibility—social and political institutions, common social practices, and broad cultural norms—may inhibit it in times of change, when synchronization is impaired, and produce pluralistic ignorance. The Dynamics of Public Opinion: Normative and Informational Forces The phenomenon of pluralistic ignorance and its exploration from a social structural and dynamic perspective contributes to a fuller and better understanding of the concept of public opinion. One can ask the somewhat naïve question what public opinion is in the context of pluralistic ignorance. Is it the majority opinion, or is it the opinion perceived as the majority? My response to this dilemma is to consider public opinion as a multidimensional concept, which incorporates attitudes as well as perceptions of opinion distributions, but also other elements including expectations as to future developments and opinion expression (Shamir & Shamir, 2000). These dimensions of public opinion will be often in synchronization, but there can also be discrepancies among them. Pluralistic ignorance is one such case that signifies divergence—at times great divergence—between the actual majority opinion and its perception. But these divergences lead us to explore their sources and thus to better understand what public opinion is and how public opinion changes. The spiral of silence theory is an explicitly dynamic theory of public opinion, whose engine is people’s fear of isolation if they deviate from the normative opinion. The theory, however, focuses on the relation between public opinion expression and perception, and neglects opinion distributions, the standard

Pluralistic Ignorance and the Spiral of Silence Meet

159

view of public opinion. It can be viewed as “a theory of social perception” (Scheufele & Moy, 2000). Relatedly, Noelle-Neumann emphasizes the concept of public opinion as social control over the rational concept of public opinion (1993, 1995), so much so that her empirical work completely ignores the impact of events and developments. The theory stresses the normative source of public opinion, but there is also the informational engine of public opinion. Attitudes (and expectations) are affected by political and social developments and event information (Page & Shapiro, 1992), as the pluralistic ignorance conservative bias cases exemplify. Bringing together the two approaches draws attention to the role of both normative and informational sources of public opinion (Deutsch & Gerard, 1955). Public opinion as a collective social entity encompasses the long-standing tension between normative forces of social control on one hand, and the powerful impact of event information and rationality on the other. It is being shaped by both informational and normative sources. Correspondingly, public opinion dynamics are versatile and a function of normative social control forces and of event information, of spirals of silence, and of conservative biases. Acknowledgment I would like to thank Charles T. Salmon for his stimulating comments on an earlier version of this chapter. References Allport, F. H. (1924). Social psychology. Boston, MA: Houghton Miffl in. Bennett, W., & Iyengar, S. (2008). A new era of minimal effects? The changing foundations of political communication. Journal of Communication, 58, 707–731. Daschmann, G. (2008). Pluralistic ignorance. In W. Donsbach (Ed.), The international encyclopedia of communication. Blackwell. Retrieved from http://www.communicationencyclo pedia.com/subscriber/tocnode?id=g9781405131995_yr2010_chunk_g978140513199521_ ss43-1 Debatin, B. (2008). The internet as a new platform for expressing opinions and as a new public sphere. In W. Donsbach & M. W. Traugott (Eds.), The Sage handbook of public opinion research (pp. 64–72). London, England: Sage. Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational influences upon individual judgment. Journal of Abnormal and Social Psychology, 51, 629–636. Eveland, W. P., Jr., & Glynn, C. J. (2008). Theories on the perception of social reality. In W. Donsbach & M. W. Traugott (Eds.), The Sage handbook of public opinion research (pp. 155–163). London, England: Sage. Fields, J. M., & Schuman, H. (1976). Public beliefs about beliefs of the public. Public Opinion Quarterly, 40, 427–48. Garrett, R. K. (2006). Protest in information society: A review of literature on social movements and new ICTs. Information, Communication, and Society, 1(2), 202–224. Glynn, C. J. (1987). The communication of public opinion. Journalism Quarterly, 64, 688–697. Glynn, C. J., Ostman, R. E., & McDonald, D. G. (1995). Opinions, perception, and social reality. In T. L. Glasser & C. T. Salmon (Eds.), Public opinion and the communication of consent (pp. 249–277). New York, NY: Guilford.

160 Jacob Shamir

Golding, P. (1981). The missing dimension — News media and the management of social change. In E. Katz & T. Szecsko (Eds.), Mass media and social change (pp. 63–82). Beverly Hills, CA: Sage. Hall, S. (1977). Culture, the media and the ideological effect. In J. Curran, M. Gurevitch, & J. Woollacott (Eds.), Mass communication and society (pp. 315–348). London, England: Edward Arnold. Katz, D., & Allport, F. H. (1931). Student attitudes. Syracuse, NY: Craftsman. Merton, R. K. (1968). Social theory and social structure. New York, NY: Free Press. Moy, P. (2008). Pluralistic ignorance and nonattitudes. In W. Donsbach & M. W. Traugott (Eds.), The Sage handbook of public opinion research (pp. 164–171). London, England: Sage. Noelle-Neumann, E. (1973). Return to the concept of a powerful mass media. Studies of Broadcasting, 9, 67–112. Noelle-Neumann, E. (1993). The spiral of silence, 2nd ed. Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1995). Public opinion and rationality. In T. L. Glasser & C. T. Salmon (Eds.), Public opinion and the communication of consent (pp. 33–54). New York, NY: Guilford. O’Gorman, H. J. (1975). Pluralistic ignorance and white estimates of white support for racial segregation. Public Opinion Quarterly, 39, 311–330. O’Gorman, H. J. (1986). The discovery of pluralistic ignorance. Journal of the History of the Behavioral Sciences, 22, 333–347. O’Gorman, H. J. (1988). Pluralistic ignorance and reference groups: The case of ingroup ignorance. In H. J. O’Gorman (Ed.), Surveying social life (pp. 145–173). Middletown, CT: Wesleyan University Press. Page, B. I., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans’ policy preferences. Chicago, IL: University of Chicago Press. Prior, M. (2007). Post-broadcast democracy. New York, NY: Cambridge University Press. Salmon, C. T., & Kline, F. G. (1985). The spiral of silence: Ten years later. In K. R. Sanders, L. L. Kaid, & D. Nimmo (Eds.), Political communication yearbook 1984 (pp. 3–30). Carbondale: Southern Illinois University Press. Scheufele, D. A. (2008). Spiral of silence theory. In W Dosbach & M. W. Traugott (Eds.), The Sage handbook of public opinion research (pp. 176–183). London, England: Sage. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12, 3–28. Shamir, J., & Shamir M. (1997). Pluralistic ignorance across issues and over time: Information cues and biases. Public Opinion Quarterly, 61, 227–260. Shamir, J., & Shamir, M. (2000). The anatomy of public opinion. Ann Arbor: The University of Michigan Press. Shamir, J., & Shamir, M. (2011, May). Public opinion perceptions in the Israeli-Palestinian intractable conflict. Paper presented at the AAPOR 66th Annual Conference, Phoenix, Arizona. Shamir, J., & Shikaki, K. (2010). Palestinian and Israeli public opinion: The public imperative in the second Intifada. Bloomington: Indiana University Press. Taylor, D. G. (1982). Pluralistic ignorance and the spiral of silence: A formal analysis. Public Opinion Quarterly, 46, 311–335. Trianni, V., Tuci, E., & Passino, K. M. (2011). Special issue on swarm cognition. Swarm Intell, 5, 1–2. Tsfati, Y., Chotiner, A., & Stroud, N. J. (2012, June). Ideological news exposure and perceived opinion climate. Paper presented at the WAPOR 65th Annual Conference, Singapore.

14 OPINION LEADERSHIP AND PUBLIC OPINION Where Weak/Strong Media Paradigms Converge Gabriel Weimann

The person-to-person influence serves as a bridge over which formal media extend their influence. (Lazarsfeld, Berelson, & Gaudet, 1948, p. 14)

Introduction The impact of mass media on public opinion is one of the most debated issues in communication research. During the past several decades, the lingering debate over the power of mass media has gone through several paradigm shifts. The early hypodermic needle and magic-bullet models of the 1920s and 1930s, attributing almost omnipotent power to the mass media, were replaced by a paradigm based on the ideas that Lazarsfeld and his colleagues in Columbia University’s Bureau of Applied Social Research put forth in The People’s Choice and subsequent studies (e.g., Lazarsfeld, Berelson, & Gaudet, 1948; Lazarsfeld & Merton, 1943; Lazarsfeld & Merton, 1948; Katz & Lazarsfeld, 1955). Media effects were seen as much more sophisticated and complex in nature than previously assumed; they were, moreover, mediated by social networks, opinion leaders, and the selective nature of media consumption and informational diets, all of which reinforced rather than changed existing attitudes (Scheufele & Tewksbury, 2007). The 1970s marked the second major paradigm shift in research on political communication when Noelle-Neumann’s (1983) call for the return of powerful mass media coincided with George Gerbner’s (Gerbner & Gross, 1974) notion of cultivation, and the early studies on agenda setting (McCombs & Shaw, 1972). All three traditions assumed that mass media had strong effects on audiences, or at least assumed so at their early stages. Then, the 1980s and early

162 Gabriel Weimann

1990s saw the positing of “negation models,” such as priming and framing, which were based on the idea that mass media had potentially strong attitudinal effects, but that these effects also depended heavily on predispositions, schema, and other characteristics of the audience that influenced how people processed messages in the mass media. However, the dominant paradigm associated with Lazarsfeld and the Columbia University tradition was mistakenly and consistently associated with “weak media”; the paradigm’s concern with factors mediating, channeling, or limiting media effects was confused with claims of “no effects.” One such example of regrettable misunderstanding and misinterpretation pertained to the concept of opinion leaders. It is my argument that not only were opinion leaders not originally presented as barriers to media effects but, as numerous studies have since revealed, their functioning as mediating between mass communication sources and the public was, in fact, said to enhance and promote media power. Thus, as this chapter suggests, opinion leadership constitutes a bridging concept, where weak and strong media paradigms converge. Noelle-Neumann’s work on opinion leaders is one of the paths leading to this bridge. Opinion Leaders Most of the criticism of the dominant paradigm focused on its minimization of media power. The concept of opinion leaders was often used to illustrate social mechanisms that served to limit media impact. Personal influence, especially that of opinion leaders, it was argued, competed with mass media for influence instead of being seen as a (selective) multiplier or amplifier of media messages. The two-step flow theory asserts that information from the media moves in two distinct stages—from the mass media to the opinion leaders and from the opinion leaders to their followers (Katz & Lazarsfeld, 1955). Decades of social science research have demonstrated that there is a group of people in any community to whom others look to for help in forming opinions on various issues and matters (Weimann, 1994). Whether called “opinion leaders” or “influentials,” these people literally lead the formation of attitudes, knowledge, and opinions. Three major studies laid the groundwork for opinion leadership theory: the People’s Choice Study (Lazarsfeld, Berelson, & Gaudet, 1948), the Decatur Study (Katz & Lazarsfeld, 1955), and the Drug Study (Menzel, 1981). A more recent description of the opinion leaders suggests that “[o]pinion leaders are more precisely opinion brokers who carry information across the social boundaries between groups. They are not people at the top of things so much as people at the edge of things, not leaders within groups so much as brokers between groups (Burt, 1999, p. 37). The concept of opinion leaders and its measurement were subjected to growing criticism and debate (reviewed and discussed by Katz, 1980, 1988;

Opinion Leadership and Public Opinion

163

Weimann, 1982). The criticism, which focused on several theoretical and empirical weaknesses, led to the development of new methods, scales, and procedures for the identification of opinion leaders and their characteristics. Valente and Pumpuang (2007) categorized about 200 studies that studied opinion leaders into 10 different methods. These methods included the use of key informants, sociometric mapping, self-identification, and observations. One of the more productive attempts was initiated by the German public opinion researcher Elisabeth Noelle-Neumann and the survey research center, the Allensbach Institut für Demoskopie. It was the news magazine Der Spiegel that initiated the process: interested in what they referred to as “the active consumers who set standards in their community,” the managers of Der Spiegel challenged the Allensbach Institute to develop an instrument that would identify influentials. The researchers, directed by Noelle-Neumann, tested numerous questionnaire items related to self-perceived personal influence (Noelle-Neumann, 1983, 1988; Weimann, 1994). These early scales were refi ned after years of pretesting with national samples in Germany. The resulting scale, the Strength of Personality Scale (Persönlichkeitsstärke), was established after a factorial reduction of a 34-item questionnaire administered to a representative sample of 3,542 residents of then-West Germany. The fi nal version of the scale included 10 items, which were later weighted according to their part-whole correlations with the total scale; scores on this scale ranged from 75 to 149. The SP scale was validated by external criteria, as tested in Germany, Israel, and the United States (Weimann, 1991); it was found to be a powerful and valid instrument for identifying and measuring how easily individuals could be influenced. The successful identification of influentials by means of the PS scale, the validating of identification by “external” criteria, and the relative ease of administrating the scale led more researchers to explore its potential. Noelle-Neumann (1999) argued that the dispositional characteristic of personality strength might be of key interest in the study of political action and social capital. Personality strength is conceived as a feature of individuals, a reflection of their confidence in leadership roles, their aptitude for shaping the opinions of others, and their self-perceived impact on social and political outcomes. Such individuals, according to her data, show higher levels of life satisfaction and are more engaged in their communities (Scheufele, 1999). Contemporary SP studies have divided societal units into quartiles of strength, based on individuals’ perceived role as opinion leaders, their sources of advice and the respect they garner (Tustin, van Vuuren, Joubert, & Weimann, 2005; Weimann, Tustin, van Vuuren, & Joubert, 2007). Those in the highest quartile are considered the most confident and the most influential; hence, they are opinion leaders. In an American study, Scheufele and Shah (2000) examined the relationship between strength of personality and social capital, using data collected as part of an annual mail survey conducted by Market Facts and funded by the DDB Needham advertising agency. The researchers reported that the “results

164 Gabriel Weimann

of structural equation modeling revealed that personality strength, an amalgam of self-confidence and opinion leadership, has a relatively strong direct impact on all dimensions of social capital, whereas informational variables have rather weak effects” (p. 107). Opinion Leaders as Mediators From the early conceptualization of opinion leadership, no one has claimed that knowing more predicts opinion leadership: “It is evident, therefore, that subjective interest, or advanced knowledge, in the arena of public affairs—as in the other arenas—cannot operate independently to generate opinion leadership. Along with information and interest, there must be some objective ‘enabling’ factor which makes it possible to translate subjective predispositions into the actual give-and-take of participation and leadership” (Katz & Lazarsfeld, 1955, p. 275). Indeed, the empirical association between opinion leadership and knowledge has been found to be amazingly weak if objective knowledge testing is applied (Schenk & Rossler, 1997). Scherer and Trepte (2010) reviewed studies revealing the weak correlation between measures of knowledge and opinion leadership, suggesting that opinion leaders may rely more on media sources rather than only on personal knowledge. The reliance of opinion leaders on the mass media for information and guidance reveals their “mediating the media” function. Opinion leaders gain information from the media, and the results of numerous studies support this relationship. According to the Decatur study, for example, opinion leaders use more media and are more influenced by them than non-leaders (Katz & Lazarsfeld, 1955). Subsequent studies stated that opinion leaders are more likely to read a newspaper, but not to watch television or listen to the radio (Troldahl & van Dam, 1965; Schenk & Rossler, 1997). Other scholars found that opinion leaders do not use media more often than others (Chan & Shekhar, 1990; Lin, 1973; Levy, 1978; Robinson, 1976), but employ media more for “cognitive orientation” (Levy, 1978). They are more likely to talk about media content with friends and relatives (Hamilton, 1971). Therefore, as Scherer and Trepte (2010, p. 121) concluded, “It can be assumed that knowledge is acquired by media use.… Thus we suggest enlarging upon Katz and Lazarsfeld’s (1955) idea of the opinion leaders. They exemplified opinion leaders as being knowledgeable because of media use and as being able to communicate their knowledge due to communicative skills. Our data suggests that media usage may in addition be considered an enabling factor of communication in the flow of information. Opinion leaders use the media more often, but do not consequently know more.” Opinion leaders do not function as automated outlets for mass media messages. They are selective mediators whose personal judgments may enhance media effects but can also weaken, minimize, or block them. This process presents the opinion leader with the opportunity to fi lter the media message and

Opinion Leadership and Public Opinion

165

any effect that stems from it, whether blunting or amplifying. Katz described the concept this way: “Influences stemming from the mass media fi rst reach opinion leaders, who, in turn, pass on what they read and hear to those of their every-day associates for whom they are influential” (1957, p. 61). The selective screening of media messages by opinion leaders is based on their own knowledge and experience in the field of their expertise, as well as on their tendency to comply with social norms (and to reinforce them). Several studies have shown that opinion leaders are more familiar with and loyal to group standards and values than are non-leaders (for a review, see Weimann, 1994). This refers to the fact that opinion leaders are vested with leadership authority by the group and, in order to maintain this position, they have to reflect and exemplify the group’s underlying norms and values. This should be another bridge to the spiral of silence theory: conformity to the group’s norms and attitudes, so essential to the spiral of silence process, relies on the mediation of the majority’s views and values, either directly by the mass media or through the opinion leaders. Opinion Leaders as Agenda Setters An illustrative example of the mediating role of opinion leaders can be found in the empirical pursuit of the suggested two-step flow model as applied to the agenda-setting process of the mass media. Research on this process stems directly from the notion, suggested by Bernard Cohen, that the mass media “may not be successful much of the time in telling people what to think, but it is stunningly successful in telling its readers what to think about” (1963, p. 13). In other words, even though the media may not be very successful in telling us what opinions to hold, they are often quite effective in telling us what to have opinions about (or what not to think about). The agenda-setting tradition, based mostly on aggregate data (media coverage, public opinion surveys) and time-series analyses, often overlooked the individual level of analysis. What is the role played by individuals in the agendasetting process? Do opinion leaders, or more precisely the influentials, play a significant role in the formation of the public agenda? This question highlights the possible interplay of mass communication and interpersonal communication in the agenda-setting process. Weimann and Brosius (1994; Brosius & Weimann, 1996) suggested and tested a two-step flow model in which certain individuals, influentials, identified emerging issues in the mass media and then diff used these issues to others via their personal networks. Based on a series of six national surveys conducted in Germany that measured issue salience and various personal characteristics, this study highlighted the role that certain individuals played in the emergence of a public agenda. The degree to which individuals can be influenced, measured by the SP scale, was found to be a powerful predictor of the formation of personal agendas, with high SP individuals differing from others

166

Gabriel Weimann

in their identification of emerging public issues. One should note that the SP scale is, in fact, a measure of communicative activity and personal influence: it identifies influentials, those individuals who are more involved in interpersonal communication and those who serve as a source of advice and guidance for others (Noelle-Neumann, 1983, 1988; Weimann, 1991, 1994). However, this finding does not suggest that personal influence overpowers or minimizes media effects. According to the above-mentioned studies, influentials were found to be mediators between the media and the public rather than agenda-setters. The Weimann-Brosius studies portrayed agenda setting as a two-step flow, in which certain individuals (influentials) collect, diffuse, fi lter, and promote the flow of information from media to the community, thus highlighting, again, their selective mediating role. New Media, New Leaders, New Convergences Many communication scholars in the current century have suggested a reexamination of the term mass communication: “Developments in computing and information technologies over the last 2 decades have blurred the boundaries between the forms of communicating around which the academic field of communication was developed” (Chaffee & Metzger, 2001, p. 365). The new media certainly challenge the traditional defi nitions of mass communication. The Internet, designed to be decentralized, increases the users’ ability to become content producers and to be more active communicators. The new media environment challenges, too, the traditional distinction between interpersonal and mass communication: the old model of broadcasting to the masses has given way to online communities exchanging messages and information, blurring the line between mass-mediated communication and personal communication. Communication theories and approaches to the study of mass communication will have to adapt to these changes. For example, how will the theory of two-step flow and opinion leadership work in the new decentralized, demassified media environment? The growing number of studies on online opinion leaders and e-fluentials demonstrates the longevity of the concept in the new media context (Burson-Marsteller, 2001; Ghosh & Lerman, 2010; Huffaker, 2010; Lyons & Henderson, 2005). However, the new opinion leaders, the e-fluentials, now operate on a new platform, using new channels, applying new forms of persuasion and combining both mass and personal communication. A fundamental assumption of the two-step flow model is that opinion leaders mediate between mass media and audiences, relying on the mediators’ personal influence. In the new media era, this conceptualization may no longer be valid: the online influentials may obtain their information from personal communication online or spread their messages to a huge audience in a mass-mediated form, using popular online outlets. The new media bring challenges to our old theories and models, as well as

Opinion Leadership and Public Opinion

167

presenting the opportunity to reevaluate, extend, and modify them as in the case of identifying opinion leadership in these media. This is also an issue that highlights the convergence of personal and mass communication channels. Consequently, opinion leaders in the new media environment will only enhance the convergence of the strong/weak media paradigms. This chapter started with the assertion that opinion leaders were erroneously associated only with the weak media tradition; however, that they do, in fact, function both as mediators between mass communication sources and the public, thus enhancing media power, and as selective carriers of media messages who can minimize or mitigate media effects substantially. The new media environment enriches both capabilities of opinion leadership, providing the online influentials with a richer arsenal of platforms and channels to gather information and to spread it to others. Let me conclude with a reference to Marshall McLuhan, who, when predicting the computer revolution, wrote in 1960, “The advent of a new medium often reveals the lineaments and assumptions, as it were, of an old medium” (p. 567). Or, to phrase this observation another way, a half century later: a new communication technology may perform an almost postmodern function of making the unpresentable perceptible or the unbridgeable-bridged. References Brosius, H. B., & Weimann, G. (1996). Who sets the agenda? Agenda-setting as a two-step flow. Communication Research, 2, 561–580. Burson-Marsteller. (2001). Influential Internet users rely on company web sites as they spread word of brands, products and services. Retrieved March 30, 2011, from http://www.imakenews.com/highpointsm/e_article001269102.cfm?x=b11,0,w Burt, R. (1999). The social capital of opinion leaders. The Annals of the American Academy of Political and Social Science, 566(1), 37–54. Chaffee, S. H., & Metzger, M. (2001). The end of mass communication? Mass Communication and Society, 4(4), 365–379. Chan, K., & Shekhar, M. (1990). Characteristics of the opinion leader: A new dimension. Journal of Advertising, 19(3), 53–60. Cohen, B. (1963). The press and foreign policy. Princeton, NJ: Princeton University Press. Gerbner, G., & Gross, L. (1974). System of cultural indicators. Public Opinion Quarterly, 38, 460–461. Ghosh, R., & Lerman, K. (2010). Predicting influential users in online social networks. Computers and Society. Retrieved from http://arxiv.org/abs/1005.4882 Hamilton, H. (1971). Dimensions of self-designated opinion leadership and their correlates. Public Opinion Quarterly, 35, 266–274. Huff aker, D. A. (2010). Dimensions of leadership and Social influence in online communities. Human Communication Research, 36(4), 593–617. Katz, E. (1957). The two-step flow of communication: An up-to-date report on an hypothesis. Public Opinion Quarterly, 21, 61–78. Katz, E. (1980). On conceptualizing media effects. Studies in Communication, 1, 119–141. Katz, E. (1988). Communication research since Lazarsfeld. Public Opinion Quarterly, 51, 25–45. Katz, E., & Lazarsfeld, P. F. (1955). Personal Influence. Glencoe, IL: Free Press. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1948). The people’s choice: How the voter makes up his mind in a presidential campaign. New York, NY: Duell, Sloan and Pearce.

168

Gabriel Weimann

Lazarsfeld, P. F., & Merton, R. K. (1943). Studies in radio and fi lm propaganda. Transactions of the New York Academy of Sciences, 6(2), 58–79. Lazarsfeld, P. F., & Merton, R. K. (1948). Mass communication, popular taste, and organized social action. In L. Bryson (Ed.), The communication of ideas (pp. 95–118). New York. NY: Harper. Levy, M. R. (1978). Opinion leadership and television news uses. Public Opinion Quarterly, 42, 402–406. Lin, N. (1973). The study of human communication. Indianapolis, IN: Bobbs-Merrill. Lyons, B., & Henderson, K. (2005). Opinion leadership in a computer-mediated environment. Journal of Consumer Behaviour, 4(5), 319–329. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176–187. McLuhan, M. (1960). Effects of the improvements of communication media. Journal of Economic History, 20, 566–575. Menzel, H. (1981). Interpersonal and unplanned communications: Indispensable or obsolete? In E. B. Roberts, R. I. Levy, S. N. Finkelstein, J. Moskowitz, & E. J. Sondik (Eds.), Biomedical innovation (pp. 155–163). Cambridge, MA: MIT Press. Noelle-Neumann, E. (1983). Spiegel Dokumentation: Persönlichkeitsstärke [Documentation: The Strength of Personality]. Hamburg, Germany: Spiegel Verlag. Noelle-Neumann, E. (1988). Identifying opinion leaders. Paper presented at the 38th ESOMAR Conference, Wiesbaden, Germany. Robinson, J. (1976). Interpersonal influence in election campaigns: Two-step flow hypothesis. Public Opinion Quarterly, 40, 304–319. Schenk, M., & Rossler, P. (1997). The rediscovery of opinion leaders: An application of the personal strength scale. Communications, 22(1), 5–30. Scherer, H., & Trepte, S. (2010). Opinion leaders—Do they know more than others about their area of interest? Communications: The European Journal of Communication Research, 35, 119–140. Scheufele, D. A. (1999). Participation as individual choice: Comparing motivational and informational variables and their relevance for participatory behavior (Unpublished doctoral dissertation). University of Wisconsin at Madison. Scheufele, D. A., & Shah, D. (2000). Personality strength and social capital — The role of dispositional and informational variables in the production of civic participation. Communication Research, 27, 107–131. Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57(1), 9–20. Troldahl, V. C., & van Dam, R. (1965). A new scale for identifying public-aff airs opinion leaders. Journalism Quarterly, 42(4), 655–657. Tustin, D. H., van Vuuren, D., Joubert, J. P. R., & Weimann, G. (2005). An exploratory study on influencer marketing in South Africa. Pretoria: University of South Africa, the Bureau of Market Research. Valente, T., & Pumpuang, P. (2007). Identifying opinion leaders to promote behavior change. Health Education & Behavior, 34(6), 881–896. Weimann, G. (1982). On the importance of marginality: One more step into the two-step flow of communication. American Sociological Review, 47, 764–773. Weimann, G. (1991): The influential: Back to the concept of opinion leaders?.Public Opinion Quarterly, 55, 267‒279. Weimann, G. (1994). The influentials. People who influence people. Albany: State University of New York Press. Weimann, G., & Brosius, H. B. (1994). Is there a 2-step flow of agenda setting? International Journal of Public Opinion Research, 6(4), 323–341. Weimann, G., Tustin, D. H., van Vuuren, D., & Joubert, J. P. R. (2007). Looking for opinion leaders: Traditional vs. modern measures in traditional societies. International Journal of Public Opinion Research, 19(2), 173–190.

15 VERNACULAR RHETORIC, CONSIDERED PUBLIC OPINION, AND MODES OF PUBLICITY Gerard A. Hauser

Publicity is the bedrock of a democracy. The viability of a political public sphere depends on open access to information and the media plus freedom of expression to publicize opinions and engage in informed public deliberation. Without it, the political subject is reduced to a monolinguist who communicates desires, interests, and ideas to immediate surroundings but is denied an audience of interlocutors or a public capacitated to effectively respond. As Elisabeth Noelle-Neumann (1984) has instructed, fear of isolation may encourage public expressions of conformity but it also encourages seeking information that will confi rm whether concerns, and quite possibly views, are shared. The possibility of a political public forming, discovering its interests, and expressing its opinions in order to authorize action is foreclosed without publicity. The rhetorical theory of public spheres in general, and the political public sphere in particular, exceeds theoretical significance in its capacity to model actual communicative practices that lead to considered public opinion. Considered public opinion (CPO), in turn, depends on publicity. CPO is different from survey results, which may index mood, popularity, or widespread opinion in response to exigencies of the moment without necessarily discriminating the immediate response to a question from a conclusion arrived at after gathering information, engaging in social dialogue, and giving the issue some thought (see Fishkin, 1995). Moreover, CPO can perform public work only insofar as it has traction with legislators and public officials who make and administer laws and public policies, and/or community members who establish the norms of social practice on which relations of equality and justice rely. However, as Jürgen Habermas (2006, p. 417) has pointed out, considered pubic opinions are hard to pin down.

170 Gerard A. Hauser

The difficulty lies in the move from understanding public opinion as what polls report to understanding it as a communication concept. Public opinion conceptualized as CPO about a public problem that shows tendency and weight is usually linked to the deliberative model, which typically invokes norms of rationality to assess the reasons that lead people to hold it as considered. However, these reasons circulate in mundane interactions of ordinary citizens that do not necessarily conform to norms of rationality and are not easily captured by polls or reflected in elite discourses, which makes them difficult to track. Norms of rational deliberation also can prove tricky, especially when it comes to determining what counts as publicity and what counts as public opinion. Rational deliberation sequences the path from publicity to opinion formation by privileging arguments advanced by leaders and spokespersons for a cause, since these are the publicly available interactions attempting to defi ne public problems, determine issues, regulate who gets to speak, what counts as evidence, and stake out positions (Gusfield, 1981). However, leaders are elite voices and interpreting their views as representative of public opinion can be problematic. Although leaders are expected to speak for those they lead, discourse can take a radical turn when leaders leave the room (Touraine, 1983). By the same token, the charisma of elite voices with media access can be a misleading indicator of CPO. Verbally pugnacious pundits who are adroit at “mixing it up” exhibit a form of argumentativeness that can charm and persuade through perverse displays of virtuosity, and attract the sort of admiration that gives them authority. Without necessarily holding formal leadership of a movement, pundits, colorful commentators, and pugnacious spokespersons can deflect captive viewers from weighing arguments and reduce them to a demographic drawn to the magnet of spectacular rhetoric. Their value is transformed from ordinary citizens whose opinions matter into market segments for programming geared more toward entertainment and concerned more with securing ratings that translate into advertisement dollars than with public service through critical issue analysis. Moreover, the increasing ascendance of new media allows snappy bloggers, catchy YouTube streams, and creative tweeters to gain (instant) celebrity status and credibility among those who access their posts, again without necessarily subjecting them to critical publicity. Let us call the foregoing account of circulating ideas publicity1. It refers to the views of elites, leaders, spokespersons, media experts, pundits, and in some cases bloggers who have access to media and the mechanisms for publicizing them. Moreover, it publicizes opinions in ways that enact political spectacle, thereby giving publicity1 a performative dimension. Certainly exchanges among groups, spokespersons, intellectuals and others are important for gauging tendencies of thinking and sentiment, but not exclusively so. The media’s power to control political learning is not absolute. There can be pushback to this mediatized arena of publicity. This pushback is captured by the publicity principle, which challenges the assumption that

Vernacular Rhetoric, Considered Public Opinion, and Modes of Publicity

171

representatives, on average, are wiser, or more committed to the common good than the ordinary citizen (Luban, 1996). Let us call this publicity2. To be sure, vernacular deliberation circulates ideas propagated by official voices, but it also circulates ideas that question their wisdom. Importantly, these moment-bymoment interactions, as well as more elaborate ways by which ordinary citizens publicize their concerns, share information, and exchange their opinions in their ongoing give and take on public problems (Hauser, 1999)—the exercise of their social skin, as Noelle-Neumann would have it—slip beneath the radar of research that monitors public opinion based on opinion polls and deliberation among elites and journalists. The rhetorically salient terms that ordinary citizens use in their unstructured, everyday deliberations are not necessarily isomorphic with published opinion, but nonetheless are capable of exerting the “gentle violence” (Bourdieu, 1977, p. 192) that resides in their power to legitimate action. Publicity1, Publicity2, and (Re)reading Public Rage Everyday communication on matters that intersect their interests is among the ways ordinary citizens perform citizenship. These mundane interactions influence public opinion by circulating ideas through diffused and fragmented exchanges that circulate mouth-to-ear. Their form is like a meme, emerging as fragmented tendencies that reflect concerns, attitudes, developing ideas, bodily performances, material forms, and so forth with a capacity to induce shared responses. To understand publicity2 requires examining the way arguments and other means by which ordinary citizens circulate concerns that matter within the political public sphere, how they engage official and elite rhetorics and quite possibly demystify the relationship with power these rhetorics often conceal, and how they come to awareness of belonging to a collectivity of thought and sympathy that has direction, weight, and endurance. Gaining insight into how publicity2 functions is complicated by its diff use and fragmented character. The capacity of the publicity principle to function as a communication mechanism through which CPO emerges is further complicated when publicity2 is overwhelmed by publicity1. Contrasting macro and micro level engagement between these two modes of publicity can give us a better understanding of their interaction. Publicity Goes Macro At the macro level, as spiral of silence theory would predict, the problem of publicity2 being overrun by publicity1 is illustrated by the great public divide of blue vs. red rhetoric that leaves no room for compromise. Media coverage of vituperative rhetoric publicizes political spectacle that encourages public rage fueled by fear and resentment. Publicizing displays of rage circulates a

172 Gerard A. Hauser

form of public opinion that invites ideologically inscribed rhetoric, which fuels extreme political and social sentiments, such as the “all or nothing” positions espoused by the Tea Party movement and measures to “crackdown” on illegal immigrants. For those who hold extreme positions, coverage of political rage supporting their position tends to reinforce the perception that they are widespread and encourages a willingness to express them with vehemence. For those not on the extremes, the relative absence of moderate views communicated with temperance discourages public expression outside the closed circle of those known to share them. By monopolizing the public sphere, extreme rhetoric skews perceptions of political sentiment because it dampens discussion that might lead to discovery of common ground. On the other hand, a macro perspective on events of the Arab Spring, with reports emphasizing the role of YouTube, Facebook, blogs,and tweets in coordinating resistance and creating a sense of solidarity across distributed space presents a more optimistic picture. By the end of 2011, this historic development served as a reminder of 1989’s lesson that spontaneous political action, as witnessed then in Central and Eastern Europe, can defy the weight of authoritarian rule, the influence of propaganda, and the crippling force of fear. Thus, we must acknowledge confl icting constructions of public opinion. Media coverage of ordinary citizens performing displays of fear and resentment and of crowds occupying political spaces in the capitals of Arab countries whose rulers had a history of repression suggest that visual displays of affect may say more about the power of spectacle than public opinion (Farrell & Goodnight, 1981). The Arab Spring, on the other hand, raises the question of why masses were mobilized in a seeming instant, how they knew not just where to go and what to do when tweeted, but had shared political desires that led to acts of solidarity. Was this more than spectacle? Had a considered public opinion formed? We can fi nd some answers to these questions by inspecting the actions and interactions of ordinary citizens at the micro level to discover how they make sense of their world in order to act within it. Publicity Goes Micro The everyday discourse of ordinary citizens is distinct from the rhetoric of institutional and official forums. Their micro-practices of moment-by-moment exchanges are resonant with value and belief systems, communal norms, and shared narratives by which community members orient to one another and their mutual problems. The spiral of silence theory highlights these mundane exchanges as providing the cues we detect in our ongoing endeavor to determine what expressions of sentiment are safe and/or expected. Their mundaneness notwithstanding, they are intensely political and exert social pressure to conform to the habitus of the community (Bourdieu, 1977). Collectively they form a vernacular rhetoric (VR) that stands apart from official languages used

Vernacular Rhetoric, Considered Public Opinion, and Modes of Publicity

173

in the public transactions typically marked by power relations—such as commerce, education, governance, law, and the professions. VR is not the product of design or a performance of elite status, such as a public address or essay. It refers to interaction within a discourse community that depends on local knowledge, concerns, meanings, modes of arguments, value schemes, logics, traditions, and the like shared among ordinary citizens and enacted through everyday discourse (Hauser 1999, 2011). Because VR stands outside realms of authority and power that require special training, such as the professions, it frequently reflects local awareness of power differentials and contains localized expressions of resistance by those outside power and resistant to being regulated by it (Cintron, 1997; Scott, 1990). These locales provide ample evidence of VR circulating information, attitudes, values, beliefs, and norms that provide publicity2 and encourage expressions of solidarity we might regard as expressions forming CPO. These micro moments of publicity2 contain elements of shared communal narratives that can explain otherwise elusive moments of solidarity. My point is illustrated in Melissa Harris-Lacewell’s Bibles, Barbershops, and BET (2004). Harris-Lacewell is interested in how everyday talk among black Americans informs black political thought.1 The chapter devoted to “bibles” examines the discourse between Rev. Carl W. Kenney, III, pastor of Orange Grove Missionary Baptist Church (OGMB), and congregants who were included in her interview sample. The members of OGMB sense their political, social, and economic condition is often in tension with the values, beliefs, and attitudes of the surrounding white Raleigh-Durham (North Carolina) community. This tension surfaces in their shared sense of vulnerability2 and how vulnerability functions in forming considered opinion. For example, Rev. Kenney, a former drug dealer who had served time, uses the pulpit to reflect on his personal vulnerabilities. And Kenny’s weekly column in the Durham Harold Sun often ponders the vulnerability of a black community to unfair treatment by a system disrespectful of their ways. The congregants’ perceived vulnerability, meanwhile, comes through in everyday talk about the minister, OGMB, and the community in which they live. Vulnerability suggests a line of analysis extending beyond community and political positioning. Vulnerability is not an emotional state, such as fear or resentment, but a state of being. Accordingly, Kenney and the members of OGMB respond with plans of action intended to defend themselves and their interests, such as a boycott of Phatt Boy beer, using the firing of a school principal to highlight how racism limits opportunities for black Americans, and sermons drawing on stories of the Israelites reaching the promised land and from the history of African Americans to bolster perseverance in the face of adversity (Harris-Lacewell, 2004, p. 47). For the members of OGMB, vulnerability provides an analytic frame that goes beyond the political agency of mood and fi xation on visual and verbal

174 Gerard A. Hauser

spectacle that distorts political perception and the opinion formation that witnessing political rage invites. The discussion of vulnerability by the members of OGMB offers Harris-Lacewell and her readers’ insight into the formation of CPO reflecting an emergent political ideology. Publicity3 Through Interaction of Elite and Vernacular Voices That everyday communication offers street level publicity to opinions is implicit in Noelle-Neumann’s discussion of our social skin. Equally consistent with Noelle-Neumann, VR also can perpetuate a local mentality that reduces complex problems to “us vs. them” and is blind to the larger network of interdependencies inherent in globalized problems. Moreover, as the influence of Rush Limbaugh and Fox News on U.S. politics illustrates, publicity1 can infi ltrate the everyday under the guise of the publicity principle through caricature and/or misrepresentation of official rhetoric and encourage a skewed perception of the political positions and policies of their targets. Consequently, VR’s circulation of publicity2 is subject to the same distortions as Noelle-Neumann noted that result from fear of isolation and that encourage congregating with like-minded individuals while remaining silent in the face of perceived opposition. Publicity2’s distinctiveness, however, lies in decoupling a discourse-based understanding of public opinion from media representations and elite or official rhetoric. It extracts public opinion from natural state communication by ordinary citizens on matters that they consider important, expressed in their own language, embedded in the narratives of their lived experiences and rooted in political values and perceived states of being rather than emotions of the moment aroused by stirring rhetoric of those with vested interests. In practice, the distinction between publicity1 and publicity2 must be tempered by recognition that moment-by-moment circulation of information, attitudes, values, and beliefs through which the community’s CPO emerges does not occur in isolation. The line of thought represented by publicity2 can be extended to participatory modes, such as new media (see Howard, 2008, 2010), that intersect the vernacular with the official. Let us call this intersection publicity3. Publicity3 refers to the hybrid dialogue and dialectic of interaction between vernacular and official rhetoric. It can occur whenever the VR of publicity2 is noticed and responded to through publicity1 or vice versa. Caller-expert exchanges on NPR, citizen testimony at public hearings, Internet fact-fi nders who correct false information, and public criticism of inappropriate words and deeds of public figures and officials are exemplary of publicity2’s power to construct a dialogue with elite voices, interrogate their opinions in ways that call for responses, and form and communicate CPO to decision makers in ways that are consequential, as was illustrated when Limbaugh lost advertisers and

Vernacular Rhetoric, Considered Public Opinion, and Modes of Publicity

175

subsequently apologized following a public uproar over his remarks portraying a Georgetown law student as a slut for saying religious institutions should cover birth control in health insurance. Finally, as a subset of publicity2 and publicity3, the possibility always exists of apparent forms of hybridity, or publicity4. Elites can speak as if they are of the people, employing its vernacular as a mode of misidentification intended to garner votes but not to advance the people’s will. Moreover, the resources capable of media control can make it seem as if there is dialogue and even dialectic as a means of control, such as “ideological conservatives” claiming they spoke for “dispositional conservatives” in refusing to compromise in negotiations on raising the federal government’s debt ceiling during summer 2011, when polling data indicated the majority of American’s, including Republicans and independents, wanted them to compromise (Newport, 2011). Admittedly there is often a fi ne line between publicity3 and publicity4. Still, the volatility of hot-button issues, such as vaccination mandates, illegal immigrants, and public entitlements requires alertness to attempts at misidentification for partisan gain. Conclusion Noelle-Neumann’s formulation of the spiral of silence theory argues that our social skin detects circulating opinions, what is safe to say, and when we may or even must express it. Although its main concern is whether people talk and why, while rhetoric’s is with the content of discourse, the theory is well suited to rhetoric studies. VR especially requires a concept of public opinion that has traction in the analysis of society’s ongoing conversations. I have extended her notion through an initial formulation of four types of publicity that I believe are at work in shaping public opinion. The ordinary citizens’ moment-bymoment vernacular dialogue offers a lens for moving beyond the political emotions that are often the focus of attention in institutional and leader discourse and encourages consideration of more local exchanges to uncover how political subjects see themselves in relation to their social environment. These non-institutional performances, whether in the mundane exchanges of the everyday, in the streets, or the hybridity of mutually inflecting institutional and vernacular voices, remind us that publicity lies at the center of political agency. It calls us to look at moments of social and economic unrest, and how those who are engaged explain themselves, what elements constitute the desire from which politics springs, the role that fear of isolation and the gentle violence of publicity1 and publicity2 to speak up or be silent. In these ways, Noelle-Neumann’s reflections on the spiral of silence and the consequences of our social skin enter into the work of rhetoricians concerned with how rhetorical practices are significant in shaping our social lives.3

176

Gerard A. Hauser

Notes 1. Harris-Lacewell’s work is part of a developing cross-disciplinary body of literature that examines everyday talk in order to discover what people think about their social and political circumstances and why they hold these beliefs. See, for example, Bryan (2004), Tracy (2010), and Walsh (2004). 2. I am indebted to students in my graduate seminar in vernacular rhetoric during the fall semester 2011 for a provocative discussion of Harris-Lacewell’s book that helped me explore her observations on vulnerability. 3. An expanded version of this chapter was presented at the annual meeting of the National Communication Association, New Orleans, Nov. 17‒20, 2011.

References Bourdieu. P. (1977). Outline of a theory of practice. (R. Price, Trans.). Cambridge, England: Cambridge University Press. (Original work published 1972) Bryan, F. M. (2004). Real democracy: The New England town meeting and how it works. Chicago, IL: University of Chicago Press. Cintron, R. (1997). Angels’ town: Chero ways, gang life, and rhetorics of the everyday. Boston, MA: Beacon Press. Farrell, T. B., & Goodnight, G. T. (1981). Accidental rhetoric: The root metaphors of Three Mile Island. Communication Monographs, 48(4), 271–300. Fishkin, J. S. (1995). The voice of the people: Public opinion & democracy. New Haven, CT: Yale University Press. Gusfield, J. R. (1981). The culture of public problems: Drinking-driving and the symbolic order. Chicago, IL: University of Chicago Press. Habermas, J. (2006). Political communication in media society. Communication Theory, 16, 411–426. Harris-Lacewell, M. (2004). Barbershops, bibles, and BET: Everyday talk and black political ideology. Princeton, NJ: Princeton University Press. Hauser, G. A. (1999). Vernacular voices: The rhetoric of publics and public spheres. Columbia, SC: University of South Carolina Press. Hauser, G. A. (2011). Attending the vernacular: A plea for an ethnographical rhetoric. In F. Girke & C. Meyer (Eds.), The rhetorical emergence of culture (pp. 157–172). Oxford, England: Berghan Books. Howard, R. G. (2008). Electronic hybridity: The persistent process of the vernacular web. Journal of American Folklore, 121, 192–218. Howard, R. G. (2010). The vernacular mode: Locating the non-institutional in the practice of citizenship. In D. C. Brower & R. Asen (Eds.), Public modalities: Rhetoric, culture, media, and the shape of public life (pp. 240–261). Tuscaloosa, AL: University of Alabama Press. Luban, D. (1996). The principle of publicity. In R. E. Goodin (Ed.), The theory of institutional design (pp. 154–198). Cambridge, England: Cambridge University Press. Newport, F. (2011, July 18). Americans, including Republicans, want debt compromise. Gallup. Retrieved from http://www.gallup.com/poll/148562/americans-including-republicansdebt-compromise.aspx Noelle-Neumann, E. (1984). The spiral of silence: Public opinion — Our social skin. Chicago, IL: University of Chicago Press. Scott, J. (1990). Domination and the arts of resistance: Hidden transcripts. New Haven, CT: Yale University Press. Touraine, A. (1983). Solidarity: The analysis of a social movement: Poland 1980–1981. (D. Denby, Trans). New York, NY: Cambridge University Press. Tracy, K. (2010). Challenges of ordinary democracy: A case study in deliberation and dissent. University Park, PA: Pennsylvania State University Press. Walsh, K. C. (2004). Talking about politics: Informal groups and social identity in American life. Chicago, IL: University of Chicago Press.

PART IV

Intercultural Contexts

This page intentionally left blank

16 WORLD PUBLIC OPINION Frank Louis Rusciano

World public opinion reflects the processes of the spiral of silence in two fundamental ways: fi rst, because there is a global spiral of silence that operates in world affairs, and second, because nations, like individuals, fear isolation. These parallels between the spiral of silence and world opinion are evident in their theoretical roots, international media research, and global polling results. Elisabeth Noelle-Neumann traces the roots of many of her theory’s concepts to classical political philosophers, particularly the British empiricists John Locke and David Hume. When Locke (1640) outlines three types of law—divine law, civil law, and the law of public opinion—he notes that the last has the greatest sway over human behavior, since no one easily escapes “the punishment of their censure and dislike … of the company he keeps, thereby enforcing this law of ‘opinion or reputation’” (as cited in Noelle-Neumann, 1984, p. 72; also see Chapter 2, this volume). David Hume (1791/1792) notes similarly that “Our reputation, our character, our name are sources of consideration of vast weight and importance; the other causes of pride; virtue, beauty and riches have little influence, when not seconded by the opinions and sentiments of others” (as cited in Noelle-Neumann, 1984, p. 75, also see Chapter 2, this volume). Historical Roots of World Opinion Theorists concerned with the notion of world opinion have spoken of it in a similar manner, although the subjects of their observations were nations and not men, and international law and not the forms Locke describes. Cardinal Richelieu argued in 1630 that states can be considered like individuals concerned with public opinion and reputation. As such, states strive to avoid a negative reputation and isolation in the international community (Niedermann,

180

Frank Louis Rusciano

1995, p. 280). Richelieu’s rather sophisticated analyses of world opinion are followed historically by those of Kant and Bentham, who argued in the fi nal decade of the 18th century that “public opinion is peace-loving … [and] public opinion is a useful instrument, the main driving force to achieve international organization” (Herberichs, 1966, p. 627). Other theorists viewed international law in a similar fashion, and even though they dealt with nations and not individuals as Locke did, the punishment for violations was the same. In 1823, James Mill, a disciple of Bentham’s, described processes of international opinion which would be supported by the approbation of various nations; he argued that if a code of international law were published, “the intelligence of the whole world being brought to operate upon it, and suggestions obtained from every corner, it might be made as perfect as possible … the eyes of all the world being fi xed upon the decision of every nation with respect to the code, every nation might be deterred by shame from objecting to any important article in it” (as cited in Herberichs, 1966, p. 634). These thoughts in turn influenced writers in the early 20th century, who recognized at the fourth Hague Convention (1907) the “principles of the law of nations, as they result from the usages established among civilized peoples, from the laws of humanity, and the dictates of public conscience” (as cited in Herberichs, 1966, p. 634). Due to this timing, world opinion has been in the past associated with Wilsonian idealism after World War I; inspired by the theorists in the early portion of the century, its proponents predicted it would restrain the behavior of nation-states (see Davison, 1973). The tragic circumstances of the post-World War I 20th century would undermine this idealism, and with it, the notion of world opinion and the idea that it would restrain nations’ worst behaviors. Further, the advent of the Cold War transformed the concept of world opinion into a propaganda tool wielded in the competition between liberal democratic and Communist systems. Both sides would claim to have moral authority for their policies based upon the judgments of most nations. While individuals have not become visibly more idealistic in their perspectives on global affairs, the concept of world opinion has increased in power due to other circumstances. A clue to these conditions lies in a quotation from Mowlana (1986), which he intends as a criticism of the concept. He argues that it has become “fashionable in the literature to apply a variety of terms to the world stage as a whole, with phrases such as ‘international community’ and ‘international system.’ It is, however, doubtful whether the aggregation of states alone possesses the common values and assumptions, which are by defi nition the essential conditions of community” (p. 176, emphases added). A “world stage” implies a new arena for discourse concerning world opinion—which, in turn, parallels Noelle-Neumann’s notion of the “public” aspect of public opinion. Partially based on Habermas’s writing, she describes four defi nitions of public: of or pertaining to the state (used in terms like “public policy” or “public administration”); legal (used in terms like a “public park” to

World Public Opinion

181

which all are allowed access); political (pertaining to the public good) and representation (as in the publication of a book). Noelle-Neumann added a “socialpsychological” interpretation that implies that one is exposed to the judgment of others unknown to the individual. The reference to a “world stage” upon which actions are observed and judged internationally implies a “global public” where nations or individuals are exposed internationally to the opinions of others (Noelle-Neumann, 1984, pp. 61–62). While it is a truism to state that global communications have changed the conduct and manner of international relations, many actions that were once taken in secret are now carried out in full view of global media networks. Further, absent the Iron Curtain, which implied secrets that needed to be kept from those on the opposite side, freer information flows expose the actions of citizens, nations, and leaders to international judgment. These first historical milestones create a public for world opinion due to the new openness for information dissemination, and the new opportunities it provided for international discourse. It is within this new arena for global discourse that the processes of world opinion can be observed; for this reason the analyses of the concept began with newspapers using the concept from various nations. A second problem arises, however, because previous interpretations of world opinion have as their subject the nation. Hence, it is nations that are isolated, and it is through the association with nations that individuals feel the effects. However, Christopher Hill notes that there exist two important and diverging meanings of world opinion. These are the opinion of states, individually and collectively, and the opinion of people, beyond their national identities (1996, p. 117). The following discussion deals primarily with the opinions of states, characterizing nations as the relevant actors in world opinion. The opinions of individuals enter the discussion through the negotiation of national identity, where a country’s international reputation impacts upon individual citizens’ perceptions of their nation. This chapter turns to the opinions of individuals in the concluding section. Disaggregating World Opinion: Primary Components The following is a brief summary of newspaper content analyses conducted over a 19-year period, from 1985 through 2011. These studies included newspapers as diverse as The New York Times, The International Herald Tribune, The London Times, The Irish Times, The Hindustan Times, The China Daily, The People’s Daily, The Times of India, the Frankfurter Allgemeine Zeitung, the Nigerian Guardian, Pravda, The Tehran Times, the Argentinian Nacional, and the Dominican Listin Diario (see, for example, Rusciano, 1996, 1997, 2001; Rusciano, FiskeRusciano, & Hill, 2004; Rusciano, Fiske-Rusciano, & Wang, 1997; Rusciano & Pollock, 1997). The format was the same in each analysis; news stories and editorials containing references to world opinion were studied using a predesigned codebook While the newspaper’s national origin does not determine

182

Frank Louis Rusciano

discourse, it provides clues to how certain issues are discussed and framed. Analyzing newspapers of record or the elite press provides some general guidance about elite perceptions of world opinion; the purpose here is not to measure world opinion, but rather to derive a matrix by which to compare different international elite media’s usages of the concept. The studies reveal six basic components of a terminology for discussing media interpretations of world opinion. Not all six components of world opinion are used in the same way across international newspapers; comparative and factor analyses reveal that different nations’ newspapers construct world opinion from these components in keeping with the historical era and, their values and interests globally (see, for instance, Rusciano, 1997; and Rusciano et al., 1997). The components used for analysis are described below. • a moral component, which refers to values shared among nations; • a pragmatic component, which refers to interests shared among nations; • the power of world opinion, which refers to its apparent influence on world events and nations’ behaviors; • the nation’s image, or reputation, in world opinion, as it is perceived by itself and other nations; • the world considered as a unit, such as an international community, which may judge and respond to other nations’ behaviors; and • the threat of international isolation, which operates as a potential punishment for nations which do not heed the dictates of world opinion. Central to these references to world opinion was a moral component, which referred to values shared among nations. In the days of Apartheid, for instance, sanctions were described as important as “moral gestures of foreign policy” even if they could not alone bring down the regime; other comments were subtler, criticizing “the developed world” for having “lost its sense of the sacred.” Alongside considerations of morality, there was a pragmatic component which referred to interests shared by all nations; during the Cold War period, The International Herald Tribune noted in an editorial that the “Soviets … deserve credit for rational calculation of important international interests” (New York Times, 1986). More recently, regarding oil markets, a report from The 2nd Workshop on the Interactions between Physical and Financial Energy Markets, held in Vienna, Austria, “commended the international efforts to increase market data transparency” (2011, p. 1). Of the two, the moral component is more central to world opinion. It appears more commonly than the pragmatic component, and is at the center of a nexus of relationships with the other elements of the concept: the power of world opinion, the internal and global images of the nations involved, the world as a unit, and the threat of international isolation. World opinion derives its power in part because of its moral imperatives; nations and leaders do not wish to offend generally accepted values of other nations and leaders. One may, of course, question

World Public Opinion

183

whether any group of countries that is culturally or historically diverse could ever agree upon moral imperatives. In response, I would ask whether nations or leaders could legitimately advocate genocide or terrorism without earning the general opprobrium of their peers in the international community. The term “international community” is perhaps the most prominent example of another element of world opinion, the world considered as a unit. While this and other similar terms such as “the community of nations” or “the world community” may be loosely defi ned, they suggest a body of nations which observe and judge the actions of their peers in the international realm. This implies that some form of world opinion regarding proper moral conduct is a necessary, if not sufficient, condition for the existence of any entity approaching a community on the global level. All communities require the common ties defi ned by public opinion. Noelle-Neumann (1984) describes the three steps to a spiral of silence that defi nes communities as: the ability of individuals to tell when an opinion is rising or falling in popularity in a population; their reactions of more confident speech or silencing, respectively, in response to this “quasi-statistical sense”; and the threat of isolation that makes individuals heed the judgments of public opinion (pp. 61–62). For an international community to exist, it too must have the capacity to defi ne its borders and punish transgressors. It is for this reason that the components of a nation’s image or global reputation, and the threat of international isolation are important. International isolation may involve such tangible actions as banning nations from international bodies or competitions, boycotting goods, or recalling ambassadors. But often the international community relies more upon indirect, or “soft,” sanctions including criticizing the image of the targeted nation, questioning the nation’s reputation, and other more abstract strategies. These actions might seem of doubtful effectiveness, especially if the country in question has considerable military or economic power. Yet, there is a proven relationship between a nation’s reputation in world opinion (Fremdbild) and its citizens’ perceptions of itself and its national consciousness (Selbstbild) (Rusciano & Ebo, 1998), and national identity is in large part a negotiation between these two (often-confl icting) forces (Rusciano, 2003). A decline in a nation’s international reputation may therefore portend a parallel decline in its citizens’ evaluation of their country, and diminished national pride. The effects are likely to be significant, especially if one accepts Greenfeld’s argument that national identity is “fundamentally a matter of dignity” necessary to ward off feelings of helplessness in a post-modern age (1992, p. 487). Here, one detects the fi rst step in the spiral of silence, as citizens of a nation must have the capacity to detect when their nation’s reputation is increasing or declining in world opinion for this punishment to be effective. In South Africa, the government during Apartheid had “a very real fear of isolation and the crippling effects it could have on white control of the country” (Karunaratne, 1981, p. 25). Further, during the run-up to

184

Frank Louis Rusciano

the Persian Gulf War in 1990, R. W. Apple noted that the United States’ actions in forming an international coalition against Saddam Hussein were directed towards “penetrating the Iraqi dictator’s tough psychological carapace and making him feel increasingly shunned, cornered, friendless” (Apple, 1990). Further, recent research shows that citizens are generally aware of rises or declines in their country’s international reputation (Rusciano, 2006, pp. 155–156, 164–165). The indirect effects of negative world opinion may thereby have a more substantial influence than the direct action of boycotts or the recall of ambassadors. A study of Germany in the wake of reunification, for instance, showed Germans who perceived they were “not liked” in the rest of the world were more likely than their fellow citizens to have less pride in being German, to believe that Germans could not be as proud of their country as citizens of other countries, and to not enjoy seeing patriotic symbols such as the flag (Rusciano & Ebo, 1998, pp. 59–88). Herein lies the second step of Noelle-Neumann’s description, for the effective “silencing” of national identity in the form of national symbols or pride in the wake of a negative international reputation follows the pattern of speech or silencing in the face of positive or negative public opinion. The fi nal step in the spiral of silence follows closely upon the second; because world opinion has a capacity to affect the very sense citizens have of their country, nations’ leaders and citizens are compelled to pay some heed to the international community’s judgments. World opinion may not alter a nation’s or a citizenry’s actions, but it becomes one more policy effect to consider. The Global Opinion Process World opinion, like public opinion, is properly approached as a process—in this case, a process involving the components discussed above. The moral component provides a value-driven justification for condemning a particular action; the pragmatic component also contributes to the power of international public opinion to influence events, by convincing nations that what is moral is also in their interests. What is at stake here is the nation’s image in the world, or its reputation in world opinion. Our research has shown that citizens tend to integrate their country’s international image into their construction of national identity. Finally, errant nations or leaders are threatened, or actually punished, with international isolation from the world community, or some other entity that describes the world as a unit. Indeed, one can summarize the global opinion process in a preliminary defi nition of world opinion: World opinion refers to the moral judgments of observers that actors must heed in the international arena, or risk isolation as a nation. This defi nition parallels Noelle-Neumann’s defi nition of public opinion: “opinions on controversial issues that one can express in public without isolating oneself ” (Noelle-Neumann, 1984, pp. 61–62). The threat of isolation tends to silence those who might support the nation being condemned by world opinion. Hence, the discourse convention arises whereby observers attribute

World Public Opinion

185

opinions, attitudes, or judgments to the world—when only one opinion tends to be expressed, it becomes easy (if not entirely proper) to attribute that opinion to an entire entity, be it a public or an international community. The complementary defi nitions follow as the elite media reflect a global spiral of silence. Concluding Thoughts: The Internet, Social Media, and the Idea of a Global Public The previous discussion leaves a lacuna regarding the opinions of people—i.e., whether there exists a global spiral of silence that affects individuals regardless of their citizenship and across the boundaries of their nations. Early research results are mixed. Lemin (2010) argues that the spiral of silence operates more in social media than in social situations, while Neill (2009) argues that the spiral of silence operates less in social media than in social situations. However, the advent of the Internet, blogs, and social media open a public for the opinions of people that provides a rich area for research and discussion. The singular characteristic of these forms of communication is that they remove any authoritative fi lter for the individual to communicate with a mass audience. Formerly, the publisher controlled access to the printed word; the radio station owners controlled access to the broadcast word; and the television or cable station owners controlled access to the broadcast word and image. Individuals who wished to reach a mass audience had no choice but to convince the editorial authority to allow their statements through or purchase access to the media. The advent of the Internet, blogs, and social media has removed such authority. The individual now has access to a potential worldwide audience through their computer. Statements or videos may “go viral” and receive mass attention. The monopoly on media originally imposed by the state in certain authoritative regimes has been undermined by the individual capacity; the Arab Spring uprisings against dictatorial governments were begun and spread primarily through this new media. The monopoly on media originally imposed by ownership is similarly diminished. This phenomenon recalls the spiral of silence in that it exposes the opinions of individuals to a world public without an editor providing an authoritative fi lter; individuals may receive praise or approbation from an international audience. Any posting now has a potential audience of anyone in the world on the Internet; indeed, even communications that were assumed private have a sometimes alarming way of receiving public exposure. The individual’s opinion is now exposed to a global public. Many rightfully decry the loss of editorial discretion as an open invitation to disseminate misinformation. But debates about net positive or negative effects are academic. The emergence of a global public for the opinions of individuals insures that the spiral of silence will remain a relevant tool for analyzing world opinion in new ways in the years to come.

186

Frank Louis Rusciano

References The 2nd Workshop on the Interactions between Physical and Financial Energy Markets. (2011, November). Retrieved from http://www.ief.org/_resources/fi les/content/events/eventcards/2nd-workshop-phys_fi n-iea-ief-opec.pdf Apple, R. W. (1990). New York Times, September 11. Retrieved from http://www.nytimes. com/1990/09/11/world/confrontation-in-the-gulf-bush-gorbachev-inc.html Davison, W. P. (1973). International and world public opinion. In I. de Sola Pool, W. Schramm, F. W. Frey, N. Maccoby, & E. B. Parker (Eds.), Handbook of communication (pp. 871–886). Chicago, IL: Rand McNally. Greenfeld, L. (1992). Nationalism: Five roads to modernity. Cambridge, MA: Harvard University Press. Herberichs, G. (1966). On theories of public opinion and international organization. Public Opinion Quarterly, 30, 624–636. Hill, C. J. (1996). World opinion and the empire of circumstance. International Affairs, 72, 109–131. Karunaratne, J. A., (1981). White South Africa’s response to threats of disinvestment. Ufahamu: A Journal of African Studies, 10(1-2). Retrieved from http://www.escholarship.org/uc/ item/4q11v0h8 Lemin, D. (2010). Public opinion the in social media era: Toward a new understanding of the spiral of silence (Unpublished MA thesis). Gonzaga University, Spokane, WA. Mowlana, H.. (1986). Global information and world communication: New frontiers in international relations. New York, NY: Longman. Neill, S. A. (2009). The alternate channel: How social media is challenging the spiral of silence theory in GLBT communities of color (Unpublished MA thesis). American University, Washington, D.C. New York Times. (1986). Opinion page, February 12. Retrieved from http://www.nytimes. com/1986/02/12/opinion/one-man-who-made-a-difference.html Niedermann, B. (1995). Oeffentliche meinung und herrschaft in der aera Richelieu: Zur verfl echtung von oeffentlicher und veroeffentlicher meinung im absolutismus [Public opinion and domination in the era Richelieu: The intertwining of public and public opinion under absolutism] (Unpublished doctoral dissertation). The University of Mainz, Germany. Noelle-Neumann, E. (1984). The spiral of silence: Public opinion — Our social skin. Chicago, IL: University of Chicago Press. Rusciano, F. L. (1996). Media observations on world opinion during the Kuwaiti crisis: Political communication and the emerging international order. Southeastern Political Review, 24, 505–530. Rusciano, F. L. (1997). First and third world newspapers on world opinion: ‘Imagined communities’ in the cold war and post-cold war eras. Political Communication, 14, 171–19. Rusciano, F. L. (2001). A world beyond civilizations: New directions in research on world opinion. International Journal of Public Opinion Research, 13, 10–24. Rusciano, F. L. (2003). The construction of national identity—A 23-nation study. Political Research Quarterly, 56, 361–366. Rusciano, F. L. (2006). Global rage after the cold war. New York, NY: Palgrave. Rusciano, F. L., & Ebo, B. (1998). National consciousness, international image, and the construction of identity. In F. L. Rusciano (Ed.), World opinion and the emerging international order (pp. 59–88). Westport, CT: Praeger. Rusciano, F. L., Fiske-Rusciano, R., & Hill, C. J. (2004, March). Proceedings of the International Studies Association World opinion on the war with Iraq. Montreal, Canada. Rusciano, F. L., Fiske-Rusciano, R,, & Wang, M. (1997). The impact of ‘world opinion’ on national identity. Harvard International Journal of Press/Politics, 2, 71–92. Rusciano, F. L, & Pollock, J. C. (1997). Media perspectives on world opinion during the recent Bosnian crisis. Current World Leaders: International Issues, 40, 56–72.

17 CULTURAL ORIENTATION AND THE SPIRAL OF SILENCE Sonny Rosenthal and Benjamin Hill Detenber

Public opinion is a product of social context and arises in the formation of different opinion factions, communication among those factions and with the public, and the public acceptance or rejection of the factions’ positions (Price, 1989). Cultural factors can affect both the variety of opinion factions and the communication processes that link them with the public. While some cultures value a diversity of perspectives and unencumbered expression, other cultures strive for unity among perspectives and actively suppress disruptive expression (Van de Vliert, 2011). Still, other cultures contain a mix of values that can result in complex states of public opinion (Lee, 2006; Manaev, Manayeva, & Yuran, 2010). This chapter considers the broad effects of cultural orientation on the formation and expression of public opinion as it relates to spiral of silence (Noelle-Neumann, 1984). Cultural Aspects of Public Opinion The meaning of culture can vary, depending on the scholarly lens through which it is viewed. In this chapter, we adopt a social scientific perspective to draw a specific defi nition of culture from relevant sociocultural literature. According to Kluckhohn (1954), “culture is to society what memory is to individuals” (p. 921); culture contains objective and subjective elements such as knowledge and norms that benefit a society’s survival, which transmit through society via shared language and geographic and temporal proximity (Triandis, 1994). When these elements form meaningful patterns of collective tendencies at the society-level, the result is a cultural orientation. Scholars describe a wide range of such orientations; however, for the purposes of this chapter, we defi ne cultural orientation along two related continua. The fi rst prioritizes

188

Sonny Rosenthal and Benjamin Hill Detenber

self-expression and individual achievement at one end, and social harmony and collective good at the other. Respectively, these concepts of individualism and collectivism defi ne cultural orientations that arise in social beliefs and interactions (Hofstede, 1980; Oyserman, Coon, & Kemmelmeier, 2002). The second continuum relates more viscerally to economic conditions. At one end is a culture of tolerance, trust, and emphasis on subjective well-being that arises in stable, modern societies that have fulfi lled their basic economic needs. At the other end, societies that face economic and political uncertainty prioritize stability and traditional values. Respectively, such self-expressive and survivalist cultural orientations can influence how a society responds to social, economic, or political changes (Inglehart & Baker, 2000). Some scholars consider the cultural ideologies of individualism and self-expression hallmarks of modern, industrialized nations (Fuligni, Tseng, & Lam, 1999; Inglehart & Baker, 2000; Kwak, 2003; Triandis, 1989). In other words, national borders can mark not only geographical and political separations, but also cultural idiosyncrasies; thus, we can ascertain some of the cultural bases of public opinion by comparing public opinion among nations with different cultural orientations. Throughout this chapter, we draw on such cross-national differences in order to explore the spiral of silence across various cultural contexts. Formation of Opinion Before we introduce specific effects of culture on willingness to speak out, fear of isolation, and perception of public opinion—key components of the spiral of silence—we will discuss some fundamental effects of culture on the formation and expression of public opinion. This discussion will lay the foundation for later sections in this chapter that directly address the spiral of silence. Differing norms and values associated with different cultural orientations can shape public opinion toward a variety of topics and policies. For example, in such individualistic societies as the United States, the relatively ubiquitous ethic of self-reliance can spawn negative attitudes toward government-funded welfare programs (Madland, 2010); whereas, in collectivistic societies, the prominence of egalitarian values can produce positive attitudes toward such programs (Blekesaune & Quadagno, 2003). Adamczyk and Pitt (2009) studied the effects of culture on public opinion regarding homosexuality, emphasizing self-expressive and survivalist cultural orientations, as well as several variables related to religion. They found that cultural orientation most strongly predicted attitude toward homosexuality; people in survivalist cultures were more likely than those in self-expressive cultures to register disapproval. Other research has found that survivalist cultural orientations are associated with disapproval of suicide (Stack & Kposowa, 2011) and prostitution (Stack, Adamczyk, & Cao, 2010). Societies generally contain a combination of individualist/collectivist and

Cultural Orientation and the Spiral of Silence

189

self-expressive/survivalist cultural orientations. Interactions between the two continua can produce a mix of public opinion. For example, residents of Arkansas (a fairly poor, Southern state) are simultaneously self-reliant and traditional, and while they express generally negative attitudes toward homosexuality, they are hesitant to endorse state and federal laws that limit the rights of homosexuals (Barth & Parry, 2009). The authors describe Arkansas as an outlier, and thus not representative of broader social tendencies in the United States. However, their fi ndings give a clear example of the interplay of cultural orientations within a population and its effect on public opinion. A similar interplay can occur between cultural orientation and some individual-level variables. For example, Leong (2008) studied the effect of cultural orientation and social dominance orientation on attitudes toward Chinese immigrants in Europe. Social dominance orientation reflects people’s endorsement of hierarchical and unequal social segmentation. Results showed that individualists expressed more negative opinions toward immigrants than did collectivists only at high levels of social dominance orientation. Communication of Opinion Cultural orientation can also affect the communication processes—both interpersonal and mediated—that are integral to public opinion formation. People generally prefer to encounter information that affirms their existing knowledge or opinions (Mojzisch, Schulz-Hardt, Kerschreiter, & Frey, 2008; Strickland, Taber, & Lodge, 2011). They also tend to adopt the beliefs of culturally similar people and to avoid or deny the beliefs of culturally dissimilar people (Kahan, Braman, Gastil, Slovic, & Mertz, 2007; McCright & Dunlap, 2011). Thus, members of a culturally-homogeneous society will express beliefs that support the existing social structure that their culture defi nes. The media with which the public engages as it forms opinion can also be culturally bound. In general, journalists enjoy more freedom of expression in individualistic, self-expressive societies than they do in collectivistic, traditional societies (Kenny & Gross, 2008; Van de Vliert, 2011). However, even in societies that do not restrict freedom of expression, cultural orientation can affect the way journalists frame issues. For example, journalists in South Korea and the United States employed different frames in coverage of the 2007 Virginia Tech shooting (Kwon & Moon, 2009) and the issue of corporate social responsibility (Lee & Kim, 2010), which both pairs of researchers attributed to differences in cultural orientation. According to the 2011–2012 Press Freedom Index, South Korea and the United States have comparably few restrictions on the press, ranking as the 44th and 47th most free, respectively, out of 179 countries (Reporters without Borders, 2012). In sum, culture can affect the formation of public opinion in at least two ways: (a) cultural orientations are inherently bound to specific beliefs and values,

190

Sonny Rosenthal and Benjamin Hill Detenber

and public opinion that arises from within a cultural orientation will reflect those beliefs and values; and (b) cultural orientations can affect the growth and spread of opinion via interpersonal and mediated communication channels. Cultural Effects on the Spiral of Silence The preceding paragraphs depict public opinion as at least partly determined by culture: mainstream beliefs and attitudes reflect the cultural orientation of the society in which they arise. Consequently, members of a society who do not orient themselves with the public majority might be less willing to openly express their own beliefs and attitudes for fear of isolation, and the spiral of silence may occurs as a result (Noelle-Neumann, 1984). Cultural orientation can affect the extent to which the spiral of silence takes root in different societies. Ethics of individualism and resistance to social pressure tend to engender, or at least facilitate, non-mainstream expression; whereas, the prioritization of collectivism and conformity can seed the spiral of silence (Huang, 2005; Manaev et al., 2010). In other words, the spiral of silence is apt to manifest in collectivistic, traditional societies that actively limit public dissent of dominant ideological tenets: when the nail that stands up gets pounded down, the nail prefers not to stand up. This chilling effect occurs both in the expressions of minority opinion factions (Huang, 2005) and in media coverage of culturally relevant issues (Van de Vliert, 2011). Media coverage further promotes the spiral of silence when it creates public perceptions of majority opinion (Scheufele & Moy, 2000). This effect is especially pronounced in societies that prioritize unity of ideology over diversity of expression, and where the media are unlikely to communicate a divergent ideology (Manaev et al., 2010). As with the previous section, we discuss separately the formation of opinion in the public and the interpersonal and mediated communication of public opinion. This section emphasizes the effects of culture on willingness to speak out, and thus describes cultural effects on the spiral of silence. Formation of Opinion An important distinction between individualistic/self-expressive and collectivistic/survivalist cultural orientations is the value they place on independent opinions and behaviors. In general, people in individualistic/self-expressive societies consider dissent a necessary feature of their society, even when it creates systemic disruptions; while people in collectivistic/survivalist societies prefer to minimize dissent or other disruptions to their relatively uniform social fabric. Hornsey, Jetten, McAuliffe, and Hogg (2006) tested the effects of individualist and collectivist norms on preference for group-concordant and

Cultural Orientation and the Spiral of Silence

191

group-dissenting statements. In a pair of experiments, they manipulated university students’ perceptions that the student body has a predominantly individualistic or collectivistic cultural orientation. Next, the experimenters presented participants with fictitious information in which an anonymous student rated a new university proposal in a manner that was either concordant or discordant with how the student body rated the proposal. Across both experiments, results showed that students who self-identified with the collectivistic student body evaluated concordant students more favorably than discordant students. Furthermore, this effect attenuated to a non-significant difference when students who self-identified with the individualistic student body evaluated concordant and discordant students. This study reveals two complimentary points: (a) when individualism is the norm, individualists do not hold negative attitudes toward independent opinions or behaviors, and (a) when collectivism is the norm, collectivists prefer opinions and behaviors that conform to society. The points above demonstrate how cultural orientation affects the social acceptability of behaviors. However, they do not show fully how individual actors will tailor their opinions and behaviors to comply with a cultural climate. Indeed, the spiral of silence does not occur at the ascription of social norms, but rather at the public’s acquiescence to prescriptive norms and subsequent modifications of opinion expression. In other words, norms are, to a large extent, a product of cultural orientation, and people’s willingness to speak out and fear of isolation are products of norms. At least one study has considered directly the effects of cultural orientation on people’s willingness to speak out. Huang (2005) examined several sociopsychological variables to explain why people in Taiwan—a collectivistic society—are less willing to express their opinions in a range of public settings than are people in the United States—an individualistic society. Results showed that, in both cultural contexts, a lack of expressive self-efficacy limited people’s willingness to speak out. Furthermore, people in Taiwan were less willing to speak out when they perceived that their opinion was incongruent with a future majority opinion, and people in the United States were less willing to speak out when they were disoriented about the issue. These fi ndings suggest that perceptions of majority opinion affect willingness to speak out in collectivistic societies, but not in individualistic societies. This study also found that fear of isolation was unrelated to willingness to speak out in both samples. The authors concluded that people in collectivistic cultures are reticent as a means of maintaining social harmony, and not out of a fear of isolation. However, more recent research has drawn a clear association between cultural orientation and fear of isolation. Fear of isolation is a core tenet of the spiral of silence, and the potential effects of cultural orientation on fear of isolation are intuitive: the greater the collectivistic orientation, the greater the personal detriment of social isolation. In individualistic societies, being independent of the mainstream is acceptable;

192

Sonny Rosenthal and Benjamin Hill Detenber

thus, discrepant behavior will not lead to social isolation. Hornsey et al. (2006) explain this effect succinctly: The norm of individualism causes people to decategorize, meaning that they perceive themselves and other group members as individuals. [However,] members of individualist groups are still bound by a group norm, but one in which individualist behavior is normatively prescribed. In other words, attitudes and behavior can be conceptualized (somewhat ironically) as a manifestation of group influence—the result of conformity to an individualist group norm. (p. 59)

Accordingly, people in individualistic societies who act in isolation of the mainstream are actually conforming to the individualistic norm and maintaining a norm-consistent social identity, and research supports this assertion ( Jetten, Postmes, & McAuliffe, 2002). Conversely, people in collectivistic societies who act in isolation of the mainstream contradict the collectivistic norm and risk social isolation. Fear of isolation can be a powerful motivator not to speak out (see Glynn & Huge’s discussion in Chapter 6, this volume), and cultural orientation can affect the strength of this motivation. However, we should also note that several prominent studies show clear attitudinal and behavioral effects of group conformity in individualistic cultures (e.g., Asch, 1956; Nolan, Schultz, Cialdini, Goldstein, & Griskevicius, 2008), which suggests that humans have an innate tendency to think and behave in a manner that conforms to a perceived majority. Notably, Asch (1956, p. 31) found in post-experiment interviews that among college students in the United States, even “the most independent … feared the group would disapprove of them standing out” were they to express a majority-confl icting, but correct response to the experimental task. Despite the apparent human propensity to conform to a perceived majority, recent research provides evidence that sensitivity to fear of isolation varies across cultures. Kim and Markman (2006) measured fear of isolation using the fear of negative evaluation scale (Watson & Friend, 1969), which measures general anxiety related to loss of social approval. Results showed that South Korean university students reported a significantly greater fear of isolation than did American university students, and attributed the difference to Western versus non-Western (i.e., Asian) cultural orientations. Federici, Stella, Dennis, and Hunefeldt (2011) extended this line of research by comparing fear of isolation between the United States and Italy—two Western societies that tend to exhibit different levels of individualism and collectivism. The purpose of their study was to demonstrate that the cultural differences in fear of isolation occur not only between Western and non-Western societies, but also within such societies as a function of cultural orientation. As with prior research, the fear of negative evaluation scale measured fear of isolation. Results showed that

Cultural Orientation and the Spiral of Silence

193

Italians reported a significantly higher fear of isolation than did Americans, and that this difference mirrored that between Western and non-Western cultures, as other studies have reported. In other words, fear of isolation—which tends to reduce willingness to speak out—is a personality factor that appears more often in members of collectivistic societies than in members of individualistic societies. Consequently, members of collective societies are less willing to express divergent opinion than are their individualistic counterparts. Communication of Opinion The spiral of silence begins when members of a minority group are less willing to speak out because of a perceived majority opinion. Because they do not speak out, their opinion becomes underrepresented and the majority opinion acquires a stronger footing. Subsequently, members of the minority group are even less likely to speak out, and this cycle gets reinforced over time. The preceding section addresses willingness to speak out; however, it does not fully explain why perceived majority opinion gains prominence. This latter process occurs largely when a sender asserts the form of mainstream beliefs and attitudes through interpersonal and mediated communication channels. Indeed, the spiral of silence is a theory of communication. We now turn our attention to cultural aspects of communication that can amplify or attenuate the spiral of silence. The Southeast Asian island city-state of Singapore provides a useful example of a collectivistic, survivalist society with a well-defi ned system of media censorship. In recent years, Singapore has invoked its Sedition Act (Cap. 290, 1985 rev. Ed. Sing.) to prosecute publications, utterances, or importation of publications that promote contempt toward the government and the administration of justice, raise discontent among citizens or residents of Singapore, or sew racial and class-based social discord (Lee & Kan, 2009). The spirit of the Sedition Act is to maintain social harmony, and the Singapore news media engage in extensive self-censorship in order to protect themselves from criminal liability that the Act imposes, as well as civil liabilities that similar legislation prescribes (e.g., the 1974 Newspaper and Printing Presses Act). Even non-mainstream channels of communication—blogs, for example—have experienced pressure to conform to principles of self-regulation (Lee & Kan, 2009). Consequently, individuals who wish to air certain grievances in the mainstream media fi nd that they have no venue for such expression (Tey, 2008). Somewhat ironically, the political and media landscape that, in theory, should contribute to the spiral of silence in Singapore also make studying the spiral of silence a difficult task: at least in survey research, the statistical treatment of reticence (i.e., non-response) often requires tenuous assumptions, if researchers are to interpret such data at all. Nonetheless, there is at least philosophical value in reflecting on the sociopolitical barriers to minority expression that can arise in such cultural contexts

194

Sonny Rosenthal and Benjamin Hill Detenber

as Singapore. In addition, regulated venues of opinion expression that exist online offer new opportunities to witness and perhaps measure the influences and variations in speaking out in different opinion environments. Lee (2002) describes the climate of political discourse in Singapore as usually either pseudo participation or partial participation, where members of society do not play an active or central role in decision making. Rather, key decision-makers give the public the opportunity to endorse a pre-determined position about a political issue. A third mode of participation, full participation, reflects “undesirably ‘Western’ ideals of individualism and the push for liberally democratic and human rights, [which] is clearly unworkable in the Singapore context” (p. 105). Lee and Kan (2009) describe Singapore’s approach to censorship as unique among modern, industrialized nations; however, the approach is not unique among similarly collectivistic, survivalist societies. For example, the Chinese government uses media in a similar fashion (e.g., structural restrictions prevent confl icting sources of information) in order to maintain the legitimacy of the political and social system (Stockmann & Gallagher, 2011), and researchers have documented widespread self-censorship among the news media in Central Asian post-communist nations (Kenny & Gross, 2008). Manaev et al. (2010) describe the spiral of silence in such post-communist nations as both an instrument of political control and an outcome of a collectivistic cultural orientation. Perceived Consensus of Opinion The previous section introduced communication as an important factor in building perceived majority opinion. Intuitively, a diverse communication landscape—regarding both interpersonal and mediated channels—can promote divergent viewpoints and give minority opinions some exposure in mainstream conversations. As a result, the public might come to understand a particular issue as having multiple interpretations and a lack of consensus. Conversely, restrictive communication landscapes may promote a limited or singular viewpoint, which might lead to perceptions of a uniform majority and the absence of minority opinions. In either scenario, culture affects perceptions of consensus indirectly via interpersonal and mediated communication. This chapter also considers one direct effect of culture on perception of opinion: the fundamental nature of cultural orientation may predispose individuals to perceive varying degrees of consensus in public opinion. These perceptions should occur both in people who align with the majority and those who align with the minority. We suggest that people in the majority will be willing to express their opinion regardless of their cultural orientation. However, people in the minority will perceive relatively weak public consensus and be more willing to express their opinions in individualistic/self-expressive cultures, and will perceive relatively strong public consensus and be less willing to express their opinions in collectivistic/

Cultural Orientation and the Spiral of Silence

195

survivalist cultures. As cultures develop progressive tendencies, a bias can arise in which individuals perceive widespread consensus that favors traditional values, where in reality, traditional values are in decline (Zou et al., 2009). Kuran (1987) describes such bias as preference falsification, which promotes the status quo much in the way the spiral of silence promotes a majority opinion. According to this perspective, social stigma and other punishments “ensure that most opponents keep their private preferences and beliefs to themselves. As a result, new generations grow up hearing much in favor of the system and almost nothing against it” (p. 663; see also Kuran, 1995). Thus, a cultural orientation that intrinsically seeks to minimize opposition to the established social structures is also likely to build in the public a false perception of consensus on matters of “mainstream” opinions and behaviors. From within this theoretical framework, Zou et al. (2009) conducted a series of four experiments in order to examine the effects of cultural orientation on perceived consensus. Two of the experiments are relevant to this chapter. The fi rst experiment demonstrated the simple effect that university students in Poland—a relatively collectivistic culture—perceived more collectivistic cultural consensus than did students in the United States—a relatively individualistic culture. Furthermore, students from Poland were more susceptible to influence by peer information than were students from the United States. This fi nding speaks to the potential effects of cultural orientation on perceived consensus: people in collectivistic cultures may perceive a relatively uniform role of individuals in society and, partly as a consequence, form opinions or engage in behaviors that reflect perceived collectivistic norms. In the second experiment, Chinese and American university students viewed an image of a lone actor engaging in divergent behavior and indicated whether they believed the behavior was due to dispositional (i.e., internal to the actor) or situational (i.e., external to the actor) factors. Participants also indicated their beliefs that other members of their culture would attribute the behavior to dispositional or situational factors. Thus, the former query served as a measure of personal belief, while the latter served as a measure of perceived consensual belief perceived cultural consensus. Analysis of variance showed that Chinese participants perceived stronger consensus for situational attribution, whereas American participants perceived stronger consensus for dispositional attribution. Further analysis showed that perceived consensus for dispositional attribution mediated the relationship between culture and personal dispositional attribution. In other words, culture affects opinion both directly and via perceived consensus. Thus, an important pathway of opinion formation can occur, for example, when (a) members of individualistic cultures believe that their peers will attribute divergent behaviors to dispositional factors, and (b) this perceived consensus informs personal beliefs that dispositional factors underlie the behavior. As a consequence, members of individualistic cultures may be

196

Sonny Rosenthal and Benjamin Hill Detenber

more accepting of behaviors that are highly self-expressive and draw clearly on individual dispositional factors. These fi ndings have some important implications for the spiral of silence. In collectivistic/survivalist cultures, when individuals perceive consensus that others’ behaviors reflect and respond to situational demands, they might form personal beliefs that derogate individualistic self-expression. These personal beliefs should reflect both internalized cultural norms and also compliance with perceived cultural consensus. We do not have reason to suspect that people in collectivistic/survivalist cultures perceive more consensus about behavioral attributions than do people in individualistic/self-expressive cultures, but only that the nature of consensus varies from one culture to another. Indeed, the effects of perceived consensus on the formation of opinion can vary dramatically across the continua of cultural orientation. In sum, cultural orientation affects the spiral of silence in at least three ways. First, to the extent that members of a society align with an individualistic/selfexpressive cultural orientation, they will be both more accepting of dissent and more likely to express dissenting opinions. Conversely, alignment with collectivistic/survivalist ideals should result in emphasis on conformity of opinion, and less willingness to speak out. Second, members of individualistic/self-expressive societies have relatively liberal media as both a means of expression and source of varied perspectives; whereas, members of collectivistic/survivalist societies are often limited in their means of expression and access to dissenting opinions via the media. Freedom of expression—which is a central feature of individualistic/ self-expressive societies—affects not only the variety of discourse in mainstream media, but also the kinds of opinions people are willing to express in such informal outlets as blogs and social media. Finally, perceived consensus regarding acceptable behavior can directly affect the formation of opinion. The belief that the general public either tolerates or admonishes divergent, potentially disruptive expressions can influence how individuals treat such behaviors. Notably, in collectivistic/survivalist societies, “consensus” may disapprove of expressions or behaviors that bear on individualistic dispositional factors. Consequently, there may be greater fear of isolation and less willingness to speak out among members of these societies. Promising Directions for Future Research We propose three directions for future research. First, the Internet presents new opportunities for democratic participation and overcomes some of the structural restrictions of traditional media (e.g., few senders with many receivers; Balkin, 2004). Some research describes “destabilizing” trends of Internet use in collectivistic societies that give way to strongly individualistic expressions (Liu, 2009; Xu, 2009); a in societies with restrictive media environments, social networking sites offer people a means of expressing dissent with relative impunity

Cultural Orientation and the Spiral of Silence

197

(Attia, Aziz, Friedman, & Elhusseiny, 2011; Smeltzer & Keddy, 2010). In the recent social uprisings that defi ned the Arab Spring, social media acted as important mobilizing agents (Bellin, 2012). New media have a unique capacity to bypass some of the structural restrictions that limit traditional media in authoritarian societies. This feature of new media gives minority voices a relatively open forum to lodge dissent, even if anonymously. Furthermore, fi ndings from recent research on opinion expression suggest that the influence of fear of isolation is significantly diminished in computer-mediated communication, compared to face to face contexts (Ho & McLeod, 2008). Given that fear of isolation may reflect cultural orientation, new media should have a greater impact on expression collectivistic, traditional societies than in individualistic, self-expressive societies. Future research might seek to determine the extent of this effect. Second, although cultural orientation is an important aspect of public opinion, it does not address the individual-level factors that play out in public endorsement of opinion factions and willingness to speak out. Scheufele and Moy (2000) suggest that public opinion researchers should measure the idiocentric and allocentric values that are the psychological manifestations of individualism and collectivism, respectively (see Triandis, 2001, for further explication of these concepts). Huang (2005) examined the effect of idiocentrism and allocentrism in the spiral of silence in Taiwan and the United States and found that neither personality dimension significantly predicted willingness to speak out in either cultural context. However, the measures of idiocentrism and allocentrism had low reliabilities, which affects the interpretation of the results. This is the only study of which we are aware that examines these personality variables in the context of the spiral of silence. Future research can shed additional light on the role of individual-level cultural variables that affect people’s willingness to speak out. Finally, although researchers have identified differences in fear of isolation, they have not considered how these differences affect willingness to speak out. The intuitive linkage among culture, fear of isolation, and willingness to speak out is that people in more collectivistic cultures experience greater fear of isolation and are thus less willing to speak out. Contrary to this assertion, Willnat, Lee, and Detenber (2002) found that in Singapore, fear of social isolation positively predicted willingness to speak out on the issue of interracial marriage and fear of opinion isolation positively predicted willingness to speak out on the issue of gay rights. Only one fi nding supported the prior assertion: fear of opinion isolation negatively predicted willingness to speak out on the issue of interracial marriage. Neuwirth, Frederick, and Mayo (2007) conducted a similar study of public opinion in the United States, fi nding that fear of isolation significantly predicted three unique strategies people use in order to avoid stating their opinion about the invasion of Iraq: lying, making non-revealing neutral comments, and saying nothing. That fear of isolation significantly affected

198

Sonny Rosenthal and Benjamin Hill Detenber

willingness to speak out in a predominantly individualistic society further complicates the role of culture in the spiral of silence. However, Willnat et al. (2002) suggest that the relationship between fear of isolation and willingness to speak out depends somewhat on the issue at stake. Research has yet to compare across cultures the effect of fear of isolation on willingness to speak out about a common issue. This gap in the literature presents a clear opportunity for future research. References Adamczyk, A., & Pitt, C. (2009). Shaping attitudes about homosexuality: The role of religion and cultural context. Social Science Research, 38(2), 338–351. doi:10.1016/j.ssresearch.2009.01.002 Asch, S. E. (1956). Studies of independence and conformity. I. A minority of one against a unanimous majority. Psychological Monographs, 70(9), 1–70. Attia, A. M., Aziz, N., Friedman, B., & Elhusseiny, M. F. (2011). Commentary: The impact of social networking tools on political change in Egypt’s “Revolution 2.0.” Electronic Commerce Research and Applications, 10(4), 369–374. doi:10.1016/j.elerap.2011.05.003 Balkin, J. M. (2004). Digital speech and democratic culture: A theory of freedom of expression for the information society. [Editorial Material]. New York University Law Review, 79(1), 1–58. Barth, J., & Parry, J. (2009). Political cult ure, public opinion, and policy (non)diff usion: The case of gay- and lesbian-related issues in Arkansas. Social Science Quarterly, 90(2), 309–325. doi:10.1111/j.1540-6237.2009.00619.x Bellin, E. (2012). Reconsidering the robustnes s of authoritarianism in the Middle East lessons from the Arab Spring. Comparative Politics, 44(2), 127‒149. Blekesaune, M., & Quadagno, J. (2003). Public attitudes toward welfare state policies: A comparative analysis of 24 nations. European Sociological Review, 19(5), 415–427. doi:10.1093/ esr/19.5.415 Federici, S., Stella, A., Dennis, J., & Hunefeldt, T. (2011). West vs. West like East vs. West? A comparison between Italian and US American context sensitivity and Fear of Isolation. Cognitive Processing, 12(2), 203–208. doi:10.1007/s10339-010-0374-8 Fuligni, A. J., Tseng, V., & Lam, M. (1999). Attitudes toward family obligations among American adolescents with Asian, Latin American, and European backgrounds. Child Development, 70(4), 1030–1044. doi:10.1111/1467-8624.00075 Ho, S. S., & McLeod, D. M. (2008). Social-psychological influences on opinion expression in faceto-face and computer-mediated communication. Communication Research, 35(2), 190–207. Hofstede, G. (1980). Culture’s consequences. Beverly Hills, CA: Sage. Hornsey, M. J., Jetten, J., McAuliffe, B. J., & Hogg, M. A. (2006). The impact of individualist and collectivist group norms on evaluations of dissenting group members. Journal of Experimental Social Psychology, 42(1), 57–68. doi:10.1016/j.jesp.2005.01.006 Huang, H. P. (2005). A cross-cultural test of the spi ral of silence. International Journal of Public Opinion Research, 17(3), 324–345. doi:10.1093/ijpor/edho65 Inglehart, R., & Baker, W. E. (2000). Modernization, cultural change, and the persistence of traditional values. American Sociological Review, 65(1), 19–51. doi:10.2307/2657288 Jetten, J., Postmes, T., & McAuliffe, B. J. (2002). ‘We’re all individuals’: Group norms of individualism and collectivism, levels of identification and identity threat. European Journal of Social Psychology, 32(2), 189–207. doi:10.1002/ejsp.65 Kahan, D. M., Braman, D., Gastil, J., Slovic, P., & Mertz, C. K. (2007). Culture and identityprotective cognition: Explaining the White-male effect in risk perception. Journal of Empirical Legal Studies, 4(3), 465–505. Kenny, T., & Gross, P. (2008). Journalism in Central Asia: A victim of politics, econom-

Cultural Orientation and the Spiral of Silence

199

ics, and widespread self-censorship. International Journal of Press-Politics, 13(4), 515–525. doi:10.1177/1940161208324644 Kim, K., & Markman, A. B. (2006). Differences in fear of isolation as an explanation of cultural differences: Evidence from memory and reasoning. Journal of Experimental Social Psychology, 42(3), 350–364. doi:10.1016/j.jesp.2005.06.005 Kluckhohn, C. (1954). Culture and behavior. In G. Lindzey ( Ed.), Handbook of social psychology (pp. 921–976). Cambridge, MA: Addison-Wesley. Kuran, T. (1987). Preference falsification, policy continuity and collective conservatism. Economic Journal, 97(387), 642–665. doi:10.2307/2232928 Kuran, T. (1995). Private truths, public lies: The social consequences of preference falsifi cation. Cambridge, MA: Harvard University Press. Kwak, K. (2003). Adolescents and their parents: A review of i ntergenerational family relations for immigrant and non-immigrant families. Human Development, 46(2–3), 115–136. doi:10.1159/000068581 Kwon, K. H., & Moon, S. I. (2009). The bad guy is one of us: f raming comparison between the US and Korean newspapers and blogs about the Virginia Tech shooting. Asian Journal of Communication, 19(3), 270–288. doi:10.1080/01292980903038998 Lee, F. L. F. (2006). Collective efficacy, support for democrat ization, and political participation in Hong Kong. International Journal of Public Opinion Research, 18(3), 297–317. doi:10.1093/ ijpor/edh105 Lee, T. (2002). The politics of civil society in Singapore. Asian Studies Review, 26(1), 97–117. doi:10.1080/10357820208713332 Lee, T., & Kan, C. (2009). Blogospheric pressures in Singapore: Internet discourses and the 2006 general election. Continuum—Journal of Media & Cultural Studies, 23(6), 871–886. doi:10.1080/10304310903294804 Lee, Y. J., & Kim, S. (2010). Media framing in corporate social responsibility: A Korea-U.S. comparative study. International Journal of Communication, 4, 283–301. Leong, C. H. (2008). A multilevel research framework for the analyses of attitudes toward immigrants. International Journal of Intercultural Relations, 32(2), 115–129. doi:10.1016/j. ijintrel.2007.10.002 Liu, C. (2009). Narcissism and mob behaviour: Two extremes among Chi nese Internet users. Hermes, 55, 47–54. Madland, D. (2010). Reforming retirement: Values and self-interest dr ive support for policy reform in opposite directions. Journal of Aging & Social Policy, 22(2), 207–221. doi:10.1080/08959421003621804 Manaev, O., Manayeva, N., & Yuran, D. (2010). The ‘spiral of silence’ in election campaigns in a post-Communist society: The case of Belarus. International Journal of Market Research, 52(3), 319–338. doi 10.2501/s1470785309201296 McCright, A. M., & Dunlap, R. E. (2011). Cool dudes: The denial of clim ate change among conservative white males in the United States. Global Environmental Change-Human and Policy Dimensions, 21(4), 1163–1172. doi:10.1016/j.gloenvcha.2011.06.003 Mojzisch, A., Schulz-Hardt, S., Kerschreiter, R., & Frey, D. (2008). Combined effects of knowledge about others’ opinions and anticipation of group discussion on confi rmatory information search. Small Group Research, 39(2), 203–223. doi:10.1177/1046496408315983 Neuwirth, K., Frederick, E., & Mayo, C. (2007). The spiral of silence and fear of isolation. Journal of Communication, 57(3), 450–468. doi:10.1111/j.1460-2466.2007.00352.x Noelle-Neumann, E. (1984). The spiral of silence: Public opinion — Our soc ial skin. Chicago, IL: University of Chicago Press. Nolan, J. M., Schultz, P. W., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2008). Normative social influence is underdetected. Personality and Social Psychology Bulletin, 34(7), 913– 923. doi:10.1177/0146167208316691 Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collec-

200 Sonny Rosenthal and Benjamin Hill Detenber

tivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128(1), 3–72. doi:10.1037//0033-2909.128.1.3 Price, V. (1989). Social identification and public opinion: Effects of commun icating group confl ict. Public Opinion Quarterly, 53(2), 197–224. doi:10.1086/269503 Reporters without Borders. (2012). World press freedom index 2011–2012. Paris: Author. Scheufele, D. A., & Moy, P. (2000). Twenty-five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research, 12(1), 3–28. doi:10.1093/ijpor/12.1.3 Smeltzer, S., & Keddy, D. (2010). Won’t you be my (political) friend? The changi ng face(book) of socio-political contestation in Malaysia. Canadian Journal of Development Studies-Revue Canadienne D Etudes Du Developpement, 30(3-4), 421–440. Stack, S., Adamczyk, A., & Cao, L. Q. (2010). Survivalism and public opinion on c riminality: A cross-national analysis of prostitution. Social Forces, 88(4), 1703–1726. Stack, S., & Kposowa, A. J. (2011). The effect of survivalism-self-expressionism c ulture on black male suicide acceptability: A cross-national analysis. Social Science & Medicine, 72(7), 1211– 1218. doi:10.1016/j.socscimed.2011.01.028 Stockmann, D., & Gallagher, M. E. (2011). Remote control: How the media sustain authoritarian rule in China. Comparative Political Studies, 44(4), 436–467. doi:10.1177/0010414010394773 Strickland, A. A., Taber, C. S., & Lodge, M. (2011). Motivated reasoning and public opinion. Journal of Health Politics Policy and Law, 36(6), 935–944. doi:10.1215/03616878-1460524 Tey, T. H. (2008). Confi ning the freedom of the press in Singapore: A “pragmatic” pre ss for “nation-building”? Human Rights Quarterly, 30(4), 876–905. Triandis, H. C. (1989). The self and social-behavior in differing cultural contexts. Psychological Review, 96(3), 506–520. doi:10.1037//0033-295x.96.3.506 Triandis, H. C. (1994). Culture and social behavior. New York, NY: McGraw-Hill. Triandis, H. C. (2001). Individualism-collectivism and personality. Journal of Personality, 69(6), 907–924. doi:10.1111/1467-6494.696169 Van de Vliert, E. (2011). Bullying the media: Cultural and climato-economic readings of press repression versus press freedom. Applied Psychology—an International Review—Psychologie Appliquee—Revue Internationale, 60(3), 354–376. doi:10.1111/j.1464-0597.2010.00439.x Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33(4), 448–457. doi:10.1037/h0027806 Willnat, L., Lee, W. P., & Detenber, B. H. (2002). Individual-level predictors of public outspokenness: A test of the spiral of silence theory in Singapore. International Journal of Public Opinion Research, 14(4), 391–412. doi:10.1093/ijpor/14.4.391 Xu, T. B. (2009). The emergence of parallel opinions. Hermes, 55, 80–82. Zou, X., Tam, K. P., Morris, M. W., Lee, S. L., Lau, I. Y. M., & Chiu, C. Y. (2009). Culture as common sense: Perceived consensus versus personal beliefs as mechanisms of cultural influence. Journal of Personality and Social Psychology, 97(4), 579–597. doi:10.1037/a0016399

18 THE KUUKI THEORY AND THE SPIRAL OF SILENCE Youichi Ito

The great earthquake that devastated northeastern Japan in March 2011 and the Fukushima Atomic Power Plant disaster have served to reacquaint the Japanese with the word kuuki and what that entails. In addition, the Fukushima disaster has reminded Europeans of the Chernobyl disaster of 1986, which caused 33 direct casualties and uncountable indirect casualties and victims of radioactive contamination. In fact, news articles from Europe with titles such as “Antinuclear Kuuki Covers Europe” have made a strong impression on the Japanese mass media. In Japan after the earthquake, people felt as if they had been thrown into a wartime situation due to the frequent blackouts and shortages of goods in supermarkets and convenience stores. In fact, there was a report of an ordinary pedestrian chastising the manager at a supermarket over his bright neon signs that, he felt, were wasting electricity. A Japanese weekly magazine introduced numerous news snippets of this kind in a column entitled “What in the World is this Kuuki?” (Higaki, 2011). Shincho 45, a monthly magazine for intellectuals, featured articles on these phenomena under the title of “New Studies of Kuuki.” In one issue, this magazine emphasized, “the Japanese [vulnerability to “kuuki”] hasn’t changed a bit since Shichihei Yamamoto pointed it out more than 30 years ago” (Yoshizaki, 2011, p. 23), referencing social critic Yamamoto’s famous book A Study of Kuuki (1977). Yamamoto (1977), and almost all the books and articles written on kuuki phenomena, argued that kuuki is peculiar to Japan. In contrast, the author of this chapter has argued that it is more universal and can generally be applied to other nations besides the Japanese and gives examples such as the Dreyfus affair in late19th century France, the Spanish American War of 1898, Nazism in Germany, McCarthyism, and the response in the United States to the attacks on September

202 Youichi Ito

11, 2001 attacks. This point came to be gradually accepted by Western intellectuals, and the word kuuki was included in the fifth edition of the Dictionary of Media & Communication Studies (Watson & Hill, 2000) as an independent entry. Furthermore, Japanese kuuki theory was included in the Encyclopedia of Communication Theory (Littlejohn & Foss, 2009). Elilsabeth Noelle-Neumann was one of the Western scholars who paid attention to the concept of kuuki at a very early stage. She asked me many questions on the concept and wrote in her preface for the Japanese edition of The Spiral of Silence that kuuki and climate of opinion are very similar and that, for this reason, the spiral of silence theory may be easy for the Japanese to understand and accept (Noelle-Neumann, 1997, vii). It is true that kuuki and climate of opinion resemble each other on the macro level. However, kuuki and fear of isolation are quite different on the micro level. Let us compare such similarities and difference in the following sections. What Is Kuukii? Some English equivalents of kuuki are air, atmosphere, climate, environment, and circumstance. Although these words originally referred to physical phenomena, they, however, can also be applied to human or social phenomena. For example, one might say, “She has a sad air about her.” An important difference between kuuki and these English terms is that while the former has a connotation of pressure, the latter does not. Another important characteristic of kuuki is that it exists and functions just as much between two persons as it does among small groups and upward to national or international levels. Kuuki between two persons is called “relational kuuki” and kuuki among three people or more is called “situational kuuki” (Reizei, 2006). Relational Kuuki Relational kuuki can be defined as the atmospheric pressure that arises between two people. When two persons are communicating, gestures are used in addition to language. Gestures have been well studied by experts of interpersonal communication. In addition to language and gestures, however, humans communicate through observation and inference. For example, if we meet a woman “with a sad air about her,” we receive the message that she is sad. Then, what we can say or do to her is limited, which is a kind of pressure or kuuki. If people around her behave as if they do not notice that she is sad, they might be accused (at least in Japan) of being “K.Y.” (kuuki ga yomenai) or insensitive and inconsiderate. In Western communication culture, however, if somebody feels sad, he or she must express it in some way or another (Baldwin, 2008; Miyahara, 1992, 1993; Zhang, 2008). Ambiguous attitudes such as sadness may well be ignored. This cultural difference seems obvious because there are many Japanese words that refer to non-verbal behaviors assuming that they should be inferred and

The Kuuki Theory and the Spiral of Silence

203

noticed by the target person but there are no English or European equivalents (Ito, 1989, 1992). For example, Japanese psychiatrist Takeo Doi wrote about a British woman who was married to a Japanese man in his highly acclaimed book Anatomy of Dependence (1973). The women went to see Dr. Doi to consult about the mental problem of her daughter. The woman told Dr. Doi that her daughter had hardly displayed amae to her since she was a baby. To Dr. Doi’s question why she used the Japanese term amae in her explanation made in English, the British woman said, “There is no way of expressing it in English” (p. 18). Amae refers to non-verbal behaviors or signs meaning, “I need your attention, love, and care.” The Western mainstream norm may be: “if you feel in such a way, express it verbally or clearly” (Doi, 1974). However, this is only an official norm (tatemae) in the West. If one reads Erving Goff man’s The Presentation of Self in Everyday Life, especially its sixth chapter entitled “The Arts of Impression Management” (1959, pp. 208–237), one will realize that Westerners also use various non-verbal tactics to attract the attention of, or exert influence on, the target person. Interestingly, most users of these communication tactics in Goff man’s examples are women. In fact, sex, seniority, power, and social position have to do with the relational kuuki. People in a superior position can say whatever they want and need not pay attention to their inferiors’ unexpressed thoughts and feelings. On the other hand, those in an inferior position cannot say what they want and have to pay attention to their superiors’ unexpressed thoughts and feelings. Just as some people are good at making a speech, some people are good at nonverbal communication tactics, which means that relational kuuki may be manipulated to their own advantage by people with such a talent. Situational Kuuki Situational kuuki can be defi ned as the atmospheric pressure surrounding a situation. The fi rst thing that a person does in situational kuuki with three people or more is to try to understand by inference the nature, purpose, and reason for the gathering. This is what Thomas (1951) calls the “definition of situation.” In doing so, the person infers what is expected of him or her, which becomes the situational kuuki. For example, a person who cracks jokes in a situation where a woman is telling condoling friends about her miscarriage is an extreme example of “K.Y.” (kuuki ga yomenai) or insensitivity. People not only defi ne the situation but also take it for granted that the defi nitions of other people in the same situation are the same as theirs. Kuuki in small group situations often overlaps with ordinary group pressure. However, there are some differences. The fi rst difference is that in most cases group pressure is based on group norms that are static or rigid, whereas kuuki is based on a situation that is dynamic, changeable, and fluid. The second difference is that group pressure is something that is clearly stated or expressed

204 Youichi Ito

whereas kuuki is something inferred. The third difference has to do with the fi rst and second differences. While group pressure is rather difficult to change or control, kuuki can be easily manipulated, changed, and controlled by people (especially leaders) with a talent for it. In the case of relational kuuki, the sanction for not behaving properly is being disliked by the other person, acquiring a bad reputation as “K.Y.” (insensitive or inconsiderate), and so on. In the small group situation, the sanction for not complying with the majority may be overt and verbal criticism, bullying, or condemnation. Usually, however, the sanction is more subtle and takes the form of unpleasant looks, unfriendly attitudes, and an intentional and thorough disregard by the majority.1 Situational kuuki can arise in a community, a town, or a city. For example, Crenson (1971) discusses what he calls an “inarticulate ideology” operating in Gary, Indiana, home of the U.S. Steel Company, where criticism of the company was taboo because people depended on the company for employment and many other fringe benefits. The Japanese would call this “kuuki in Gary” because opinions are controlled by factors particular to the situation itself. At the national and international levels, individuals cannot directly experience most situations. At these levels, the mass media recognizes, experiences, understands, and tries to defi ne each situation on behalf of the individual (Boorstin, 1961; Fujitake, 1968; Lippmann, 1922). In history this was made possible through the development of newspapers. In L’Opinion et la Foule [Public Opinion and the Crowd] published in 1901, Gabriel Tarde points out that by writing about ongoing and urgent issues, newspapers, unlike books, caused strained emotions among their readers and often brought about mass frenzy (1964, p. 88). This “mass frenzy” is a kind of kuuki because it functions as strong atmosphere or a climate of opinion forcing compliance or conformity. The influence of yellow journalism, which substitutes sensationalism for well-researched facts, on the Spanish-American War (1898) is a typical example of newspapers creating kuuki on a national level and exerted strong pressure on both the public, and political leaders. According to Kaplan (2008): Starting in 1895, Cuban exiles in New York City provided the press with exaggerated or fabricated reports, detailing the attacks of Cuban rebels and reprisals of the Spanish government. The yellow journals seized on the information and twisted it to fit their own jingoistic plots. (pp. 5370–5371)

Then, the American battleship Maine mysteriously exploded in Havana harbor. As a result, President McKinley was under tremendous pressure from public opinion to embark on a war to liberate the Cubans from “Spanish tyranny and avenge the Maine.” The “war spirit” was whipped up across the entire nation “to the boiling point” (The Encyclopedia Americana, p. 360w). The “war

The Kuuki Theory and the Spiral of Silence

205

spirit” at the “boiling point” obviously indicates the existence of kuuki. Thus, on April 25, 1898, the American Congress declared the existence of a state of war as of April 21. Second, the defi nition of a situation can be given to the public in the form of propaganda, agitation, or by some other means of persuasive communication. During the Nazi period, a unique and strange kuuki gripped all of Germany. The Nazi defi nition of the situation itself was an illogical and mysterious mixture of several contradictory ideas. The pursuit of racial purity and expansionism were obvious contradictions, but such logical contradictions were masked by propaganda and agitation, that is, the manipulation of images and emotions. Thomson (1999) names Hitler, Goebbels, and Speer as the three outstanding propagandists during the Nazi period. Noelle-Neumann (1984, p. 112) refers to Goebbels’ “rallying call” technique. Moving spectacles as seen at the Nuremberg party rallies, which were prepared and staged by Albert Speer, certainly contributed to the creation and maintenance of a unique kuuki that supported the Nazi regime. Third, although it is rare, the defi nition of a situation can spontaneously arise from among the public in the form of a social movement. The anti-war kuuki (Ito, 2002) or climate of opinion (Monroe, 1975) in the United States at the time of the Vietnam War is an example. The anti-pollution movement in the 1970s (Yamamoto, 1977) is another. If some social movement successfully arises among the public and sways the majority of public opinion, it becomes a national consensus with a strong propensity for conformity, which is kuuki on the macro level. Regardless of the differences in the way kuuki is created, what sustains and reinforces it is some form of social sanction. On September 14, 2001, the U.S. Congress approved a resolution, requested by President Bush, to use the U.S. Armed Forced in response to the terrorist attacks on the country. In the Congress, only one member, Barbara Lee (Democrat), voted against the bill. According to Reuters, who distributed this news, since the night of that day, many faxes and telephone calls of protest and harassment began to reach Lee’s office and home. Consequently, the police offered her round-the-clock protection (“Hansen,” 2001). In Japan, at the time of the Russo-Japanese War (1904–05), The Kokumin Shimbun, a newspaper critical of the war, was attacked by angry mobs, set fi re to, and subsequently went bankrupt. The Yorozu Choho, another newspaper that was critical of the war, faced a drastic decrease in circulation as a result and changed its policy from anti-war to pro-war (Toriumi, 1973). In the 1930s, the Asahi Shimbun, a relatively liberal newspaper that was critical of the military, often became the target of boycotts by veterans’ and right-wing groups (Ogata, 1952, pp. 107–108; Maesaka, 1989, pp. 88–89). Terrorism can reinforce kuuki because it intimidates people into silence and enforces conformity.

206 Youichi Ito

Measuring Kuuki Is it possible to quantify and measure kuuki? The answer lies in the difference between the micro and macro levels. Most kuuki in interpersonal relations or in small group situations is recognized through inference, which is impossible to quantify. Therefore, interpretive methods, including discourse analysis or ethnomethodology, must be used to describe kuuki at the micro level. At the macro level, however, measurement is possible through (a) questionnaire surveys and (b) content analysis of the mass media. The procedure is as follows: 1. Defi ne the issue, including its beginning and end. 2. Measure the degrees of support/non-support or favorable/unfavorable in the following three sectors: (a) the government, (b) the mass media, and (c) the public. When there is a high degree of inter-media and inter-sector consonance regarding some specific opinion or policy, it is very likely that it exerts strong political or social pressure, especially when the following five conditions are met; then it is called kuuki and exerts a particularly strong influence: 1. The majority opinion is held in two or more of the following three sectors: government, mass media, and the public. 2. The majority opinion is held by the majority across the three sectors. 3. The majority opinion increases over time. 4. The majority opinion intensifies. 5. The subject matter tends to stir up the emotions and inflames peoples’ inherent emotions regarding basic values, societal norms, antagonisms, loyalty to the collective, and patriotism. 6. Social sanctions are imposed on those who do not conform to the majority point of view. When these conditions are met and kuuki is created, it exerts a strong political or social pressure, resulting in the minority becoming ever more silent and acquiescent and changing or modifying its opinion or policy. For example, the above criteria, named as the “tripolar kuuki model” (Ito, 1996a) explains why Premier Kaifu of Japan was unable to pass the United Nations Peace Cooperation Bill (to dispatch Japanese military to the Gulf War of 1991).2 Conclusion The kuuki and spiral of silence theories are similar in the sense that they are both applicable at the micro and macro levels, which makes it possible to explain the influence of public and news media opinion on government policies. This area has been rather weakly addressed in the American mass-communication

The Kuuki Theory and the Spiral of Silence

207

research tradition (Meyen, 2008, pp. 3321–3322). Dominant climate of opinion and kuuki both tend to silence people. Even in interpersonal relationships, certain kuuki created by one member of a relationship, which may be a result of something implied or inferred by one or the other, embarrasses the partner into silence. Although the spiral process is not a main feature of the kuuki theory, interactions among government, the mass media, and the public can be selfreinforcing spirals that aggravate any existing kuuki and leading to the “rule of kuuki” (Yamamoto, 1977). The difference between these two theories is that while the spiral of silence theory concentrates on fear of isolation, the kuuki theory addresses actual sanctions, nuisances, and inconveniences such as damaged reputation, being disliked, overt criticism, intentional and thorough disregard, boycott, and terrorism. The kuuki theory also addresses the role of (group or national) leaders who are able to create an atmosphere that applies pressure on group members or the public. These two theories are applicable to any society in the world at the interpersonal and small group levels. Some kuuki phenomena at the macro level sometimes appear in advanced democracies as in cases of the anti-nuclear, antipollution, and anti-war movements. However, as discussed earlier, kuuki at the macro level has often been created by irresponsible mass media, propaganda and agitation by the government, and reinforced by terrorism. These are characteristics of immature democracies rather than mature ones where freedom of speech is given a paramount position and every rational member of society despises political terrorism and supports severe punishment as a deterrent against it. In traditional Japan, motivation was often emphasized more than act itself. For this reason, punishment of terrorism based on patriotism, honor, loyalty, and other selfless samurai or chivalrous values was rather light. This traditional way of thinking completely changed after the World War II as far as political terrorism was concerned. In present-day Japan, political terrorism is considered to be the worst public enemy. In contrast, in many Islamic countries, terrorism based on religious motivation tends to be justified. This aggravates “rule by kuuki” (Yamamoto, 1977). There are many countries in the world today where democracy is still immature but strong economic and military capabilities are available. The macro theories of kuuki may be especially helpful in analyzing and explaining the political process in such countries. Notes 1. Intentional and thorough disregard is called shikato and is considered to be a serious form of bullying especially among primary school and junior high school children. 2. Also see Ito (1996b) in which the same method is used to explain the amendment process of the Consumption-Tax Law in 1989 and Ito (2002) that describes the measurement method in more detail.

208

Youichi Ito

References Baldwin, J. R. (2008). Communication modes, Western. In W. Donsbach (Ed.), The International encyclopedia of communication (pp. 785–789). London, England: Blackwell. Boorstin, D. J. (1961). The image. London, England: Weidenfeld & Nicolson. Crenson, M. A. (1971). The un-politics of air pollution. Baltimore, MD: Johns Hopkins University Press. Doi, T. (1973). The anatomy of dependence. Tokyo, Japan: Kodansha International. Doi, T. (1974). Some psychological themes in Japanese human relationships. In J. C. Condon & M. Saito (Eds.), Intercultural encounters with Japan: Communication-contact and conflict (pp. 17–26). Tokyo, Japan: Simul Press. The Encyclopedia Americana. (1964). “Spanish-American War,” vol. 25, 360v–360x. Glendale, CA: Encyclopedia Americana Corporation. Fujitake, A. (1968) Gendai komyunikeshon no riron [Contemporary mass communication theories]. Tokyo, Japan: Nihon Hoso Shuppan Kyokai. Goff man, E. (1959). The presentation of self in everyday life. New York, NY: Anchor. “Hansen no kain giin wo keigo” [Anti-war Congresswoman is protected]. (2001, September 18). Sankei Shimbun, p. 3. Higaki, T. (2011). Nannanda kono kuuki wa? [What in the world is this kuuki?], Shukan Gendai. Ito, Y. (1989). Socio-cultural backgrounds of Japanese interpersonal communication style. Civilisations, 39(1), 101–127. Ito, Y. (1992). Theories on interpersonal communication styles from a Japanese perspective: Sociological approach. In J. Blumler, J. McLeod, & K. E. Rosengren (Eds.), Comparatively speaking: Communication and culture across space and time (pp. 238–268). Newbury Park, CA: Sage. Ito, Y. (1996a). Mass media’s influence on government decision making. In D. L. Paletz (Ed.), Political communication in action: States, institutions, movements, audiences (pp. 37–52). Cresskill, NJ: Hampton. Ito, Y. (1996b). Masses and mass media influence on government decision-making. In D. L. Paletz (Ed.), Political communication research (pp. 63–89). Norwood, NJ: Ablex. Ito, Y. (2002). Climate of opinion, kuuki, and democracy. In W. Gudykunst (Ed.), Communication yearbook 26 (pp. 267–296). Mahwah, NJ: Erlbaum. Kaplan R. L. (2008). Yellow journalism. In W. Donsbach (Ed.), The international encyclopedia of communication (pp. 5369–5371). London, England: Blackwell. Littlejohn, S. W., & Foss, K. (Eds.). (2009). Encyclopedia of communication theory. Newbury Park, CA: Sage. Lippmann, W. (1922). Public opinion. New York, NY: Harcourt. Maesaka, T. (1989). Senso to shimbun, 1926–1935: Hei wa kyoki nari [War and the press, 1926–1935: A soldier is a weapon]. Tokyo, Japan: Shakai Shiso-sha. Meyen, M. (2008). Noelle-Neumann, Elisabeth. In W. Donsbach (Ed.), The international encyclopedia of communication (pp. 3320–3322). London, England: Blackwell. Miyahara, A. (1992). Toward a Japanese theory of interpersonal communication competence. Seinan Gakuin Daigaku Eigo Eibungaku Ronshu, 32(2-3), 25–46. Miyahara, A. (1993). An empirical investigation of communication strategies employed by Japanese as “collectivists.” Seinan Gakuin Daigaku Eigo Eibungaku Ronshu, 33(3), 83–110. Monroe, A. D. (1975). Public opinion in America. New York, NY: Dodd, Mead & Company. Noelle-Neumann E. (1984). The spiral of silence: Public opinion—our social skin (2nd ed.). Chicago, IL: University of Chicago Press. Noelle-Neumann, E. (1997). Chimmoku no rasen riron [The spiral of silence: Public opinion — Our social skin] (I. Kenichi & Y. Tomoko , trans.). Tokyo, Japan: Burehn Shuppan. Ogata, T. (1952, January). Genron hissoku jidai no kaiso [Memoir of the age of press supression]. Chuo Kohron, 67(1), 106‒111. Reizei, A. (2006). “Kankei no kuuki“ “ba no kuuki” [“Relational kuuki” “situational kuuki”]. Tokyo: Kohdan-sha.

The Kuuki Theory and the Spiral of Silence

209

Tarde, G. (1964). Yoron to gunshu ( Japanese translation of L’fopinion et la foule) [Public opinion and the crowd] ( I. Michio, trans.). Tokyo, Japan: Mirai-sha. Bungei Shunju-sha. (Original work published 1901) Thomas, W. I. (1951). Theory and social research. In E. H. Volkart (Ed.), Social behavior and personality (pp. 211–228). New York, NY: Social Science Research Council. Thomson, O. (1999). Easily led: A history of propaganda. Thrupp, Stroud, England: Sutton. Toriumi, Y. (1973). Taigai kiki ni okeru shimbun roncho [Newspaper editorials at national crises]. In N. Bunka Kaigi (Ed.), Nihon ni okeru jahnarizumu no tokushitsu [Characteristics of journalism in Japan] (pp. 4–23). Tokyo: Kenkyu-sha. Watson J., & Hill, A. (2000). Dictionary of media & communication studies. London, England: Arnold. Yamamoto, S. (1977). Kuuki no kenkyu [A study of kuuki]. Tokyo, Japan: Bungei Shunju-sha. Yoshizaki, T. (2011). Nihon wo oou “3.11 Syndrome” [The 3.11 Syndrome covers Japan]. Shincho 45, September, 20–23. Zhang, Y. B. (2008). Communication modes, Asia. In W. Donsbach (Ed.), The international encyclopedia of communication (pp. 775–779). London, England: Blackwell.

This page intentionally left blank

A Personal Tribute

This page intentionally left blank

19 ELISABETH NOELLE-NEUMANN AND THE MYSTERIES OF SOLITUDE Mihaly Csikszentmihalyi

For centuries, social scientists have puzzled over why we need each other so much. The Greeks adopted the word “idiot” to describe someone who lived by himself, outside the community. The Romans considered banishment from the city—Inter hominem esse desinere, or ceasing to be among men, next to being thrown off to death from the Tarpean Rocks by the Capitol Hill, the worst punishment the state could enforce. Aristotle defi ned humans as zoon politikon, which could be translated as “animals who live in cities.” They lived in cities because they needed each other’s company, and, in order to live in cities, they had to develop ethics (a word that derives from the Greek “the way things are done in our town”), various technologies, and a political system that allowed them to work together towards shared goals. In modern times, anthropologists have reported that in many pre-literate cultures when a person behaved in ways that the tribe disapproved of, the majority agreed to believe that the offender was dead, even though biologically he was hale and hearty. They appeared not to notice him (or her), they talked of him in the past tense, they shook their heads when telling stories of his bad deeds, and expressed relief at his demise. And of course, because he was dead already, no one offered him food or drink, or help when needed. Slowly— the anthropologists report—the culprit also began to believe that—contrary to appearances—he was dead, lost hope in the future, and shortly afterwards passed from social to biological death as well. But being accepted by the group is not just a human need, as ethologists have noticed. Our primate ancestors, who share 95% or more of human genetic material, have had to learn the hard way that expulsion from the community has usually deadly consequences. A baboon or a chimpanzee that is chased away from the group, soon becomes the prey of leopards or snakes. We know in our

214 Mihaly Csikszentmihalyi

very fibers that we depend on our well-being, and on our very identity, on the acceptance and approval of others like us. The importance of harnessing this human need did not go unnoticed to rulers and politicians of the past. Demagogues learned how to use the desire for acceptance and security that community offered. As recently as the last century, Stalin was able to get the support of Russians against the Ukrainians and Poles whom he wanted to destroy, by focusing on how different they were from the ethnic group in power; Hitler’s emphasis on the differences between Jews and Aryans was a masterful move to create an us-vs.-them mentality that allowed Christians to consider Jews as belonging to a not-quite-human race. In other words, our need for acceptance by the community has made us vulnerable to manipulation at the hand of those who need our help to achieve their political ends. In this sense, Aristotle’s “zoon politikon” could be translated as “animals that are easy to manipulate.” The Social Nature of Man in Noelle-Neumann’s Work Social thinkers from Machiavelli’s time to today have understood this weakness in the human constitution, and have tried to fi nd ways to teach those in power how to use this knowledge to pursue their own goals. As we all know, almost half a century ago, there were those who accused Elisabeth Noelle-Neumann of being one of the scientists who used their knowledge to help those interested in the pursuit of power. In the many years of my acquaintance with Elisabeth, however, I learned that this was not true. She was often stubborn, willful, and prejudiced in favor of the political forces she felt best represented her beliefs. But more than anything else, she was passionately interested in how people made decisions, in what they cared about, in how personal beliefs were translated into political actions. She was not motivated by power, but by an insatiable curiosity about the human condition. Occasionally I disagreed with how she interpreted the data she collected, but over the years I have come to trust implicitly the integrity of her motivation and of her data. What more could one expect, even in the supposedly objective world of science? Although Elisabeth worked with many concepts in her research and writing, none was as central to her thinking as the human need for inclusion. It is the basis of her main theoretical work—The Spiral of Silence—and it informs her uncanny ability to predict trends in the climate of opinion that were not visible to less theoretically attuned observers. This sense for what people were thinking, and wishing, was Elisabeth’s genius, and it was because of this that in the popular imagination she was seen as a Sybil, a seer; and by her detractors, as a witch. Yet there was no great mystery to her achievements—just a great deal of thought, theory, and hard work. The importance of social contact, of relationships, has been recognized by many psychologists in the past few decades.1

Elisabeth Noelle-Neumann and the Mysteries of Solitude

215

People who are isolated physically and psychologically develop more diseases and die earlier. In my own work, I have been astonished over and over by how much happier people report being when they are in the company of friends and family than when they are alone. Even seriously ill people seem to forget their disease when they are with other people, and start to feel the pain again when they are alone. But I do not know of any psychologist who took these facts as seriously as Elisabeth did, or who used this knowledge to explain human behavior as thoroughly as she did. Methodological Innovations The other side of her genius was the ability to adapt technological tools and strict methods to the collecting and processing of data. Her system of hiring part-time interviewers distributed across Germany, each located in a demographically identifiable neighborhood, allowed her to send out a telefax (we are talking of those Dark Ages when personal computers were few and far between) to 1,000 interviewers, requesting each to interview five, or ten random people walking about in the neighborhood. In a matter of hours, as the telefaxes came back to the Allensbach Institute, the staff entered the answers and sorted them by relevant characteristics: e.g., Catholic Bavarian from a middle-class town; or a Protestant dockworker from Hamburg. Within the day it would be possible to create a composite picture of what the entire country thought or felt. I remember having breakfast one morning with Elisabeth, Heinz MaierLeibnitz (her late husband), and my wife in the garden of their villa in Piazzogna, the delightful retreat in Italian Switzerland where the couple used to restore their peace of mind, away from the frenetic pace of their lives in Germany. In the middle of the breakfast, the phone rang inside the house, and Elisabeth rushed in to answer it. It took a long time for her to return to the rough granite table where we were eating, and she had an unusually thoughtful expression on her face. After a few sips of coffee, she told us what the call was about. The German Chancellor was on a state visit to a mid-Eastern country. He had called to ask Elisabeth what to do. Apparently the night before, a minister with a sensitive portfolio had been arrested by the police as he was leaving, drunk, an establishment frequented by gay men. The fact had been picked up by the news media. What should he do, the Chancellor asked—demand the minister to resign? Keep him, but demand he apologize to the public? Or treat the event as a minor infraction? Elisabeth told the Chancellor that she will have suggestions, backed by data, next morning at the latest. So, after fi nishing breakfast, she repaired to her small office behind the kitchen, drew up a short questionnaire, and telefaxed it to Allensbach with a request that it be sent out immediately to the usual panel of interviewers. To make the story short, by dinner time she was able to phone the Chancellor that except for Catholic

216

Mihaly Csikszentmihalyi

middle-class rural men between 35 and 75 years of age, respondents were not upset very much by the minister’s behavior; so extreme measures were not necessary. All of the hundreds of surveys that were used at the Allensbach Institute were fresh and original; it was difficult to resist completing them. And Elisabeth treated the “Christmas Tree” at the Institute, a display where the various drafts of surveys hung so that all the researchers could walk past them several times a day and think of changes and improvements, as a sacred totem of her office. One of the most ingenious methodological advances came as a result of her occasional visits to East Berlin. She noticed that as soon as she crossed the border, people walking on the street looked different: their eyes were downcast and half-closed, their posture hunched up and defensive—they rarely laughed or smiled. On these observations she based a very intriguing measure of personality strength that could be used as an unobtrusive indicator by interviewers. I can’t help wondering whether Elisabeth’s vision inspired the Stimmungsgasometer that flashes in the sky and indicates the mood of the Berlin citizens these nights.2 It is said that Menenius Agrippa, who was consul in 503 BC, prevented revolution in Rome by giving a speech to the leaders of the two confl icting parties, in which he compared the Republic to the human body—the belly and the limbs could not survive without the direction of the head, and the head could not survive without the contributions of the belly and the limbs. From then on, many thinkers have used the organismic analogy to discuss the State. Watching Elisabeth at work, one realized that this analogy is becoming a reality: thanks to emerging technologies, the time has come when information about what is happening inside and outside the “body politic” can be obtained and processed instantaneously. Her institute operated like an embryonic peripheral nervous system by providing the necessary information for the central nervous system of government institutions on which to base their decisions. The Researcher’s Curiosity As it is often the case, Elisabeth’s genius was based on early curiosity. She used to say that the most enjoyable times in her childhood had been those when she spent days alone, building an imaginary city on the floors of an empty room in her parents’ house. A shoe-box was the railway station, a big open book standing on its edges was the church or the school. There were trees, and streets, and along the streets there were homes made of match boxes or pencil cases. In each home there lived a family, and Elisabeth knew each member of every family—their likes and dislikes, their fears and hopes. Occasionally something happened: a train derailed, or there was an epidemic of measles. Then Elisabeth walked around her town, imagining how each of the inhabitants was affected

Elisabeth Noelle-Neumann and the Mysteries of Solitude

217

by the event—some children were not at all sad to stay home from school when they came down with measles. But there is a great distance between early curiosity and imagination, on the one hand, and mature genius, on the other. Elisabeth was able to transform her childhood interest into a profession that would help society know itself, and act in a manner that was conscious of its interests. In this she was helped by a whole-hearted belief in what is best in traditional German culture, coupled with an equally fi rm faith in Western democracy. Sometimes these allegiances resulted in paradoxes in her opinions that were difficult to reconcile, but I have never known her to act against her most fundamental belief, which was her obligation to hold up an accurate mirror to public opinion. It is a pity that, at least for the time being, Elisabeth’s contributions are not better known outside researchers in public opinion. Psychologists, sociologist, and political scientists could benefit greatly by her theoretical insights and methodological advances. This is in great part due, in my opinion, to the fact that the social sciences have become rather rigid and parochial despite their recent status as “science.” Because Elisabeth preferred to work largely outside the academic environment, following her own daimon and the unfolding of history, what she accomplished has been left out of the textbooks and journal references of the mainline social sciences. As with many other geniuses of the past, we can be sure that her ideas might live in the shadows for a time, but sooner or later will be recognized as a unique and necessary perspective for understanding the human condition. Notes 1. See Chapter 1, this volume, for influences on Noelle-Neumann and her original presentation of the theory of the spiral of silence in 1966 and later in her Spiral of Silence book. 2. The Stimmungsgasometer is a smiley on a huge screen from which one can read the average mood of Berlin citizens as measured by the expressions on random peoples’ faces that are then analyzed by a special software.

AUTHOR INDEX

A Abramowitz, A., 81 Abrams, S. J., 78, 81 Adamczyk, A., 188 Aday, S., 60– 61, 135 Allen, S., 112, 125 Allport, F. H., 154, 157 Althaus, S. L., 96 Asch, S. E., 192 Ashman, M. G., 79 Attia, A. M., 197 Aziz, N., 197 B Baker, W. E., 188 Baldwin, J. R., 202 Balkin, J. M., 196 Ball-Rokeach, S., 2, 39 Banks, C., 107 Bantz, C., 137 Bartels, L. M., 77 Barth, J., 189 Bathelt, A., 146 Beck, P. A., 81, 122 Bellin, E., 197 Benford, R. D., 138 Bennett, A., 94 Bennett, W. L., 41, 42, 77, 96, 137, 156 Berdayes, V., 69, 147 Berelson, B. R., 37, 79, 161, 162 Berger, P. L., 135 Bicchieri, C., 133

Bimber, B., 93 Blakely, E. J., 81 Blekesaune, M., 188 Blumer, J., 135 Bogart, L., 5 Bolt, D., 150 Bond, J., 81 Boorstin, D. J., 204 Borzekowski, D. L. B., 150 Bourdieu, P., 171, 172 Bowring, J., 24 boyd, d., 116 Brade, A. M., 13 Braman, D., 189 Brashears, M. E., 94 Breed, W., 137 Breen, M., 56 Brody, R., 138 Brooks, D. J., 96 Brosius H-B., 97, 146, 165, 166 Brossard, D., 137 Brown, R., 108 Brownstein, R., 81 Brunell, T., 82 Brunsting, S. Bryan, F. M., 176 Bryant, J., 131 Buller, D. B., 105 Burgoon, J. K., 105 Burson-Marsteller, 166 Burt, R., 162

Author Index

C Cao, L. Q., 188 Cappella, J. N., 95, 96, 108–109 Cappella, J., 138 Cartwright, D., 27 Chaffee, S. H., 11, 16, 37, 119, 120, 122, 123, 124, 166 Chan, K., 164 Chang, T., 137 Chen, C., 137 Chia, S. C. Y., 147, 150 Chotiner, A., 156 Christen, C. T., 147, 149 Christie, B., 104 Cialdini, R. B., 133, 134, 137, 192 Cintron, R., 173 Clement, R. W., 78 Cohen, B., 165 Cohen, J., 150 Cook, R., 79 Cooley, C.H., 6 Coon, H. M., 188 Costa, D., 82 Cova, B., 94 Cova, V., 94 Crenson, M. A., 204 Crespi, L., 29 Cromwell, 27 Croteau, D., 132 Croucher, S., 135 D Dalton, R. J., 81, 122 Daschmann, G., 154 Davenport, T. C., 80 Davison, W. P., 10, 39, 145, 148, 149, 150, 180 de Waal, E., 96 Debatin, B., 156 DeFleur, M., 2, 39 Degen, S., 97 DellaVigna, S., 81 Dennis, J., 193 Detenber, B., 9, 56, 60, 69 Detenber, B. H., 135, 197 Deutsch, M., 159 Diamond, E., 95 Diener, E., 107 Dillard, J. P., 150 Dilliplane, S., 80, 81, 87 Dimitrova, D., 102, 114–115 Doi, T., 203 Domke, D., 69

219

Domke, D. S., 94 Donsbach, W., 4, 13, 68, 97 Dovifat, E., 19, 27 Druckman, J. N., 79, 138 Dunlap, R. E., 189 Dunwoody, S., 137 Dylko, I. B., 60 E Ebo, B., 183, 184 Edwards, H. H., 137, 139 Edwards, K., 96 Eldersveld, S., 27 Elhusseiny, M. F., 197 Entman, R. M., 39 Esser, H., 102, 102, 116 Eveland, W. P., 6, 9, 56, 95 Eveland, W. P., Jr., 57, 121, 122, 123, 125, 126, 149, 156 F Faber, R. J., 149 Farrell, T. B., 172 Fazio, R. H., 133 Federici, S., 192 Fenwick, R., 14 Festinger, L., 96 Fields, J. M., 78, 154, 157 Fiorina, M.P., 78, 81 Fishkin, J. S., 169 Fishman, M., 77 Fiske, S., 137 Fiske-Rusciano, R., 181 Fleisher, R., 81 Foss, K., 202 Fraenkel, E., 27 Frederick, E., 55, 57, 69, 70, 135, 187, 197 Frey, D., 189 Friedman, B., 197 Friend, R., 192 Fujitake, A., 204 Fuligni, A. J., 188 Fung, K. F., 140, 141 G Gallagher, M. E., 194 Gamson, W., 40 Gamson, W. A., 132 Garrett, R. K., 97, 101, 156 Garry, S. L., 9 Gastil, J., 189 Gaudet, H., 37, 79, 161, 162 Geer, J. G., 96

220

Author Index

Gentzkow, M., 81 Gerard, H. B., 159 Gerber, A. S., 77, 80 Gerbner, G., 165 Gergen, K. J., 108 Ghanem, S. Ghosh, R., 166 Glanvill, J., 25 Gleason, L. S., 101 Glynn, C. J., 6, 9, 14, 55, 58, 59, 65, 66, 67, 68, 69, 70, 94, 120, 121, 123, 127, 156, 157, 158 Goff man, E., 137, 203 Golding, P., 156 Goldman, S. K., 78 Goldstein, N. J., 192 Gonzenbach, W. J., 12, 14, 58, 59, 68 Goodnight, G. T., 172 Graf, J., 60–61, 135 Green, D. P., 77, 80 Greenfeld, L., 183 Griskevicius, V., 192 Groeling, T., 77 Gross, L., 165 Gross, P., 189, 194 Guggenheim, L., 37, 41 Gunther, A. C., 147, 149, 150 Gusfield, J. R., 170 H Habermas, J., 6, 19, 21, 28, 169 Hackett, R., 79 Hahn, K. S., 70, 81 Hall, S., 156 Hamilton, H., 164 Han, M., 69, 150 Haney, C., 107 Hardin, C.D., 6 Harris-Lacewell, M., 173, 176 Hauser, G. A., 173 Hayes, A. F., 55, 58, 60, 61, 65, 69, 70, 120, 123, 126 Hechter, M., 135 Henderson, K., 166 Hennis, W., 27 Herberichs, G. Hersh, S. M., 135 Hertzog, C., 60 Hetherington, M. J., 81 Higaki, T., 201 Higgins, E. T., 6 Hill, A., 202 Hill, C. J., 181

Hindman, M., 93 Hively, M. H., 95 Hizi, S., 14 Ho, S. S., 56, 69, 97, 102, 114, 116, 197 Hockheimer, J. L., 37 Hofstede, G., 188 Höfstetter, C. R., 30, 79, 80 Hogg, M. A., 190 Holbert, R. L., 101 Holtzendorff, F. von, 23, 27 Horne, C., 134 Hornsey, M., 190 Howard, R. G., 174 Hoynes, W., 132 Hu, Y.-W., 97 Huang, H. P., 56, 61, 135, 190, 191, 197 Huckfeldt, R., 81, 119, 121, 122, 124 Huff aker, D. A., 166 Huh, J., 150 Hunefeldt, T., 192 Hutchens, M. J., 121, 122 I Ihering, R., 22-23, 24 Inglehart, R., 188 Ito, Y., 203, 205, 206, 207 Iyengar, S., 70, 81, 96, 136, 141, 156 J Jablonski, P., 14, 68 Jäckel, M., 116 Jamieson, K. H., 95, 96 Jetten, J., 190 Joubert, J. P. R., 163 K Kahan, D. M., 189 Kahn, M., 82 Kallgren, C. A., 134 Kan, C., 193, 194 Kaplan R. L., 204 Kaplan, E., 81 Katz, D., 27, 154, 157 Katz, E., 37, 161, 162, 164, 165 Keddy, D., 197 Kelley, H. H., 134 Kemmelmeier, M., 188 Kenny, T., 189, 194 Kenski, K., 79 Kepplinger, H. M., 5 Kerschreiter, R., 189 Key, V. O., 95 Kiesler, S., 105

Author Index

Kim, K., 192 Kim, S-H., 59, 69 Kinder, D. R., 141 King, C., 14, 68 Kiousis, S., 102 Klapper, J., 37 Kline, F. G., 10, 112, 156 Kluckhohn, C., 187 Kluckhohn, F. R., 138 Kposowa, A. J., 188 Krosnick, J. A., 81 Krueger, J., 78 Kruglanski, A., 96 Kunda, Z., 78 Kuran, T., 195 Kwak, K., 188 Kwon, K. H., 189 L Laclos, C. de Ladd, J. M., 77, 81, 93 Lam, M., 188 Landis, P. H., 22, 26 Landshut., S, 20 Lang, G. E., 8 Lang, K., 8 Lang, S., 93 Langteau, R., 150 LaPiere, R. T., 24 Lapinski, M. K., 133, 134 Larimer, C. W., 80 Lasorsa, D. L., 10, 57 Lauf, E., 3, 96 Lawrence, E., 77 Lawrence, R. G., 137 Lazarsfeld, P. F., 37, 79, 161, 162, 164 Lea, M., 108, 111 Lee, E., 56, 69, 122 Lee, F. L. F., 187 Lee, T., 193, 194 Lee, W. P., 69, 135, 197 Lee, W., 56 Lee, Y. J., 189 Lemert, J. B., 79 Lemin, D., 185 Lenz, G. S., 77, 81 Leong, C. H., 189 Lepper, M. R., 97, 148 Lerman, K., 166 Levendusky, M., 81 Levine, T. R., 134 Levy, M. R., 161, 162 Liebhart, J. L., 147, 150

221

Lin, N., 68, 164 Lippmann, W., 10 Littlejohn, S. W., 202 Liu, C., 196 Livingstone, S., 137 Locke, J., 21, 22, 23, 27 Lodge, M., 77, 189 Löffler, M., 20, 29 Luban, D., 171 Luckmann, T., 135 Luntz, F., 138 Lyons, B., 166 M Madland, D., 188 Maesaka, T., 205 Manaev, O., 187, 190, 194 Manayeva, N., 187 Mann, C. B., 80 Markman, A. B., 192 Marshall, G., 133 Martin, P., 77 Matera, F. R., 57, 68 Matthes, J., 55, 56, 60, 61, 69, 70 Maus, H., 14 Mayer-Uellner, R., 106 Mayo, C., 55, 69, 70, 135, 187, 197 McAuliffe, B. J., 190, 192 McClung Lee, A., 27 McCombs, M. E., 2, 39, 77, 141, 161 McCright, A. M., 189 McDevitt, M., 102, 104, 105, 107, 115 McDonald, D. G., 59, 157 McGuire, T. W., 105 McLeod, D. M., 7, 9, 56, 57, 102, 114, 116, 149, 197 McLeod, J., 15, 55, 135, 138 McLeod, J. M., 66, 68, 69, 102, 120, 125 McLuhan, M., 167 McPhee, W. N., 79 McPherson, M., 94 McQuail, D., 132 Mead, M., 134 Mendez, J. M., 79 Menzel, H., 162 Merton, R. K., 26, 30, 132, 155, 161 Mertz, C. K., 189 Metzger, M., 15, 166 Meyen, M., 206 Milkie, M. A., 150 Mills, C. W., 36 Min, S. J., 60 Miner, M., 5

222 Author Index

Miron, D., 131 Mitzman, B. N., 79 Miyahara, A., 202 Modigliani, A., 40 Moh, C. Y., 2, 94 Mojzisch, A., 189 Monroe, A. D., 205 Moon, S. I., 189 Morrison, K. R., 69, 70 Mowlana, H., 180 Moy, P., 56, 57, 58, 60, 122, 134, 135, 156, 159, 190, 197 Mutz, D. C., 6, 12, 68, 77, 78, 87, 96, 96, 97, 119, 121, 124, 126, 146, 147 N Nathanson, A. I., 149 Neill, S. A., 185 Nesselroade, J. R., 60 Neuman, W. R., 37, 41, 93 Neumann, E. P., 44–45 Neuwirth, K., 55, 59, 68, 69, 70, 135, 187, 197 Newcomb, T. M., 94, 120, 133 Ng, I., 137 Ng, S. H., 109 Niedermann, B., 179 Nir, L., 79, 108-109, 124 Noelle, E., 19, 44 Noelle-Neumann, E., 2, 3, 4, 5, 13, 38, 44, 45, 46, 47, 48, 49, 50, 51, 55, 56, 58, 61, 65, 66, 68, 69, 75, 76, 79, 82, 92, 93, 101, 102, 104, 105, 109, 11,2, 116, 121, 126, 132, 134, 135, 146, 147, 149, 155, 156, 157, 158, 159, 163, 166, 169, 174, 179, 181, 183, 184, 190, 202, 205 Nolan, J. M., 192 Norris, P., 85 O O’Gorman, H. J., 9, 154, 155, 157 Ogata, T., 205 Oncken, H., 20 Opp, K., 135 Osborn, T., 79 Oshagan, H., 9, 68, 123, 124 Ostman, R. E., 59 Oyserman, D., 188 P Page, B. I., 159 Pan, Z., 102, 125, 138 Panagopoulos, C., 80

Papacharissi, Z., 77, 95 Park, E., 94, 123 Park, J. E., 9 Park, S.-Y., 69, 101, 105, 114 Parry, J., 189 Passino, K. M., 156 Paxton, P., 94 Pepitone, A., 133 Perry, S. D., 12 Perse, E. M., 40 Peter, C., 97 Petersen, T., 13, 46, 47 Petrič, G., 56, 58 Pinter, A., 56, 58 Pitt, C., 188 Pitts, S. C., 60 Pollock, J. C., 181 Pope, J. C., 78 Postmes, T., 106, 108, 111, 192 Potter, W., 40, 132 Preacher, K. J., 60 Price, V., 77, 108–109, 112, 125, 136, 187 Priest, S. H., 69 Prior, M., 77, 92, 156 Pruijt, H, 95 Putnam, R. D., 82 Q Quadagno, J., 188 R Rainie, L., 93 Reese, S., 136, 137 Reeves, B., 96 Reichenbach, H., 14 Reicher, S. D., 106, 107, 108, 109, 110, 112 Reizei, A., 202 Reno, R., 134 Rentsch, M., 97 Rimal, R., 133, 134 Rittenberg, J., 98 Robards, B., 94 Robinson, J., 164 Robinson, M. J., 79 Rojas, H., 60, 149 Rokeach, M., 137, 138 Rokka, J., 94 Rosengren, K., 135 Rosenthal, R., 68, 70 Ross, E. A., 22, 23, 134 Ross, L., 97, 148 Rössler, P., 101, 164 Rousseau, J.-J., 29

Author Index

Rusciano, F. L., 181, 182, 183, 184 S Salmon, C. T., 9, 10, 14, 68, 112, 156 Salwen, M. B., 57, 68 Sasson, T., 132 Saunders, K. L., 81 Schemer, C., 69, 70 Schenk, M., 164 Scherer, H., 164 Scheufele, B., 102, 116 Scheufele, D. A., 14, 56, 57, 58, 60, 69, 77, 122, 123, 132, 134, 135, 136, 138, 153, 159, 161, 163, 190, 197 Schielicke, A.-M., 97 Schlegel, F., 23 Schmidtchen, G., 24 Schönbach, K., 96 Schultz, P. W., 192 Schulz-Hardt, S., 189 Schuman, H., 78, 154, 157 Scott, J., 133, 173 Seither, M. A., 79 Selig, J. P., 60 Shah, D. V., 149, 163 Shamir, J., 58, 68, 148, 155, 156, 158 Shamir, M., 68, 148, 155, 156, 158 Shanahan, J. E., 14, 56, 58, 69, 70, 122, 123, 147 Shapiro, J. M., 81 Shapiro, R. Y., 159 Shaw, D. L., 141, 161 Shekhar, M., 164 Shen, F., 60, 137, 139 Sherif, M., 133 Shikaki, K., 158 Shoemaker, P. J., 56, 136, 137 Short, J., 104 Shulman, H. C., 134 Siegel, J., 105 Signorielli, N., 39, 57 Simpson, C., 5 Sire, J. W., 137 Slater, M., 70, 126 Slovic, P., 189 Smeltzer, S., 197 Smith, E. E., 96 Smith-Lovin, L., 94 Sniderman, P. M., 138 Snow, D. A., 138 Snyder, M. G., 81 Soss, J., 12, 147 Spears, R., 106, 108, 109, 111

Speier, H., 27 Spencer, A., 135 Sprague, J., 81, 121, 124 Stack, S., 188 Stamm, K., 56, 69, 94 Stamper, M., 56 Stella, A., 192 Stevenson, R. L., 58, 59, 68, 97 Stockmann, D., 194 Storey, J. D., 150 Streeck, W., 13 Strickland, A. A., 189 Strodtbeck, F. L., 138 Stroud, N. J., 79, 81, 93, 96, 156 Sunstein, C. R., 77, 84 T Taber, C. S., 77, 189 Tajfel, H., 104 Tarde, G., 204 Taylor, D. G., 153, 158 Taylor, G. D., 10 Taylor, S., 137 Tein, J.-Y., 60 Tetlock, P. E., 138 Tewksbury, D., 96, 98, 132, 136, 161 Tey, T. H., 193 Theriault, S. M., 138 Thibaut, J. W., 134 Thomas, W. I., 203 Thomson, O., 205 Toriumi, Y., 205 Touraine, A., 170 Tracy, K., 176 Trepte, S., 164 Triandis, H. C., 187, 189, 190 Trianni, V., 156 Troldahl, V. C., 164 Trost, M. B., 133, 134 Tseng, V., 188 Tsfati, Y., 150, 156 Tuci, E., 156 Turner, J. C., 108 Tustin, D. H., 163 U Uldall, B., 58 V Valente, T., 163 Vallone, R. P., 97, 148 van Dam, R., 164 Van de Vliert, E., 187, 189, 190

223

224

Author Index

Van Gorp, B., 140 van Vuuren, D., 163

Wojcieszak, M., 77 Woodall, W. G., 105

W Wahl-Jorgensen, K., 102 Walker, J., 81 Walsh, K. C., 176 Wang, J., 137 Wang, M., 181 Wanta, W., 97, 102, 114–115 Watson J., 202 Watson, D., 192 Watt, S. E., 108 Weimann, G., 47, 162, 163, 165, 166 Wellman, B., 94 West, S. G., 60 Williams, E., 104 Willnat, L., 56, 57, 58, 60–61, 69, 135, 147, 197, 198

X Xu, T. B., 196 Y Yamamoto, S., 201, 205, 206 Yang, F., 14 Yoshizaki, T., 201 Young, L., 87, 96, 97 Yun, G. W., 69, 101, 105, 114 Yuran, D., 187 Z Zhang, Y. B., 202 Zimbardo, P. G., 107 Zipp, J. F., 14 Zou, X., 195

SUBJECT INDEX

Page numbers in italics refer to figures or tables. A A-B-X model, 120, 123 Acceptance, 1–2, 5, 187, 214 Accessibility. 141 Accountability, 87 Accountability to political system, 96 Accuracy, 9, 11, 78, 120, 121–124, 150, 154–155 Affect-based collectives Agenda setting (agenda-setting), 4, 39, 77, 141, 161, 165–166 Agreement, 30, 39, 54, 81, 87, 93, 119–124, 126, 135 Allensbach Institute, 1, 7, 20, 30n, 44–45, 46, 163, 215–216 Allensbacher Jahrbuch der Demoskopie, 45 Allensbacher WerbeträgerAnalyse, 44 Anonymity, 15, 58, 94, 101, 102, 105–106, 107, 108–109, 110, 112, 113, 114, 115–116, 117n Applicability, 8, 11–12, 13, 60, 136, 139 Arab Spring, 172, 185, 197 Articulation function, 92 Atmosphere, 202, 204, 207 Atmospheric pressure, 202, 203 Audiences, 4, 11, 13, 15, 38, 40, 42, 77–78, 81, 132, 137–138, 161–162, 166 Audiences, fragmentation, 92–93, 96–97, 98 Automated outlets, 164–165

B Balkanization, 94 Bias, 50, 77, 147–148, 151, 154, 155–156, 157–159, 195 Bias, media, 4, 85, 87, 97 C Campaign participation, 4 Capital, 82, 95, 163–164 Capital, community, 82, 95 Centralized, 156, 166 Centralized media map, 92–93 Classical political philosophers, 179–181 Climate of opinion, 7, 12, 25, 30, 48, 51, 55, 57–58, 66–67, 69–70, 76, 78–83, 87, 95–98, 101–102, 104, 114–116, 120, 122–123, 125, 145–151, 153, 155–157, 202, 204–205, 207, 214 Closure, need for, 96 Cognitive, 36, 96, 108–110, 134, 136, 138 Cognitive orientation, 4, 40, 155 Cohesion, 26, 28, 29, 93, 135 Collectives, affect-based, 36, 94–95 Collectivist cultures, 60–61, 93, 156, 159, 193 Collectivistic orientation, 187, 206 Community, 4, 5, 10, 27 Community capital, 82, 95, 163, 166, 169, 213–214 Community-level opinion, 95, 120, 122, 123, 147, 172–174, 179–180, 183–184, 204 Computer-mediated communication (CMC), 101–102, 105, 116n, 197

226

Subject Index

Congruency, 120, 121–122, 124, 126 Connectivity, social, 94 Conservative bias, 157, 158, 159 Conservative opinion, 95, 157, 158, 159 Conservative opinion media, 95, 157, 158, 159, 174 Considered public opinion, 47, 169–176 Consonance, 15, 48, 76–78, 92, 95, 156, 206 Consonant, 4, 15, 55, 56–57, 62, 66, 76–77, 102, 103, 117n Consonant information, 96, 70 Content analysis, 38, 48, 131, 137, 206 Content, uniformity of, 95 Control, 8, 38, 41–42, 51, 88, 96, 170–171, 175, 185, 194, 204 Control, social, 6, 19–30, 93, 115, 159 Controversy Coorientation, 119–127 Coorientation model, 11, 120, 124 Credibility, source, 97, 170 Criticisms, 5, 7, 9–10, 42, 49–50, 80, 162–163, 174–175, 180, 204, 207 Cross-cutting exposure, 78, 81–82, 86–87 Cross-cutting perspectives, 81 Cultivation, 4, 6, 39, 94, 146, 156, 161 Cultural orientation, 187–198 Culture, 3, 7, 11, 12–13, 58, 60–61, 131, 135–136, 140–141, 203, 213, 217 Cumulation, 48, 156 D Decentralization, 92 Decentralized media map, 92 Defi nition of situation, 20 Defi nitions of public, 6, 20–21, 206 Democracy, 4, 8, 20, 21, 25, 28, 30n, 76, 83, 121, 169, 207, 217 Democratic engagement, 93 Dependency theory, 2, 39 Designing questionnaires, 45, 54 Deviance, 94–95 Diachronic process, 157–158 Diff usion of Innovations, 2 Digital, 70, Digital media, 70, 94 Discussion, 1, 4, 8, 11, 20, 28, 51, 54, 55, 78–79, 93–96, 102, 115, 119, 121–127 Dissonance, 38, 103, 112,116 Dissonance, cognitive. 96 Dissonant, 7, 96, 111, 114–115 Diversity, 35, 40, 41, 82, 87, 135, 187, 190 Dual climate of opinion, 48, 51

E Echo chamber, 95 Ecosystem, 93 Effect of publicity, 16, 48, 169, 171–175 Effects of culture on opinion expression, 11, 12–13, 60–61, 136–139, 202–203 Effects of culture on opinion formation, 135–136, 139–141, 187–190, 213 Effects of culture on the spiral of silence, 13, 60–61, 131, 190–196, 198 Effects of culture on willingness to speak out, 197, 202–203 Effects of mass media, 8, 36–38, 57, 59, 66, 75, 92, 161–162, 165 Effects on interpersonal communication, 4, 38, 66, 119–122, 165–166, 201 Effects on mass media, 20, 42, 47–49, 77, 96–97 Effects, media, 4, 8, 11, 35–42 Egocentric network data, 123–125 Elite press, 182 Empowerment, 5, 95 Enclave communication, 86 Engagement, democratic, 93 Environment, 5, 6, 9, 11, 15–16, 23, 35, 40, 56, 70, 76–78, 80, 82–87, 92, 93, 96–98, 101–117, 125, 131, 134–35, 139, 141, 153, 155–157, 166–167, 202 Environment, scanning, 66, 93, 95–98 Expectations of sanctions, 104–112, 113, 114 Experiments, 6, 10, 45, 62, 147, 191, 195 Expert-curated information, 92–93 Exposure, selective, 8, 78, 87, 96–97, 147, 156 Expression of opinions, public, 12–13, 20, 39, 55–60, 66, 70, 75, 79, 93–95, 101– 117, 119–126, 138, 157, 158, 171–175, 187–196 F Face-to-face interviews, 46 Fear of isolation, 10, 12, 38, 66, 70, 93–97, 104, 109, 135, 150, 158, 169, 174, 175, 183, 188, 190–193, 197, 202, 207 Fourth Hague Convention, 180 Fragmentation, 95, 98 Fragmentation of audiences, 92–93, 96–97, 156 Fragmentation of society, 93 Frame and schema theory, 39 Framing, 11, 40, 131–141, 162 Function, articulation, 92 Future research, 196–198

Subject Index

G Gentle violence, 171, 175 Global spiral of silence, 179, 185 Goals, 14, 93, 135, 137, 213, 214 Government, 6, 20–21, 24–29, 44, 82, 175, 183–184, 193–194, 206–207, 216 Groups, 9, 20–21, 23, 24, 29, 57, 60, 66, 77, 94–95, 104,107–116, 124–125, 131, 135–141, 155, 162, 170, 202, 205 H Historical roots, 179–181 Homogeneity, 77, 81, 94–95, 112, 115, 119, 135, 189 Homogenous communities, 82 Homogenous networks, 86 Hostile opinion, 59, 62, 76, 81 Hostile media, 6, 78–80, 97, 148 I Identifi ability, 107, 112, 113, 114 Inclusion, 67, 86, 214 Indexing, 77 Individualist cultures, 11, 12–13, 60–61, 192, 195–196 Individualistic orientation, 11, 38, 46 Individual-level, 4, 7, 10, 12, 21–24, 26–28, 55–59, 62, 63, 66, 70, 80, 88, 93 Information, 12, 15, 37, 39, 40, 45, 47, 56–57, 66, 79–82, 87, 92–93, 96–98, 101, 120, 124, 132, 134, 137, 139, 146, 148, 154–159, 162, 164, 166–167, 169, 171, 173, 181, 189, 191, 195, 216 Informational diets, 161 Institut für Demoskopie Allensbach, 1, 20, 44 Integration, 1, 7, 10–11, 23, 24–26, 97, 120, 124 Integration, society, 22–23 Intellectual roots, 5–6 Interactive aspects, new media technologies, 41–43, 93, 94, 98, 102, 156, 166–167 Inter-media agenda setting, 77 International law, 179–180 Internet, 15, 42, 77, 93, 94, 101, 102, 104, 156, 166, 185, 196–197 Internet fragmentation, 95 Interpersonal networks, 76, 81, 83, 86, 88, 155 Interpretations, 181, 182, 194 Intolerance, 79, 81–82, 83, 85 Isolation, 15, 27, 51, 80, 83, 182–185 Isolation, fear of , 10, 12, 38, 55–56, 61–62,

227

66, 70, 78, 93–95, 97–98, 103, 104, 109, 117n, 120, 126, 135, 149, 159, 169, 174, 175, 179, 188, 190–193, 197–198, 202, 207 Issue opinion, 83, 84, 85, 88 J Jahrbuch der öffentlichen Meinung, 45 K Kuuki, 201–207; relational Kuuki, 202–204; situational Kuuki, 202, 203–205; tripolar Kuuki model, 206 L Landscape, media, 15, 16, 92, 95, 103, 193 Law of nations, 180 Law-of-small-numbers, 147 Laws of humanity, 180 Levels of publicity, 49, 170–174 Liberal bias, 158 Like-minded others, 75, 81, 82 M Macro publicity, 171–172 Magic-bullet model, 161 Majority opinion, 9, 10, 47, 51, 56, 57, 59, 92, 103, 105, 114, 116n, 158, 190–191, 193, 194–195, 206 Mass amateurization, 92–93 Mass communication, 4, 15, 16, 36–37, 42, 101, 131, 138, 162, 165, 166–167, 206 Mass media, 4, 8, 20, 35, 36–40 , 42, 47, 49, 56–57, 59, 66, 76, 86, 92, 95–97, 102, 103, 119, 132, 145–147, 155–157, 161–166, 201, 204, 206–207 Mass opinion, 76, 78, 85, 87–88, 136 Mass persuasion, 36 Mass society, 8, 36, 39, 47, 94 Meanings, 39–40, 173, 181 Media consonance, 76–78 Media consumption, 161 Media content, 4, 12, 97, 101, 132, 146–148, 149, 164 Media cultivation, 39 Media dependency theory, 2 Media ecosystem, 93 Media effects, 4, 8, 11, 35–42 Media environment, 56, 70, 76–78, 83–87, 96, 98, 102, 155–156, 166–167, 196–197 Media interpretations, 182 Media landscape, 15, 16, 92, 95, 103, 193 Media map, 92–93

228

Subject Index

Media networks, 181 Meta-analysis, 9, 12, 65–70 Methodological innovations, 215–216 Micro publicity, 172–174 Minimal effects, 36–37 Minority, 12, 104–105, 108, 126, 155, 197, 206 Minority opinion, 6, 47, 61, 65, 75, 81–83, 92, 94–95, 116n, 134–135, 190, 193–194 Minority positions, 55–58, 76, 86, 154 Misers, 96 Model of sociological explanation, 103, 116n Moderating effects of new media, 41–42, 93, 95, 156–157, 166–167, 185, 197 N Nation’s image, 182–184 National identity, 181, 183–184 Need for closure, 96 Negation model, 162 Negative world opinion, 184 Neo-tribalism, 94 Network data, 122, 123–126 Networks, 9, 76–77, 78, 79, 81, 83, 86, 88, 92, 94, 121, 122, 124, 137, 139, 155, 161, 165, 181 New media ecosystem, 93 Newspaper content analyses, 181–184 Niche news, 93 Normative and Informational sources, 154, 159 O Online environments, 93, 101–117 Operationalization, 16, 56, 57, 61, 62, 63, 66 Opinion change, 7, 38, 50–51, 80, 158 Opinion climate, 12–13, 15, 55–59, 62–63, 66, 67, 70, 75–76, 78, 79–83, 98, 104, 117n, 120, 122–123, 125, 134, 138, 146, 148–149, 158 Opinion leaders, 11, 47, 148, 161–167 Opinion leadership, 11, 161–167 Opinion of states, 181 Opinion suppression, 148 Organized, 37, 92, 94 Overlapping groups, 94 Overload, information, 97 P Panel survey, 9, 46, 59 Partisan media, 77, 78, 81, 87, 93, 95–96, 97 Partisan news, 80, 95, 156

Paternalism, 149 Perceived consensus of opinion, 195–196 Perceived influence, 145–147 Perceived media influence, 147–148 Perceived public opinion, 9, 12, 47, 57, 58–59, 65, 66, 67, 69–70, 78–82, 104–105, 114, 115, 117n, 125, 134, 148–149, 158 Perceptions, 4, 6, 7, 9–11, 21, 30, 38, 55–57, 59, 66, 76, 78–79, 83, 88, 101, 119–127, 133, 138, 145 Personal identity, 108–109, 110 Personal networks, 76, 78, 165 Personality strength, 163, 164, 216 Persuasive press inference, 147–148, 151, 156 Pluralistic ignorance, 6, 9, 11, 39, 153–159 Polarization, 79, 81–83, 156 Political communication, 1, 2, 40, 54, 76, 82, 87, 119–127, 161 Political discussion, 11, 28, 49, 79, 119–127 Political participation, 75–76, 79–80, 86, 120 Political polarization, 81–82 Political preferences, 80–81, 121 Political system, 82, 96, 213 Politics, 8, 13–14, 19, 36, 44, 48, 76, 82, 124, 154, 174 Powerful mass media, 4, 8, 38, 47–48, 132–133, 156, 161 Powerful media effects, 36, 37–38, 132–133 Priming, 141, 156–157, 162 Principles, 20, 51, 180, 193 Processes, 13, 15, 20, 50, 60–61, 63, 88, 96–98, 102, 103–104, 108, 115, 116, 121, 125, 126, 132, 137, 138, 140, 141, 146, 148, 151, 154–157, 180–181, 188, 189 Propaganda, 5, 36–37, 149, 172, 180, 205, 207 Public conscience, 180 Public expression of opinions, 56, 93 Public opinion expression, 57–58, 101–116, 122, 158–159 Public sentiment, 55, 147, 151 Publicity principle, 170–171, 174 Publicity, modes of, 169–175 Publicity1, 170–174 Publicity2, 171–175 Publicity3, 174–175 Publiciyt4, 175 Punishment for violations, 180 Q Quasi-statistical organ, 146, 148

Subject Index

Quasi-statistical sense, 6, 9, 15, 56–57, 66, 102, 109, 153, 183 Questionnaire construction, 45 Quota samples, 46 R Reality, social, 10, 14, 39, 40, 92, 121 Reference groups, 9, 57, 94, 112, 119, 124, 125–126 Reliability, 12, 56–58, 62 Resonant opinion climates, 98 S Safe discussions, 95 Sanctions, 23–24, 95, 103, 104–106, 111, 113, 114, 115, 117n, 134, 149, 182–183, 206, 207 Scanning, 93, 95–98 Schemas, 11, 111, 131, 136, 139 Social Categorization Theory (SCT), 108 Segmentation, selective exposure, 8, 78, 87, 96–98, 147–148, 156 Self-censorship, 148, 149, 193, 194 Self-protective mechanism, 149 Social Identity Model of Deindividuation (SIDE), 105–117 Silencing, 78, 83, 88, 183, 184 Singular, 96, 97, 185, 194 Social Identity Theory (SIT), 108, 109 Slant of media coverage, 147 Snowball sample data, 124, 126 Social capital, 82, 95, 163–164 Social conformity, 58–59 Social constructivist, 40 Social control, 6, 19–30, 47, 93, 115, 159 Social identity model of deindividuation (SIDE model), 105–116 Social identity theory, 108 Social isolation, 12, 15, 27, 51, 55–56, 61–62, 78, 80, 83, 104, 120, 126, 149, 191–192, 197 Social media, 42, 92, 104, 116, 185, 196–197 Social nature, 6, 103, 107, 154, 214–215 Social network, 9, 13, 79, 161, 196–197 Social network analysis, 119–117 Social network data, 122, 124–125, 126 Social perspective, 153–155 Social pressure, 51, 78, 80, 88, 172–173, 190, 206 Social reinforcement, 85–86 Social sanctions, 23–24, 134, 149, 205, 206 Social-categorization theory (SCT), 108

229

Societal level opinion, 122 Society, 8, 9, 14, 21–24, 26, 28, 30n, 35–36, 38–39, 51, 66, 92–98, 101, 102, 103, 108–109, 133–134, 135, 149, 150, 155–156, 157, 175, 188, 190–191, 193, 195–196, 198, 207, 217 Source credibility, 97 Specialization, 29, 98 Spiral of silence theory, 1–16, 35, 39, 44–51, 54, 55, 57, 58, 62–63, 69, 75, 83, 92, 94, 96, 97, 101, 119, 131–132, 134–136, 139, 148, 149, 153, 155, 156, 158–159, 165, 171,175, 202, 207 Split-ballots, 46 Strength of personality, 163 Strong media, 46, 161–167 Student movement, 49 Survey, 1, 9, 14, 20–21, 25, 29, 39, 59, 115, 146, 163, 165, 169, 206, 216 Survey research, 44–51, 122–124, 131, 193 Swarm intelligence, 156 System, political, 82, 96, 213 T Technological advancements, 94, 98 Terrain, 35, 92 Terrain, media, 93 Third Reich, 1, 49 Third-person effect, 145–151 Threat of isolation, 183–184 Traditional. 2, 19, 94, 96, 101, 107, 166, 188–189, 190, 195–197, 207, 217 Train-test, 58 Trust, 37, 82, 96–97, 188, 214 Two-step flow, 11, 162, 165–166 U Uniformity of content, 95 V Validity, 4, 10, 12, 45, 56, 58, 62, 122 Values, 15, 47, 93, 133–139, 153, 155, 165, 173, 180, 182, 187–190, 195, 197, 206 Vernacular rhetoric, 169–175 Visibility, 50, 155, 158 Voices, 41, 77, 170, 171, 174–175, 197 Voices, dissonant, 96, 97 Voting, 7, 36, 49, 63, 66, 80, 150, 154 Vulnerability, 173–174, 176n, 201 W Weak media, 162, 167

230 Subject Index

Whole network data, 124–125 Willingness to express opinion/willingness to speak out, 12, 58–59, 63, 65, 80, 102, 103, 105, 119–120, 122, 135, 141, 188, 190–191, 193, 196–198

Wilsonian idealism, 180 World opinion, 179–185 Y Yellow journalism, 204