UCLA
UCLA Electronic Theses and Dissertations
Title
First, the Bad News: Opposition Media in Authoritarian Regimes
Permalink
https://escholarship.org/uc/item/488535rm
Author
McGreevy-Stafford, Eoghan Wallace
Publication Date
2020
Peer reviewed|Thesis/dissertation
eScholarship.org
Powered by the California Digital Library
University of California
UNIVERSITY OF CALIFORNIA
Los Angeles
First, the Bad News:
Opposition Media in Authoritarian Regimes
A dissertation submitted in partial satisfaction
of the requirements for the degree
Doctor of Philosophy in Political Science
by
Eoghan Wallace McGreevy-Stafford
2020
c Copyright by
Eoghan Wallace McGreevy-Stafford
2020
ABSTRACT OF DISSERTATION
First, the Bad News:
Opposition Media in Authoritarian Regimes
by
Eoghan Wallace McGreevy-Stafford
Doctor of Philosophy in Political Science
University of California, Los Angeles, 2020
Professor Barbara Geddes, Chair
Why do dictatorships sometimes allow opposition groups to publish media, but at other
times forbid opposition media? I propose a theory that dictators tolerate opposition media
selectively in order to limit protests. I formalize the theory in a signaling model, from which
I derive several testable empirical implications. I illustrate the logic of the theory with a
case study of the Ben Ali dictatorship in Tunisia during its first five years (1987-1992), based
in part on interviews I carried out in Tunisia. I show how the theory explains variation in
Ben Ali’s willingness to allow opposition media, across both time and opposition groups. To
test the implications of the model quantitatively, I construct a panel dataset on ten Arab
countries with authoritarian regimes during 1992-2017. The data measure which regimes
allowed opposition groups to produce media in which years and are based on my research
on a wide range of opposition groups (of various ideologies and legal statuses) and of media
ii
(including newspapers, websites, and TV channels).
I find that dictators allow opposition media when their regimes are most likely to survive an
uprising, in order to signal their strength to citizens and discourage them from protesting.
In particular, in years when authoritarian regimes experience strong economic performance
— including low unemployment, high economic growth, and plentiful revenue from oil and
natural gas — they are far more likely to permit opposition media. After the “Arab Spring”
uprisings of 2011 revealed that the region’s authoritarian regimes were more vulnerable to
mass unrest than they previously appeared, those regimes became much less likely to tolerate
opposition media. By advancing a new theory and analyzing original empirical evidence, this
study contributes to our understanding of why media freedom varies in authoritarian regimes.
iii
The dissertation of Eoghan Wallace McGreevy-Stafford is approved.
Michael S. Y. Chwe
James D DeNardo
Tim Joseph Groeling
Daniel Simon Treisman
Barbara Geddes, Chair
University of California, Los Angeles
2020
iv
I dedicate this work to Ceren, with boundless gratitude and love.
v
Contents
Abstract
ii
Dissertation Committee
iv
Dedication
v
List of Figures
xii
List of Tables
xviii
Acknowledgments
xix
Vita
xxiii
vi
1 Introduction
1
1.1
My Argument in Brief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
1.2
Why study this question in Arab countries? . . . . . . . . . . . . . . . . . .
5
1.3
Outline of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
2 Amplifying
Pessimism:
A
Theory
of
Opposition
Media
Under
Dictatorship
8
2.1
Argument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
2.2
Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
20
2.3
Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
28
2.4
Equilibria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
31
2.5
Empirical Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36
3 “A Weird Period”: Opposition Media in Tunisia, 1987-1992
42
3.1
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
45
3.2
Variation in state tolerance for opposition media across groups and time
46
vii
. .
3.3
Why the difference in media policies for the secularist and Islamist opposition? 50
3.4
Why Ben Ali first prohibited and later allowed Islamist media . . . . . . . .
51
3.5
Why Ben Ali cracked down on Islamist media . . . . . . . . . . . . . . . . .
58
3.6
Why secular opposition newspapers were shut down, but only briefly . . . .
61
3.7
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
4 “Constructive Criticism”: A Cross-National Study of Opposition Media
in Arab Authoritarian Regimes
65
4.1
The Arab Region Since the Cold War . . . . . . . . . . . . . . . . . . . . . .
68
4.2
A Dataset of Arab Opposition Media . . . . . . . . . . . . . . . . . . . . . .
76
4.3
Explanatory Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
94
4.4
Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.5
Main Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4.6
Alternative Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
4.7
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
viii
5 Conclusion
140
5.1
Summary of Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.2
Contributions of this Study . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
5.3
Limitations of this Study and Suggestions for Future Research . . . . . . . . 147
5.4
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Appendix 1: Proof of result in model
152
Appendix 2: Regime Narratives
154
Appendix 3: Opposition Group Narratives
171
Appendix 4: Opposition Media Narratives
202
Appendix 5: Regime Leaders
229
Appendix 6: Histograms of continuous explanatory variables
241
Appendix 7: Dropping one regime at a time: tolerated
246
ix
Appendix 8: Dropping one regime at a time: banned
269
Appendix 9: Alternative coding of main opposition
292
References
294
x
List of Figures
4.1
Significant opposition groups per regime-year . . . . . . . . . . . . . . . . . .
87
4.2
tolerated over time: Algeria-Egypt . . . . . . . . . . . . . . . . . . . . . . . .
95
4.3
tolerated over time: Jordan-Morocco . . . . . . . . . . . . . . . . . . . . . .
96
4.4
tolerated over time: Sudan-Yemen . . . . . . . . . . . . . . . . . . . . . . . .
97
4.5
banned over time: Algeria-Egypt . . . . . . . . . . . . . . . . . . . . . . . . .
98
4.6
banned over time: Egypt-Jordan . . . . . . . . . . . . . . . . . . . . . . . . .
99
4.7
banned over time: Kuwait-Morocco . . . . . . . . . . . . . . . . . . . . . . . 100
4.8
banned over time: Sudan-Yemen . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.9
Residuals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
xi
4.10 Cook’s Distances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
4.11 Partial regression plot: unemployment . . . . . . . . . . . . . . . . . . . . . 114
4.12 Partial regression plot: fuel revenue per capita . . . . . . . . . . . . . . . . . 114
5.1
Distribution of unemployment rate (percentage) . . . . . . . . . . . . . . . . 241
5.2
Distribution of annual GDP per capita growth (percentage)
5.3
Distribution of fuel revenue per capita (2018 USD) . . . . . . . . . . . . . . 242
5.4
Distribution of oil revenue per capita (2018 USD) . . . . . . . . . . . . . . . 243
5.5
Distribution of natural gas revenue per capita (2018 USD) . . . . . . . . . . 243
5.6
Distribution of years regime leader has been in power . . . . . . . . . . . . . 244
5.7
Distribution of fuel revenue per capita (2018 USD), given greater than 0 . . 244
5.8
Distribution of oil revenue per capita (2018 USD), given greater than 0 . . . 245
5.9
Distribution of natural gas revenue per capita (2018 USD), given greater than 0245
xii
. . . . . . . . . 242
List of Tables
4.1
Regimes included in the dataset . . . . . . . . . . . . . . . . . . . . . . . . .
86
4.2
Bans and Toleration of Opposition Media . . . . . . . . . . . . . . . . . . . .
94
4.3
Explanatory Variables (continuous and integer) . . . . . . . . . . . . . . . . 105
4.4
Explanatory Variables (binary) . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.5
Tolerated Opposition Media (Regime FE) . . . . . . . . . . . . . . . . . . . 110
4.6
Tolerated Opposition Media (Regime FE), continued . . . . . . . . . . . . . 111
4.7
Effect sizes of focal variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
4.8
Opposition Media Bans (Regime FE) . . . . . . . . . . . . . . . . . . . . . . 118
4.9
Opposition Media Bans (Regime FE), continued . . . . . . . . . . . . . . . . 119
xiii
4.10 Tolerated Opposition Media (Country FE) . . . . . . . . . . . . . . . . . . . 121
4.11 Tolerated Opposition Media (Country FE), continued . . . . . . . . . . . . . 122
4.12 Opposition Media Bans (Country FE)
. . . . . . . . . . . . . . . . . . . . . 123
4.13 Opposition Media Bans (Country FE), continued . . . . . . . . . . . . . . . 124
4.14 Tolerated Opposition Media (Regime FE), disaggregated fuel . . . . . . . . . 125
4.15 Tolerated Opposition Media (Regime FE), disaggregated fuel, continued . . . 126
4.16 Opposition Media Bans (Regime FE), disaggregated fuel . . . . . . . . . . . 127
4.17 Opposition Media Bans (Regime FE), disaggregated fuel, continued . . . . . 128
4.18 Tolerated Opposition Media (Regime FE, OLS vs Logistic) . . . . . . . . . . 130
4.19 Tolerated Opposition Media (Regime FE, OLS vs Logistic), continued . . . . 131
4.20 Opposition Media Bans (Regime FE, OLS vs Logistic) . . . . . . . . . . . . 132
4.21 Opposition Media Bans (Regime FE, OLS vs Logistic), continued . . . . . . 133
4.22 Tolerated Opposition Media (Regime FE, alternative coding of main groups) 135
xiv
4.23 Tolerated Opposition Media (Regime FE, alternative coding of main groups),
continued . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4.24 Opposition Media Bans (Regime FE, alternative coding of main groups)
. . 137
4.25 Opposition Media Bans (Regime FE, alternative coding of main groups),
continued . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.1
Tolerated Opposition Media (Regime FE), dropped Algeria 1992-NA . . . . 247
5.2
Tolerated Opposition Media (Regime FE), continued, dropped Algeria 1992-NA248
5.3
Tolerated Opposition Media (Regime FE), dropped Bahrain 1971-NA . . . . 249
5.4
Tolerated Opposition Media (Regime FE), continued, dropped Bahrain 1971-NA250
5.5
Tolerated Opposition Media (Regime FE), dropped Egypt 1952-2011 . . . . 251
5.6
Tolerated Opposition Media (Regime FE), continued, dropped Egypt 1952-2011252
5.7
Tolerated Opposition Media (Regime FE), dropped Egypt 2011-2012 . . . . 253
5.8
Tolerated Opposition Media (Regime FE), continued, dropped Egypt 2011-2012254
5.9
Tolerated Opposition Media (Regime FE), dropped Egypt 2013-NA . . . . . 255
5.10 Tolerated Opposition Media (Regime FE), continued, dropped Egypt 2013-NA 256
xv
5.11 Tolerated Opposition Media (Regime FE), dropped Jordan 1946-NA . . . . . 257
5.12 Tolerated Opposition Media (Regime FE), continued, dropped Jordan 1946-NA258
5.13 Tolerated Opposition Media (Regime FE), dropped Kuwait 1961-NA . . . . 259
5.14 Tolerated Opposition Media (Regime FE), continued, dropped Kuwait 1961-NA260
5.15 Opposition Media Bans (Regime FE), dropped Mauritania 2008-NA . . . . . 261
5.16 Opposition Media Bans (Regime FE), continued, dropped Mauritania 2008-NA262
5.17 Opposition Media Bans (Regime FE), dropped Morocco 1956-NA . . . . . . 263
5.18 Opposition Media Bans (Regime FE), continued, dropped Morocco 1956-NA
264
5.19 Opposition Media Bans (Regime FE), dropped Sudan 1989-NA . . . . . . . 265
5.20 Opposition Media Bans (Regime FE), continued, dropped Sudan 1989-NA . 266
5.21 Opposition Media Bans (Regime FE), dropped Yemen 1978-2012 . . . . . . . 267
5.22 Opposition Media Bans (Regime FE), continued, dropped Yemen 1978-2012
268
5.23 Opposition Media Bans (Regime FE), dropped Algeria 1992-NA . . . . . . . 270
5.24 Opposition Media Bans (Regime FE), continued, dropped Algeria 1992-NA . 271
xvi
5.25 Opposition Media Bans (Regime FE), dropped Bahrain 1971-NA . . . . . . . 272
5.26 Opposition Media Bans (Regime FE), continued, dropped Bahrain 1971-NA
273
5.27 Opposition Media Bans (Regime FE), dropped Egypt 1952-2011 . . . . . . . 274
5.28 Opposition Media Bans (Regime FE), continued, dropped Egypt 1952-2011 . 275
5.29 Opposition Media Bans (Regime FE), dropped Egypt 2011-2012 . . . . . . . 276
5.30 Opposition Media Bans (Regime FE), continued, dropped Egypt 2011-2012 . 277
5.31 Opposition Media Bans (Regime FE), dropped Egypt 2013-NA . . . . . . . . 278
5.32 Opposition Media Bans (Regime FE), continued, dropped Egypt 2013-NA . 279
5.33 Opposition Media Bans (Regime FE), dropped Jordan 1946-NA . . . . . . . 280
5.34 Opposition Media Bans (Regime FE), continued, dropped Jordan 1946-NA . 281
5.35 Opposition Media Bans (Regime FE), dropped Kuwait 1961-NA . . . . . . . 282
5.36 Opposition Media Bans (Regime FE), continued, dropped Kuwait 1961-NA . 283
5.37 Opposition Media Bans (Regime FE), dropped Mauritania 2008-NA . . . . . 284
5.38 Opposition Media Bans (Regime FE), continued, dropped Mauritania 2008-NA285
xvii
5.39 Opposition Media Bans (Regime FE), dropped Morocco 1956-NA . . . . . . 286
5.40 Opposition Media Bans (Regime FE), continued, dropped Morocco 1956-NA
287
5.41 Opposition Media Bans (Regime FE), dropped Sudan 1989-NA . . . . . . . 288
5.42 Opposition Media Bans (Regime FE), continued, dropped Sudan 1989-NA . 289
5.43 Opposition Media Bans (Regime FE), dropped Yemen 1978-2012 . . . . . . . 290
5.44 Opposition Media Bans (Regime FE), continued, dropped Yemen 1978-2012
xviii
291
ACKNOWLEDGMENTS
I am deeply grateful to the members of my dissertation committee. As scholars they are
inspiring role models and as advisors they are dedicated and supportive guides. Professor
Barbara Geddes was always generous with her time and provided countless helpful
suggestions about my research.
I have benefitted immensely from her knowledge and
insight, her sympathy and understanding, and her unfailing advocacy for me. Professor
James DeNardo has encouraged and guided me since my first year of graduate school. His
commitment to providing feedback and advice was extraordinary.
Professor Daniel
Treisman always challenged me to question my assumptions and sharpen my arguments
and he provided a wealth of practical advice. Professor Michael Chwe provided critical
guidance on the many versions of my model and helped me learn how to streamline a
model to its essential elements. Professor Tim Groeling provided essential perspective on
the study of the media in terms of both substance and methodology.
In addition to my committee members, I have also received valuable advice on my research
from Professor Chad Hazlett, Professor Leslie Johns, Professor Jeffrey B. Lewis, Professor
Susanne Lohmann, Professor Barry O’Neill, Professor Michael Ross, Professor Thomas
Schwartz, and Professor John Zaller. I learned a great deal from working with Professor
Barbara Geddes and Professor Robert Trager as a research assistant. I am lucky to have
taken classes with UCLA’s dedicated and talented faculty, not only in the Political Science
department, but also in the departments of French and Francophone Studies, History, Near
Eastern Languages and Cultures, and Statistics. I could not have navigated graduate life
without the tireless support of the Political Science department staff. Special thanks are
due to Joseph Brown for his heroic assistance in overcoming bureaucratic obstacles.
For helping me grow as a teacher and deepening my love of teaching, I thank the professors
xix
with whom I taught, my fellow teaching assistants, and my students. In particular, I am
grateful to Professor James DeNardo and Professor Kathleen Bawn for serving as outstanding
role models and mentors. I am also very grateful for the guidance I received in designing my
first course from Dr. Kumiko Haas.
I am grateful for my friendships with fellow students in the Political Science department. I
learned so much from them and they supported me in many ways. I particularly want to
thank Maxim Ananyev, Raffaele Asquer, Shahin Berenji, Sarah Brierley, Albert Chang,
Chao-Yo Cheng, Marika Csapo, Sebastián Garrido de Sierra, Kristen Kao, Zsuzsanna
Magyar, Paasha Mahdavi, Naveed Mansoori, Manoel Gehrke Ryff Moreira, Yuree Noh,
George Ofosu, Amanda Rizkallah, Anton Sobolev, Clint VanSonnenberg, Andrea Vilán,
Kathryn Wainfan, and Feng Yang.
And thank you to everyone who attended my
presentations at COMPASS meetings for your many useful suggestions.
I was also very fortunate to make friends in the History and Anthropology departments. I am
so grateful for my friendships with Sam Anderson, Arnon Degani, Alma Heckman, Pauline
Lewis, Hanna Petro, and Anoush Sunni. Not only are they brilliant scholars who taught
me new ways of looking at the world, they provided never-ending support and fantastic
company.
I want to thank Kim Tran and Derrick Spiva for their support and for many wonderful
conversations, both serious and ridiculous.
I want to thank the faculty of Qasid and the staff of the Critical Language Scholarship Arabic
program in Amman for their fantastic work. I also want to thank my fellow CLS students,
especially Elizabeth Proehl and Katherine Downs, for making my time in Jordan such a
delight. I have received valuable feedback and training during the Empirical Implications of
xx
Theoretical Models workshops at Princeton University and Duke University and during the
Digital Research Start-Up Partnerships program at UCLA. I also benefitted greatly from the
comments of discussants, panelists, and audience members at many conferences, including
those of the Midwest Political Science Association, the Pacific Northwest Political Science
Association, the American Political Science Association, the Alexander Hamilton Center
at NYU, the Association for the Study of the Middle East and Africa, the Middle East
Studies Association, Southern California Comparative Political Institutions, and UCLA’s
Comparative Politics Association. I have also received so much helpful advice from the
fellows and staff of the Middle East Initiative at the Harvard Kennedy School of Government.
I am grateful for the financial support I received to support my studies and research from
UCLA’s Political Science Department, Graduate Division, and the Office of Instructional
Development, as well as from the Department of Education’s Critical Language Scholarship,
the Princeton Department of Politics, the Duke Department of Political Science, the New
York University Department of Politics, the Association for the Study of the Middle East
and Africa, and the Harvard Kennedy School of Government’s Middle East Initiative.
My fieldwork in Tunisia benefitted enormously from the outstanding research assistance of
Mohamed Ashraf Ayadi.
I thank Professor Peter Baldwin and his family for their incredible generosity.
I thank Dr. Carl Alasko for providing perspective and wisdom at a crucial time.
I thank Rachel Dart, Maggie Finnegan, and Quinnie Lin, for their advice, encouragement,
humor, and love.
xxi
I am so grateful for the love and support of my mother, Joyce, my father, Maeliosa, and my
siblings, Cian, Aelia, and Benen. You all amaze and inspire me and you have encouraged
me every step of the way.
The best way my life changed at UCLA began on my very first day there, at the orientation
for new graduate students, when I sat next to Ceren Abi, who is now my wife. As we made
our journey through graduate school together, I could always count on her to help me solve
problems, to lift me up when I struggled, and to remind me of the real sources of joy in
life. She displayed superhuman endurance as she listened to an inordinate number of my
presentation practices and read an endless stream of drafts. Her brilliance, wisdom, and
kindness inspire me, her sense of humor delights me, and her love sustains me.
xxii
VITA
Education
2010 M.Sc. in Political Theory, London School of Economics and Political Science
2006 A.B. in Social Studies, Harvard University
Publications
2017 “Stop the Presses! Media Freedom in Authoritarian Regimes: A Case Study of Ben
Ali’s Tunisia.” Journal of the Middle East and Africa. Vol. 8, No. 4.
2004 “Unemployment Rate Deception.” In Real World Macro. Dollars & Sense.
Awards
2019 Middle East Initiative Research Fellowship, Harvard Kennedy School of Government
2017 Dissertation Year Fellowship, University of California, Los Angeles
2016 Best Paper by a Student, Association for the Study of the Middle East and Africa
Annual Conference
2016 Digital Research Start-Up Partnership, University of California, Los Angeles
xxiii
2016 Research Grant, Association for the Study of the Middle East and Africa
2016 Empirical Implications of Theoretical Models, Duke University
2015 Collegium of University Teaching Fellowship, University of California, Los Angeles
2013 Travel Grant, Alexander Hamilton Center, New York University
2012 Graduate Summer Research Mentorship Fellowship, University of California, Los
Angeles
2012 Empirical Implications of Theoretical Models, Princeton University
2011 Critical Language Scholarship (Arabic), US Department of Education
xxiv
Chapter 1
Introduction
“For Mubarak, its value is based on the extent to
which it can contain forces, not unleash them.”
-Mohammed Sid Ahmed, journalist, on the loosening
of media restrictions in Egypt in the 1980s (quoted
in Ross [1986])
Less than three years after Hamad Isa Ibn Al-Khalifah ascended to Bahrain’s throne, he faced
a potent new challenger to his regime. Founded in November 2001, Al-Wefaq was an Islamist
organization that called for a genuine constitutional monarchy. The group also demanded
an end to the discrimination faced by the Shiite Muslims who make up the majority of
Bahrain’s population and are ruled by a state whose top officials are predominantly Sunni.
Despite Al-Wefaq’s call for fundamental political change, and its swiftly growing membership
and electoral support, King Hamad allowed the opposition group to openly operate media
1
outlets throughout the 2000s. During this time, his government did not block Bahraini
citizens’ access to Al-Wefaq’s website, nor did it shut down the group’s weekly newspaper.
But beginning in 2010, the king abruptly abandoned this permissive approach toward AlWefaq’s media outlets. In September of that year, the government banned the newspaper and
began blocking access to the website from inside Bahrain. The newspaper never reemerged.
The government restored access to Al-Wefaq’s website in 2012, only to resume blocking it in
2016.1
Why do authoritarian regimes sometimes allow opposition groups to run their own media
outlets, but other times prevent them from doing so? Indeed, why do dictators — a group
not generally known for their enthusiasm about the airing of views different from their
own — ever allow their opponents a public platform? What’s in it for the dictator? In
what circumstances do dictators expect to benefit from allowing opposition media outlets,
and when do dictators think they are better off banning them? As far as I am aware, no
previous research has tried to answer these questions in a general way. In this study, I offer
a theory to explain the motivation of some dictators to allow opposition media, and why
in other cases they choose not to do so. I represent this argument abstractly in a simple
game-theoretic model, from which I derive testable predictions about when we should and
should not expect to see opposition media out in the open in dictatorships. I illustrate my
argument with a qualitative case study of the Ben Ali dictatorship in Tunisia, based in part
on fieldwork I conducted there. Finally, I test the model’s predictions quantitatively, using
an original dataset on authoritarian governments’ tolerance and intolerance of opposition
media in several Arab countries during the last few decades.
1
For the sources this paragraph is based on, see the narratives on Al-Wefaq and its media in Appendices
3 and 4.
2
1.1
My Argument in Brief
My answer to this puzzle starts from the premise that opposition leaders do not always
want to mobilize their followers to protest against authoritarian regimes. Often they judge
that protest is too risky. They may fear that the protest would fail to achieve anything,
that protestors could be violently repressed, and that failure and repression will discourage
citizens from joining future protests that might have had a better chance of succeeding.
Opposition leaders might consider a couple factors in particular when considering the likely
outcome of a protest. First, are many citizens likely to join the protest once it gets started?
Second, how likely are the security forces — the police, the army, etc — to use violence
against protestors? No one can know the answers to these questions perfectly in advance,
but elites in both the opposition and the regime will tend to have sources of information to
base their guesses that most ordinary citizens lack. For example, elites in a dictatorship will
tend to have a better idea than ordinary citizens about how well the economy is performing
nationally, which could affect the number of citizens who are dissatisfied with the regime and
potentially willing to protest. Elites would also have a better idea of whether the security
forces are personally under the control of the individual who leads the regime or whether a
larger group of regime elites shares control over the security forces. This factor will affect
the likelihood that the security forces would be willing to use violence against protestors to
keep the dictator in power.
When elites have information that protest is unlikely to succeed, opposition leaders will want
to discourage their followers from protesting. In these cases, authoritarian governments will
allow opposition groups to publish their own media, so that opposition leaders can reach as
large an audience as possible with the message that the time is not ripe to protest. After
all, citizens are more likely to believe that conditions do not favor protesting if they hear
this message from opposition groups, rather than just from state-run media, since the latter
3
would always want to send that message. When protest is likely to be more successful,
opposition leaders will not be willing to use their media to discourage their supporters from
protesting. In such cases, the government will not allow the opposition to publish at all.
We might suppose that an authoritarian regime will only allow opposition groups to have
media when the regime is very weak: that it will allow opposition media as a concession
when it faces a high risk of mass unrest. But the theory I present in this study predicts
the opposite pattern. Regimes will only allow opposition groups to have media when the
opposition is either unable to mobilize citizens to protest or the regime can count on the
security forces to violently repress protests if they occur. Regimes that face the greatest
threat of unrest — because the opposition can plausibly mobilize more citizens to protest
than the security forces would be willing or able to suppress — will prevent opposition groups
from publishing media.
This study aims to add to political science’s understanding of why media freedom varies
under authoritarian rule, how and why opposition groups sometimes act in ways that benefit
the regimes they seek to change, and how dictators use the trappings of democracy to
prolong authoritarianism. If my argument is right, we should be very skeptical about whether
dictators who allow opposition media will continue to do so for long, let alone embark on
other forms of political liberalization. Rather, such tolerance is more likely intended to
prolong their rule than to guide their countries towards democracy. And just as quickly as
they opened space for opposition media, future circumstances may lead the rulers to decide
they would be better off shutting that space down.
4
1.2
Why study this question in Arab countries?
I explore my theory empirically in the context of modern authoritarian regimes in the
Middle East and North Africa, specifically in Arab countries. I conduct a qualitative case
study of Ben Ali’s policies toward opposition media during his first five years as president
of Tunisia (1987-1992) and a quantitative study of authoritarian regimes in ten Arab
countries over a twenty-six year period (1992-2017). The prevalence and persistence of
authoritarianism in the Arab region makes it a fertile ground for the study of opposition
media under authoritarian regimes. Additionally, the breakdown of the social contract
between governments and citizens throughout much of the region and the rise of opposition
movements, particularly Islamists, beginning in the 1970s brought renewed relevance to
dictators’ problem of avoiding being overthrown by their own citizens. The danger of mass
unrest that dictators face in the region has remained pressing throughout the last three
decades, as the “Arab Spring” and other uprisings have dramatically underscored. Many
Arab authoritarian regimes during this period have carried out political liberalization
measures that were often later reversed and in most cases did not bring an end to
authoritarian rule.
Thus the recent history of the region is particularly relevant for
understanding how authoritarian leaders use pseudo-democratic reforms to promote the
longevity of their regimes.
1.3
Outline of the Study
In the next chapter I present in more detail my theory of why dictatorships sometimes
allow opposition groups to produce media and sometimes do not. I discuss the literatures
that this theory builds on, including studies on media freedom in authoritarian regimes,
5
relations between authoritarian governments and opposition groups, authoritarian regimes’
strategies for preventing and containing threats to their survival from mass unrest, and the
ways opposition elites influence collective action. I present a game-theoretic model that
formalizes my theory and then derive from it empirically testable implications that I will
explore in the next two chapters.
Chapter 3 presents a case study of Tunisia during the first five years of the Ben Ali
dictatorship to illustrate how the strategic dynamics in my model play out in practice.
Drawing on news articles, human rights group reports, secondary sources, and field
interviews, I show how predictions from the model help explain why Ben Ali allowed some
opposition groups to produce their own newspapers but not others, and why he allowed
opposition groups to publish in some years but not others. At any given time, he was more
likely to allow media produced by those groups that posed the least threat of mobilizing
protests against his government. For any given group, he was more likely to allow it to
publish a paper at times when its capacity to mobilize was lower and the government’s
capacity to repress protests was higher.
In Chapter 4, I test the theory cross-nationally and over a longer and more recent time
period. Specifically, I explain how I identified all significant opposition groups under Arab
authoritarian regimes during 1992-2017. I then describe how I created a dataset measuring
whether each regime allowed or banned media produces by these groups in each year. To
my knowledge, this is the first cross-national dataset on toleration of opposition media by
authoritarian governments. I present results from linear probability models, using regime
fixed effects, that support many of the predictions from the model. Specifically, regimes are
more likely to tolerate opposition media when their revenues from extracting oil and natural
gas are higher than usual, when unemployment is lower than usual, and when GDP growth
is higher than usual. After the “Arab Spring” uprisings, they became less likely to allow
6
opposition media. I also present various robustness tests, using different model specifications
and an alternate version of my dataset based on slightly different coding rules.
Chapter 5 concludes. I summarize the findings, discussing the extent to which the hypotheses
helped to explain the Tunisian case and were supported by patterns in the cross-national
analysis. I discuss how the findings of this study extend and differ from previous results in
the literature. I also discuss the limitations of the study and recommend future directions
for research on the topic of opposition media under authoritarian regimes.
7
Chapter 2
Amplifying Pessimism: A Theory of
Opposition Media Under Dictatorship
No one could accuse Jordan’s Muslim Brotherhood of being sore losers. In November 1993,
Jordan held a parliamentary election, in which the Brotherhood’s Islamic Action Front
(IAF) saw its seat share fall from 22 to 16 (Redden 1993). Yet, when King Hussein gave a
post-election press conference, a reporter from the IAF’s newspaper As-Sabeel prefaced his
question as follows: “To begin with, let us congratulate you and ourselves on this national
celebration. Jordan has provided an unprecedented example in the Arab world of relations
between the government and the opposition, especially the Islamists” (BBC Monitoring:
Middle East 1993). Leaders of Bahrain’s Islamist Al-Wefaq party offered similarly kind
words for their monarch on their website in 2002: “The political societies were honored to
meet with His Majesty the King on Friday, 13 September 2002, where he praised the great
role played by political societies in this historical stage” (Accord National Islamic Society
2019a).
8
Why would opposition groups engage in such an anodyne public discourse? Although these
groups may seek substantial political change as a long-term goal, I argue, they often calculate
that it is not in their short-term interest to challenge the state. Authoritarian rulers are
only too happy to amplify such accommodating messages, in hopes of dissuading the groups’
followers from taking to the streets. In this way, allowing opposition organizations to produce
their own media can sometimes serve to shore up a dictator’s powers.
When opposition media under Arab authoritarian governments do engage in criticism, it is
often quite mild, in comparison to the stated long-term goals of the opposition organizations.
Consider the top stories on the news website As-Sahwa, produced by Yemen’s Islamist Islah
party, on a day in February 2010. Though not full of praise for Yemen’s government, the
site was not exactly scathing in its criticisms, either. The top article focused on problems
experienced by Yemeni students studying abroad as a result of the overlapping responsibilities
of the Ministry of Finance and the Ministry of Higher Education. Another leading article
reported on a Yemeni businessman’s call for a conference to improve the country’s investment
climate (Al-Sahwa Net 2019). Opposition media in authoritarian Arab countries have often
limited themselves to criticizing minor policies of their governments, while stopping short
of calling for fundamental political change. Writing in the newspaper Ash-Sha‘b during the
Sadat era, a leading Muslim Brotherhood official wrote that “complete and comprehensive
reform” was “a distant goal” and that, for the time being, “the Muslim Brotherhood offers
its advice to all Egyptian governments, wishes them all success” (quoted in Lust-Okar [2005:
164-165]).
Of course, opposition groups frequently are not permitted to publish forceful criticisms of
their governments even if they want to. Authoritarian governments typically place tight
limits on the content the media publish and have the power to enforce those limits. And
yet opposition groups that want to challenge a regime that limits the content they can
9
publish often have an alternative to playing by the government’s rules: silence itself can
speak volumes. By refusing to publish at all, a group can draw attention to the restrictions
the government is imposing on what they can say. Journalists under authoritarian regimes
have often gone on strike to put a spotlight on the censorship they face — and by extension,
the dictator’s fear of independent voices. Examples from recent decades include Algeria in
1998 (Goldstein 1999), Russia in 2003 (Associated Press 2003), Egypt in 2006 (Committee to
Protect Journalists 2007), Tunisia in 2009 (BBC Monitoring: Media 2009c), China in 2013
(Coonan 2013), and Sudan in 2018 (Hendawi 2018).1 Thus, while the existence of opposition
media can signal a dictator’s strength, censorship can reveal his weakness.
2.1
Argument
In this chapter, I present a novel theory to explain why authoritarian regimes sometimes
allow their opponents to publish their own media, while in other circumstances regimes deny
their opponents a voice. In this section, I discuss the intuition of the argument. In the next
section, I review the relevant scholarly literature that this theory builds on and the new
contribution it makes to this literature. In subsequent sections, I present a formal (gametheoretic) model that reflects the logic of my theory. Finally, I derive hypotheses from the
model that I will empirically test in the next two chapters.
I define opposition groups as organizations that seek to substantially change the policies
and/or political institutions of their government. I define toleration of opposition media at
1
Instances abound of newspapers publishing a blank section in their pages where a censored article would
have gone. Examples of this practice can be found in the Ottoman empire under Allied occupation after
World War I (Baykal 2019), South Africa under Apartheid during the 1980s (Parks 1986), Tunisia during
the Gulf War in 1991 (Donnadieu 1992), and Egypt under the Supreme Council of the Armed Forces in 2011
(Purkiss 2011). Because Brazil’s military dictatorship in the 1970s prohibited newspapers from publishing
blank spaces, one paper began publishing “recipes for inedible dishes” in place of censored articles (Dassin
1979: 4).
10
the level of the individual media outlet. Thus, a government can tolerate media produced
by some opposition groups, while not allowing other groups to have media. It can also
tolerate some media outlets produced by an opposition group, while blocking other outlets
produced by the same group (or preventing them from launching them at all). Toleration
of opposition media as I use the phrase can be best be defined negatively, in terms of what
constitutes not tolerating a given media outlet. A lack of toleration can be manifested in
either of two ways, at the production level or at the access level. First, a government can
prevent or attempt to prevent the media outlet from forming as an organization, producing
media materials (articles, audio recordings, videos, etc), or making the materials public. For
example, the government could arrest all of the outlet’s personnel or destroy its physical
equipment.2 Second, a government can prevent or attempt to prevent individuals within
its borders from consuming the content that the outlet has made public, for instance by
blocking a website or jamming a TV signal.3 Only if the government allows both production
and access do I consider it to be tolerating a media outlet.4
I argue that dictators will sometimes allow opposition groups to produce their own media
outlets when anti-government protest would be likely to fail. The success and failure of
protests exists on a spectrum, of course. Success might mean overthrowing a regime or
simply forcing the regime to make some policy concession. Failure might simply mean
protestors don’t achieve the changes they demand or it could mean that large numbers of
2
Not all interference with production constitutes a failure to tolerate the outlet overall. A government
might use threats to prevent an outlet from publishing a particular article, for instance, while still allowing
it to publish other articles. As this example suggests, toleration of an outlet could be thought of as existing
on a spectrum. Nonetheless, I will hue to a binary conception of tolerating media for simplicity, focusing
on whether an opposition media outlet is allowed to publish some content — though it may face some
restrictions — or is simply not allowed to publish at all.
3
Note that interference with users’ access to a media outlet is almost never absolute — rather such efforts
are usually intended simply to make access prohibitively difficult. For instance, a government might block a
website and outlaw citizens from using virtual private networks (VPNs). It may still be possible for a citizen
to acquire and use a VPN to access the blocked website. Nonetheless, I would consider the government not
to be tolerating the website.
4
Note that legality is not a part of my definition; what matters is whether the government allows the
media outlet to function and citizens to access it, regardless of its legal status. Thus, an outlet could be
legal in theory but not tolerated in practice, or it could be illegal in theory but tolerated in practice.
11
them are arrested, beaten, or killed. In addition, a protest movement that initially appears
to be failing might ultimately succeed, or vice versa. Generally speaking, opposition leaders
tend to prefer that their followers protest when success of some sort is very likely and to
refrain from protesting when some sort of failure is very likely. The group’s strength would be
sapped if their supporters were imprisoned or killed en masse.5 In addition, a failed protest
might make citizens less likely to protest in the future. Moreover, the government may
retaliate by imprisoning or killing opposition leaders themselves (Ginkel and Smith 1999).
If the government gives an opposition group a public platform when protest would be very
likely to fail, the group will tend to use it to discourage its followers from protesting.6 Thus,
dictators are most likely to allow opposition media when protest is least likely to succeed,
in order to ensure that the opposition’s message to its followers to stand down reaches as
many of them as possible.7
What sort of factors will determine the likelihood that protest would succeed? One important
factor is the extent of discontent with the current regime. If many citizens support the
status quo and are opposed to changing current policies, leaders, or institutions, a protest
movement will attract few followers.8 Gurr (1970: 13) argues that participation in collective
action is often driven by a sense of “relative deprivation”: when individuals’ circumstances
fall short of “the goods and conditions of life to which people believe they are rightfully
entitled.” The extent of grievances in a society, and thus the likelihood that many citizens
5
An opposition movement may gain strength if repression provokes a public backlash (Tarrow [1994: 92]),
but a group will hardly benefit if repression is so heavy that its movement is crushed.
6
Different opposition groups may have different risk thresholds, with some groups willing to protest even
when the odds of success are much lower than what other groups would accept. The key point is that most
groups have some threshold at which failure is so likely that they prefer not to protest.
7
As long as the probability that protest would succeed is not zero, it is in the interest of a dictator to
convince opposition supporters not to protest.
8
Throughout, I use the term “citizens” loosely to refer to all individuals that are subject to the rule of
a particular government, whether or not they literally hold citizenship. In some countries, including many
countries in the Persian Gulf region, large shares of the population are non-citizen migrants. (89 percent of
the population of the United Arab Emirates are non-citizens.) At the same time, the vulnerability of these
populations to deportation limits their ability to protest, relative to those with citizenship. See: Fargues
(2015).
12
would join a protest movement once it began, may thus depend, for example, on the level of
unemployment and other aspects of economic performance (Hendrix, Haggard, and Magaloni
2009; Caren, Gaby, and Herrold 2017; Geddes, Wright, and Frantz 2018).
Another factor that affects the likelihood of successful collective action is how the security
apparatus — the police, the army, and other armed instruments of the state — would respond
if large numbers of citizens did join protests. If angry citizens fill the streets and the security
forces are unwilling to stop them, a regime is very likely to fall; conversely, the regime is very
likely to survive if the security forces are willing to use violence to crush the demonstrations
(Brownlee, Masoud, and Reynolds 2015). (Whether or not protests are likely to meet with
repression is a key factor in the “political opportunities” literature. See, for example, Tarrow
[1994].) A number of factors in turn influence the willingness of the security apparatus to
either stand by the regime or side with the protestors. Committing violence against civilians
can harm the prestige, morale, and unity of the security apparatus, particularly militaries,
which typically see themselves as the defenders of the nation (Bellin 2012). Thus, security
officers will not take the decision to inflict violence on protestors lightly, but will weigh the
risks of violence against the risks of political change. For instance, when the security forces
are personally controlled by the dictator and promotions depend on loyal to him, security
officers are likely to fear that a transition to a new regime would cost them their livelihoods,
or worse. By contrast, if the security forces are controlled collectively by a large group of
regime elites, or if the security forces enjoy considerable institutional autonomy, they are
more likely to retain their jobs and privileges even if the dictator — or the whole regime
— falls (Bratton and van de Walle 1997; Bellin 2012; Brownlee, Masoud, and Reynolds
2015). Other factors that influence the security forces’ ability and willingness to use violence
include the state’s fiscal resources (Bellin 2012; Josua and Edel 2015) and diplomatic support
from foreign governments (Bellin 2012). The security forces may also be less likely to use
repression the larger the crowds of protestors (Bellin 2012), so more widespread grievances,
13
to the extent that they contribute to higher participation in protests, may also decrease the
likelihood of state violence against protestors.
Leaders of the regime or the opposition are likely to have information about the level of
support for the opposition and the regime’s ability and inclination to repress to which
ordinary citizens do not have access. For example, the level of support for the opposition
may be high if there are many jobless citizens (or support may be low if there are few out
of work). Yet, while an ordinary citizen knows if she, her family, and her friends are out of
work, she may not know the national unemployment rate. Citizens may also be unaware of
whether the dictator has established tight personal control over the military and thus
whether the military would be likely to follow orders to use violence against protestors.
Regime leaders have state bureaucracies to provide them with information on economic
performance and networks of spies that can report on the level of dissent. Of course, as
Wintrobe (1998) notes, a dictator’s information about how well the government’s economic
policies are functioning and how discontent citizens are may be distorted by the incentives
of subordinates to paint a rosy picture of their performance to the regime leader.
Nonetheless, it is reasonable to assume that the machinery of state conveys some
information to the dictator about these issues that citizens cannot directly observe. And
while the dictator may not be certain whether the security forces would follow orders to
shoot protestors — as Ben Ali and Mubarak learned the hard way — he certainly has an
informational advantage over the person in the street, due to direct experience interacting
with the security apparatus and other regime elites. Opposition leaders typically also have
some insider information that citizens lack, stemming from greater education, resources,
investment in gathering political information, and direct interaction with high government
officials. So citizens who are discontent with the regime would have an interest in basing
their decision to protest in part on cues from opposition leaders, who share some interests
with discontent citizens, but have better information about some factors that affect the
14
chances for successful protest.
I therefore argue that, when exogenous factors that tend to foster support for the opposition
are low and/or the regime has reliable repressive capacity, the government is more likely to
allow opposition groups to have media outlets, so that opposition leaders can more widely
send a message to their followers not to protest. This demobilizing message may be explicit
or followers may read it between the lines if the opposition media do not mention protest and
call for only modest governmental reforms.9 Conversely, if the regime’s repressive capacity
is low and support for the opposition is high, opposition leaders will prefer to mobilize their
followers to protest. A ruler will not allow the opposition to publish such messages, so the
opposition will prefer not to publish at all, since the citizens would interpret the publication
of tame opposition media as a signal that the opposition leaders think they should not
protest.
Therefore, the more likely protest is to succeed, the less likely we will see opposition media.
It follows that, if an opposition group does not produce any media, citizens can infer that the
group may have been prevented from doing so, precisely because conditions are propitious
for collective action. However, the citizens cannot make such an inference with certainty,
because it is also possible that the group did not produce any media because it simply
lacked the necessary resources. Citizens’ ability to infer the possibility (though not the
certainty) of censorship from the absence of opposition media paradoxically explains the
credibility of opposition media when they do appear and discourage protest. It may seem
counterintuitive that such a message would ever be convincing to citizens, since I also argue
that the government would not tolerate media that sent any other message. But when
citizens see opposition media discouraging protest they know that opposition leaders had
another option: if they had believed protest was likely to succeed, they could have refused to
9
Presumably even the weakest opposition groups must engage in some public criticism of the government,
lest they lose all credibility with their followers, or their followers stop consuming the group’s media.
15
send a discouraging message at all. Such a stance would have compelled the government to
prevent them from publishing their media, and citizens would infer from the lack of opposition
media that such censorship may have occurred and raise their expectation that protest would
succeed.10 Therefore, if opposition leaders do publish media discouraging protest, citizens
know the opposition leaders genuinely believe the chances of success are too low to justify
protesting. Thus, allowing the opposition to have media — provided the group is willing
to send a demobilizing message — would most effectively persuade the group’s supporters
not to protest. But if the opposition is not willing to send a demobilizing message, then
not allowing them to have any media would be to the regime’s advantage, preserving some
uncertainty about whether protest was likely to succeed (and so the regime prevented the
group from publishing) or whether the group simply lacked the resources to produce any
media.
It must be acknowledged that, in some cases, dictators may succeed in using some
combination of bribes and threats to convince opposition leaders to publish demobilizing
messages even when they would prefer to mobilize their sympathizers.
However, such
efforts are costly for the dictator. And the better the chances of successful protest, the
more extensive and costly such efforts would have to be.
If the chances of protest
succeeding are high enough, the dictator may not be able to afford to co-opt or cajole
opposition leaders into publishing demobilizing messages. Similarly, in the model of Guriev
and Treisman (2019), an incompetent dictator can offer bribes to independent elites to
report that he is actually competent. In some cases, the elites will accept the bribes and
falsely report that the leader is competent. But in other cases, the cost of such co-optation
is too high for the dictator. Therefore, if citizens observed opposition media discouraging
10
If the government allowed the opposition to publish media that encouraged protest, citizens would
have an even higher expectation of protest success: it would be certain, rather than merely probable, that
opposition elites thought the chances of successful protest were at least high enough that protest was worth
trying. Thus, when the chances of protest success are high, disallowing opposition media is partly effective
at lowering citizens’ estimates of the chances of successful protest.
16
protest and knew it was possible that the opposition leaders had been coopted or coerced,
their estimate of the probability of protest success would still be lower than if they had not
seen such media, because they also know that the opposition leaders send sincere messages
some of the time. The greater the dictator’s ability to bribe and threaten the opposition
elites, the more often we will see tolerated opposition media. But in some cases, we would
still see governments ban opposition media, and those would still be the cases in which
exogenous factors made protest most likely to succeed.
The logic of my theory is similar to that of Cowen and Sutter (1998) and Cukierman and
Tommasi (1998), who argue that voters are more likely to support a policy proposal made by
a politician whose ideological preferences make her the least likely to advocate the policy. For
instance, with his long record as an anti-communist hawk, Nixon had particular credibility
in advocating diplomatic relations with Mao’s China (an example cited in both studies).
Similarly, voters may be more willing to accept pro-market reforms when they are proposed
by left-wing parties (Cukierman and Tommasi 1988: 180).
One might suppose that the main effects of opposition media on citizens’ attitudes are to
increase their dissatisfaction with the current regime, by reporting its abuses and failures, and
to increase their support for the opposition, by highlighting the benefits of the opposition’s
proposals and the virtues of its leaders. But the same logic that implies that opposition
media are particularly credible when they indicate that the regime is strong also implies that
opposition media reports disparaging the regime and touting the opposition will have little
effect on citizens’ beliefs. Such messages do not convey new information, because citizens
already assume opposition groups have negative views about the regime and positive views
about themselves. This is not to say that all citizens will disagree with such reports from
opposition media. However, if some citizens do agree with these reports, it will be because
they already had a negative view of the regime and a positive view of the opposition. Such
17
reports from the opposition will not tend to change citizens’ attitudes.
I do not mean to convey the impression that authoritarian governments only choose whether
to allow or not allow opposition media on the basis I have described in my theory. Political
calculations are almost always far more complex than that, and no doubt other factors often
shape this policy choice as well. In particular cases, those factors may be of such importance
that the empirical patterns my theory predicts (discussed in the last section of this chapter)
are violated. In some types of regimes, other dynamics may systematically outweigh those
of my theory, suggesting certain scope conditions. For example, I consider armed opposition
groups beyond the scope of my theory and do not focus on them in the empirical chapters.
DeNardo (1985) argues that groups that rely on violent tactics are generally not able to
recruit as many participants as groups that rely on peaceful strategies (218), and at the time
same time, violent groups have less need to recruit large numbers of participants in order
to effectively threaten a state (189). Therefore, the ability of such groups to use media to
encourage or discourage the public to take part in an uprising may be more limited and
less important to a government, in comparison to peaceful groups. Thus, other incentives
may have a bigger effect on rulers’ willingness or unwillingness to allow such groups to have
media. I also do not claim that the theory can explain whether authoritarian regimes allow
opposition groups with very few members or sympathizers to produce media. The media of
the smallest groups may be unable to convince many citizens either to protest or to refrain
from protesting. So again, other considerations may dominate regimes’ decisions to tolerate
their media or not. For instance, a government might choose not to allow a tiny group to
produce even a tame media outlet, for fear that such an outlet would increase the number
of citizens who were even aware of the group and the group would begin to attract more
followers.11
11
Such groups may also be particularly unlikely to have the resources to produce media, in which case we
would almost never observe them producing media even when the regime would have been willing to tolerate
them.
18
I want to underscore that toleration of opposition media does not necessarily mean that a
government does not manipulate its citizens’ access to information. In general, there are a
few broad ways that governments can manipulate citizens’ access to information. First, they
can prevent the media from reporting certain information that would harm officials’ interests.
Second, they can block individuals’ access to such reporting. I refer to both types of efforts as
censorship.12 Thirdly, governments can also make media outlets — both the directly staterun media and private media — produce content that will favor state interests, including
disinformation. Such efforts constitute propaganda. Not allowing opposition media at all —
whether disrupting the operations of a media outlet or impeding citizens’ access to its content
— can be seen as an extreme form of censorship. However, tolerating an opposition media
outlet, as I have defined toleration, does not necessarily mean it is completely uncensored:
as noted, a government may allow an opposition group to regularly publish articles or other
content, while compelling them not to cross certain red lines regarding what cannot be
included in that content. Moreover, if a government follows the strategy I have described,
it only tolerates opposition media selectively, when the opposition shares its interest in
avoiding protest. When conditions differ, the government prevents opposition media in
order to increase citizens’ uncertainty about the chances of successful protest. Thus, the
strategy involves limiting citizens’ access to information at times, in order to manipulate
their beliefs.13
12
I define censorship more broadly than some, who use it to mean only the seizing or blocking of material
that has already been produced. Efforts to induce what is often called “self-censorship” fall under my
definition of censorship. As Sakr (2003: 31) argues: “Censorship is achieved not only through direct
suppression of content, but also by more fundamental and less visible means, including regulation of media
ownership, regulation of entry to the profession of journalism and regulation of printing and distribution, as
well as extra-judicial intimidation of media practitioners and bars on access to information.”
13
A government may also let some opposition groups produced media but not other opposition groups.
Furthermore, the way a government treats opposition media outlets may be quite different from how it treats
other, less partisan media outlets.
19
2.2
Literature Review
All governments place some limits on what the mass media can publish, but authoritarian
regimes typically place much more onerous restrictions on the media than democracies do
(Stein 2016). According to most studies, the primary reason for authoritarian governments
to manipulate the media is to prevent citizens from engaging in collective action against
the regime. For instance, the media can influence citizens’ beliefs about the behavior,
trustworthiness, or competence of government officials, and other factors that may make
citizens more opposed to or supportive of the government.
Censorship of negative
information about the government, as well as propaganda touting its virtues, may lead
more citizens to support the government, or at least to acquiesce in its rule (Geddes and
Zaller 1989; Petrova 2008; Stockmann and Gallagher 2011; Cho, Lee, and Song 2017;
Guriev and Treisman 2019). Even if many citizens are dissatisfied with their government,
and want to bring about a change in its policies or even fundamental institutions, they may
be unwilling to take part in collective action unless they believe that enough other citizens
will also participate.
Citizens may therefore only be willing to protest if it becomes
common knowledge that grievances are widespread (Chwe 2001). As noted, the greater the
number of citizens that participate, the more likely the protest is to end in a successful
regime transition, rather than repression (Bellin 2012).
numbers”:
Moreover, there is “safety in
conditional on the security forces using violence against the crowds, a
protestor’s probability of suffering violence personally declines in the number of
participants (DeNardo 1985 [194]). By manipulating media reports about opposition to
and support for the regime, a government can prevent citizens’ dissatisfactions from
becoming common knowledge (Kuran 1991; Lohmann 1994; Egorov, Guriev, and Sonin
2009; Lorentzen 2014). Finally, the media can potentially spread logistical information
about collective action, for instance concerning the timing and location of a planned
protest. By censoring such information, governments can make it less likely that citizens
20
will learn how they can participate in a protest or even that it is happening (Walgrave and
Manssens 2000; Little 2016).
There is considerable debate over the role of the Internet and other modern communication
technologies in shaping governments’ ability to manipulate information. Some scholars argue
that the Internet is making government interference in the media increasingly ineffective,
because the Internet is more difficult for governments to control than print or broadcast
media (Lorentzen 2014).14 Thus, the spread of Internet access may make it increasingly
difficult for governments to censor information that casts them in a bad light or that reveals
the public’s grievances (Stein 2016). The Internet generally and social media in particular
also make it easier for ordinary citizens to communicate with each other directly, which may
facilitate logistical coordination of protests (Little 2016; Stein 2016). However, other studies
note the myriad ways in which authoritarian governments are adapting to the spread of
Internet access, deploying a wide range of tools for controlling what citizens can post and
see online (Whitten-Woodring and Van Belle 2014; Rød and Weidmann 2015; King, Pan,
and Roberts 2017). Moreover, governments themselves can use the Internet as a tool to more
effectively disseminate propaganda to the public (Pearce 2014; Rød and Weidmann 2015).
The rise of the Internet and other communication technologies certainly does not seem to
have made government manipulation of information obsolete.
In particular, while opposition groups around the world have steadily expanded their online
media presence, governments can and do block opposition websites and intimidate and arrest
their authors (Sweis 2015; Devitt and Pinchuk 2018; OpenNet Initiative 2020). The use of
14
Lorenzten (2014) however argues that governments can compensate for the difficulty of controlling online
media to some extent by restricting traditional media more heavily. However, as legacy media outlets continue
to migrate to online platforms, such a strategy may prove increasingly ineffective. Shadmehr and Bernhardt
(2015) argue that a ruler would always be better off if he could commit to censoring slightly less than he
does in equilibrium, because citizens would update less negatively after unfavorable news. In some cases, a
higher cost of censorship — i.e. due to new information technologies — can make the ruler better off for
this reason.
21
https by international social media platforms like Facebook and Twitter has lately been
a boon to dissident groups, because governments can no longer block individual profiles
without blocking the entire platform (Clark et al 2017: 4), which risks a public backlash (for
example: Beaumont 2011). However, authoritarian governments’ track record of adapting
to technological advances suggests it may only be a matter of time before they develop a
solution to this predicament. In any case, as local social media platforms proliferate in
countries around the world, governments may increasingly be able to block platforms based
outside their borders with relative impunity (BBC News 2017c).
Despite their incentives and ability to interfere with the media, there exists considerable
variation among authoritarian regimes in the extent to which they do so. A growing literature
examines the potential benefits to authoritarian regimes of allowing a modicum of media
freedom and explaining why some regimes allow more media freedom than others. One
set of theories argues that some authoritarian governments allow the media freedom to
conduct investigative journalism into the behavior of low-ranking government officials, in
order to deter corruption and shirking (Egorov, Guriev, and Sonin 2009; Lorentzen 2014).
Similarly, Huang, Boranbay-Akan, and Huang (2019) argue that authoritarian governments
may allow the media to report on protests against local officials precisely because the media
can facilitate the growth of protests. Allowing protests on local issues to spread may deter
abuses by local officials, limiting discontent with the regime as a whole that might otherwise
lead to nationwide protests. Wintrobe (1998) has noted that limits on citizens’ expression
can produce a “dictator’s dilemma”: repressive governments may not learn how widespread
public dissatisfaction is until it explodes into mass unrest. Thus, governments may allow
some freedom of expression in order to uncover the extent of discontent and the issues that
are feeding it, so they can adjust unpopular policies before citizens rise up against the regime
(King, Pan, and Roberts 2013; Chen and Xu 2017).15
15
Chen and Xu (2017) also argue that, if citizens observe that they are allowed to communicate their
grievances publicly, but that few other citizens are doing so, they may realize that their grievances are not
22
Manipulation of the media can also backfire in various ways. Governments may be better off
allowing the media to report some news that reflects unfavorably on the government, because
in the absence of any bad news, citizens might assume the government’s performance is even
worse than it is (Shadmehr and Bernhardt 2015). Gelbach and Sonin (2014) argue that
citizens can learn over time how much the government manipulates the news; if manipulation
is too extreme, citizens will not believe favorable news about the government, and may even
stop consuming news media altogether. Such developments would make it harder for the
government to influence citizens’ behavior. Thus, governments may want to occasionally
allow some unfavorable stories to be reported so that citizens will believe the favorable
reports. If citizens have access to alternative information sources, such as foreign media,
censorship may be exposed and stoke opposition to the government (Gläßel and Paula 2019;
Knight and Tribin 2019). Hobbs and Roberts (2018) also provide empirical evidence from
China that, when the government begins to block previously uncensored information sources,
it can inadvertently motivate citizens to seek out censorship-evasion technologies, such as
virtual private networks (VPNs), which then allow them to discover other censored content.
These studies make the case that the existence of some media freedom in authoritarian
regimes does not necessarily reflect an inability to control the media more thoroughly.
Rather, some governments choose to exercise less interference with the media than they are
capable of, because they derive a political benefit from allowing some media freedom.
Nonetheless, some factors do make the cost of manipulating the media higher for some
governments, providing a financial incentive for some regimes to tolerate a degree of media
freedom. Guriev and Treisman (2019) argue that when there are many elites who are
informed about an autocrat’s competence and could communicate their knowledge to the
public, it can become too costly for the ruler to either block the elites’ media or co-opt
widespread, which would dissuade them from protesting.
23
them into reporting favorably. Besley and Prat (2006) similarly argue that governments are
less likely to try to bribe the media into covering up bad news the more media outlets there
are.
Some studies have tested some of these theories empirically, using data from countries around
the world, and found empirical regularities regarding which authoritarian governments allow
greater or lesser media freedom. Using an index of media freedom produced by Freedom
House, Egorov, Guriev, and Sonin (2009) find that oil rich dictatorships allow less media
freedom than those without significant oil wealth.16 Whitten-Woodring and Van Belle (2014)
find a number of empirical patterns based on an original dataset on media freedom. In nondemocratic countries, international and civil wars are both negatively correlated with media
freedom, as are protests and strikes, while constraints on the executive from the legislature,
judiciary, or military are associated with greater media freedom.17 In addition to these
findings, the authors also discover that economic development and literacy are not correlated
with the extent of media freedom in non-democracies.
My study adds to this literature on media freedom in authoritarian regimes by examining a
particular aspect that has been underexamined: whether or not opposition groups are allowed
to produce their own media. Not only has this phenomenon not been explicitly theorized,
but empirical studies have primarily relied on indices that measure media freedom in a very
general way, such as Freedom House’s “Freedom of the Press” Index (Freedom House 2017c)
and Whitten-Woodring and Van Belle’s (2015) Global Media Freedom Dataset. This study
contributes a new theory of why authoritarian regimes sometimes allow opposition media
and sometimes do not.
16
This finding is consistent with a hypothesis that emerges from their model: they assume that revenue
from natural resources are less affected by the provision of public goods by bureaucrats. Since allowing
investigative reporting on bureaucrats carries the risk of triggering protests, dictators with access to natural
resource rents therefore have less reason to allow investigative reporting on their subordinates.
17
Despite the finding about protests and strikes, the authors note that there are many cases in which
unrest forced a dictator to make democratizing reforms that included greater media freedom.
24
I build on previous work that has argued that audiences are more likely to believe claims
that run counter to the usual ideological leanings of a news source. Baum and Groeling
(2009), employing an experiment in the United States, demonstrate that media consumers
are more likely to believe information from a media outlet if it goes against what they
perceive to be the outlet’s partisan bias. For instance, criticism of a Republican president
is particularly compelling if it comes from a conservative news source, while praise for a
Republican president is most compelling when delivered by a liberal outlet. As noted above,
this logic is similar to that of Cowen and Sutter (1998) and Cukierman and Tommasi (1998)
about why politicians are more likely to be believed when they point to the benefits of a
policy that deviates from their ideological inclinations.
Sobolev (2019) makes a similar claim to Baum and Groeling (2009) that is particularly
relevant to this study. Sobolev argues that, in non-democratic regimes, media reports
indicating the government is popular are more credible when they come from independent
outlets rather than state-run media. Such reports from independent media can therefore be
be particularly effective at discouraging citizens from protesting. Employing data from a
natural experiment in Russia, Sobolev provides evidence that reporting by the independent
radio station Echo of Moscow on a large pro-government demonstration in 2012 was more
convincing to the public than reports of the same information from state media. In cities
that, by accidents of geography, had access to the independent station, the frequency and
size of anti-government protests fell more after the pro-government rally than in cities that
did not have access to the station. Sobolev’s findings are consistent with evidence that
citizens in authoritarian regimes can often distinguish between government-aligned and
independent media and generally place greater trust in the latter. Analyzing World Values
Survey data from countries around the world, Tsfati and Ariely (2014) find that, among
non-democracies, trust in the media declines as the share of state ownership in the TV and
25
newspaper markets increases.18
By definition, independent media are only credible if they are willing to report both favorable
and unfavorable news about the government, depending on the true state of the world. This
leaves the question of why an opposition group, which by definition would like to change
some of the regime’s policies and may even want eventually to end the regime, would send a
message discouraging citizens from protesting against the regime. Sobolev describes Echo of
Moscow as “an anti-government radio station” and does not address why the station would
be motivated to report information that would serve the regime’s interests by dampening
anti-government protests. My theory builds on this study by offering an explanation of why
a group that favors political change would nonetheless use its media outlets to discourage
its supporters from protesting. I also derive empirical implications from that model about
when an authoritarian regime will or will not allow opposition media.19
My theory also contributes to the literature that has argued that authoritarian regimes —
particularly in Arab countries — use opposition groups, paradoxically, to enhance regime
stability (Zartman 1988; Albrecht 2005; Lust-Okar 2005). To these studies I add a new
mechanism by which regimes use opposition groups to enhance their own power: they give
a voice to the opposition when their opponents will credibly convey a message that it is not
a good time to challenge the regime.
More generally, this study contributes to the literature on how authoritarian regimes
defend themselves from threats of mass unrest. As Svolik (2012) explains, dictators face
18
The results for state ownership of newspapers obtains only among the most authoritarian countries.
Unlike Sobolev,(2019), this study treats the decision to allow oppositional media as the phenomenon
to be explained, rather than taking it as given. Sobolev acknowledges that this decision is endogenous and
that the government can benefit from an independent media outlet’s reporting if the facts it reports make
dissidents less inclined to protest. Thus, he conjectures that more popular regimes will be more inclined to
allow independent media. However, he does not model this decision or test the hypothesis empirically.
19
26
threats to their power from forces both inside and outside their regimes.20 Historically, the
most frequent way that authoritarian regime leaders have lost power has been through
internal challenges, namely coups. However, mass protest movements by ordinary citizens
have become an increasingly common way for dictators to fall in the post-Cold War era
(Stein 2016). Kendall-Taylor and Frantz (2014: 39) report that almost half of dictators
who lost power in the 1960s and 1970s were overthrown by coups, but that share fell to less
than 10 percent in the decade to 2014. Meanwhile, the share of authoritarian rulers who
lost power in mass revolts, initially less than 5 percent, more than doubled. The authors
(pages 39-40) attribute this shift in the balance of threats dictators face to a few factors:
American and European support for coups has declined since the end of the Cold-War,
“hybrid” authoritarian regimes that allow greater civil liberties have proliferated, and the
emergence of social media has facilitated collective action.
Manipulating the information citizens receive is just one of many implements in dictators’
toolbox for insulating themselves from the threat of mass unrest. They can also jail and
kill dissidents (both opposition leaders and their followers) to disrupt and deter collective
action, they can make policy concessions or redistribute resources to increase their popularity,
and they can co-opt elites who might otherwise mobilize protests against them (Acemoglu
and Robinson 2006; Gerschewski 2013). Implementing pseudo-democratic institutions and
policies has become a very common way for authoritarian regimes to contain popular threats
to their power in the last few decades, including establishing legislatures, holding multiparty
elections, and allowing a degree of media pluralism (Brancati 2014). Brancati (2014: 313)
argues that pseudo-democratic institutions enhance the stability of dictatorships in several
ways: “signaling, information acquisition, patronage distribution, monitoring, and credible
commitment.” In particular, my theory highlights how dictators can selectively allow media
20
Svolik (2012: 2) refers to dictators’ task of defending themselves against internal threats as the “problem
of authoritarian power-sharing,” while he refers to the challenge of defending themselves against external
threats as the “problem of authoritarian control.”
27
pluralism as a way to signal their strength.
Finally, this study contributes to the literature on the many ways that opposition leaders
influence anti-government protest movements. Bueno de Mesquita (2010: 447-449) lists
some of the mechanisms studies have highlighted by which opposition leaders can mobilize
collective action, including making collective action focal and providing selective incentives to
participants. Nonetheless, this study draws on the important insight that opposition leaders
do not always want to mobilize protests against authoritarian governments, because some
junctures are riskier than others (Tarrow 1994; Ginkel and Smith 1999; Stein 2008). Under
some circumstances, opposition leaders actively discourage their supporters from engaging
in collective action against the government (Lust-Okar 2005).
2.3
Model
In this section, I present a “cheap-talk” signaling model in which the presence or absence of
opposition media is an endogenous outcome. There are two senders: a Regime Leader and an
Opposition Leader. There is one receiver: a group of citizens who support the opposition,
referred to as the Supporters, and treated as a unitary actor.21 The Supporters do not
represent the entirety of the citizenry, but the subset of citizens who would prefer a regime
led by the opposition and who would consume opposition media if it were produced. In
other words, they are those citizens who are quite dissatisfied with the status quo and have
the motivation to seek out alternative media. Other citizens and the security forces do not
appear as players in the model, but considerations about how they would act are implicitly
21
I assume a single opposition group for simplicity. In practice, a government often faces multiple
opposition groups, often across a diverse ideological spectrum, and may tolerate some groups while repressing
others (Lust-Okar 2005). And as Lust-Okar argues, the strategic interactions among opposition groups can
be crucial to understanding why the government treats opposition groups the way it does. I will address
these important issues in the concluding chapter when I discuss possible directions for future research.
28
captured by the state variable S, which I turn to now.
The Regime Leader and the Opposition Leader, but not the Supporters, observe the
probability that protests would succeed in bringing about a regime transition, S, which is
drawn from (0, 1) with a CDF FS [S] and a PDF fS [S]. S reflects factors such as how
widespread grievances are — and thus how likely it is that other citizens would join in if
the Supporters protested — and how likely the security forces are to use violence against a
given number of protestors. The assumption that the Regime Leader and the Opposition
Leader observe S is consistent with the empirical reality that dictators and opposition
elites are sometimes surprised by the outcomes of protests. A protest can succeed even
when S is low or fail when S is high. Also, while the leaders are strictly better informed
about the probability of protest success than the Supporters in this model, this assumption
is made only to simplify the analysis of the model. The logic of the theory only requires
that elites have access to some sources of information about the probability of successful
protest that citizens lack, not that they necessarily have more information than citizens
overall. Even well-informed citizens can make their assessments about the likelihood of
protest success even more accurate if they are able to factor in additional information from
the elites’ signals.
The Regime Leader and the Opposition Leader, but again not the Supporters, also observe
whether the opposition has sufficient resources to produce a media outlet, R = 1, or lacks
the resources, R = 0. The probability that R = 1 is ρ ∈ (0, 1).22
If R = 1, the Opposition Leader chooses a message m ∈ {0, 1}, which the Regime Leader
sees but the Supporters do not. m = 1 represents a message encouraging the Supporters to
22
The assumption that R and S are drawn independently is made only for simplicity. In practice, the
probability that protests would succeed is likely correlated with opposition groups’ resources. A betterfinanced group may organize protests more effectively. And more widespread grievances impact not only
citizens’ willingness to join protests but also their willingness to donate resources to opposition groups.
29
protest. In substantive terms, this might be an explicit call to protest or it might take the
form of calling for an immediate regime transition. m = 0 represents a message encouraging
the Supporters not to protest. Such a message might take the form of explicitly calling on
opposition supporters not to protest, or simply refraining from calling them to protest, and
advocating incremental policy reforms only.
The Regime Leader then chooses whether to allow the opposition to have a media outlet,
M = 1, or not, M = 0. If R = 0, then M = 0 by default.23
The Supporters observe M . If and only if M = 1, the Supporters observe m.
The Supporters observe their cost of protesting, C, which is drawn from a distribution on
(−∞, ∞), with CDF FC (C) and PDF fC (C).24
The Supporters choose whether to protest, P = 1, or not to protest, P = 0. If they protest,
there is a regime transition, T = 1, with probability S. With probability 1 − S, the security
forces crush the protest violently and the regime remains in place, T = 0. If the Supporters
do not protest, the regime remains in place with certainty. As noted previously, protest
success and failure can take many forms other than a regime transition or violent mass
23
The timing of the Regime Leader’s choice to allow or disallow opposition media after the Opposition
Leader’s choice of a message reflects the idea that authoritarian regimes typically have the capacity to
censor a media outlet quickly if it publishes content the regime doesn’t want citizens to see. For instance,
the government might stop the distribution of an issue of a newspaper with an article encouraging protest,
or block access to a website after a post it considers too critical. Reversing the timing of these two moves
would not affect the outcome of the model, because the Regime Leader and the Opposition Leader have the
same information, so the former could perfectly predict the signal the latter would send. In an alternative
version of the model, in which the Regime Leader is uncertain about the Opposition Leader’s preferences or
the two players see different signals about the probability that protest would succeed, the timing of these
moves could affect the results. In such a model, it might make sense to assume that the Regime Leader first
chooses whether to allow the Opposition Leader to operate a media outlet, the Opposition Leader chooses
a message to publish, and then the Regime Leader chooses whether to try to censor it, and with some
probability censorship fails and the Supporters see the message.
24
C should be thought of as the total direct and opportunity costs of protesting minus any benefits that
come from protesting — material, social, or psychological — regardless of the outcome of the protest. A
negative value of C would denote that the inherent benefits of protesting outweigh the costs.
30
repression. I assume a binary set of outcomes only to simplify the analysis of the model.
The theory itself applies regardless of whether the success or failure of a protest is likely to
be dramatic or more limited.
The Regime Leader gets a payoff normalized to 0 if the regime survives and -1 if the regime
is overthrown. That is, the Regime Leader’s utility is simply −T .
The Opposition Leader gets a payoff normalized to 0 if there is no protest. She gets a payoff
of 1 if there is a regime transition. She pays a cost πOL > 0 if the Supporters protest but
fail to overthrow the regime. Thus, the Opposition Leader’s utility is T − πOL P (1 − T ).
The Supporters also get a payoff normalized to 0 if they do not protest and a payoff of 1
if they succeed in overthrowing the regime. They pay a cost C if they protest, regardless
of the outcome. They pay an additional cost πsup > 0 if they protest but fail to oust the
regime. Thus the Supporters’ utility is T − P πsup (1 − T ) + C).
2.4
Equilibria
In this section, I derive the game’s perfect Bayesian equilibria. Because the game has two
senders, whose actions jointly determine the information that the receiver sees, if the receiver
observes an out-of-equilibrium signal, it may be indeterminate which sender deviated, or
whether both senders deviated. I propose as a refinement to employ a principle of least
deviation. The basic idea is that deviations from equilibrium strategies are improbable, and
deviations by more players are less probable than deviations by fewer players. In formal
terms, if an out-of-equilibrium signal (or set of signals) could have been produced either by
31
deviations by a set of senders A or by deviations by a set of senders B, and B is a proper
subset of A, then the receiver does not believe that the signal was caused by deviations by
all the senders in A. In this model specifically, if an out-of-equilibrium signal could have
resulted from either deviations by both senders or from a deviation by one sender only, the
receiver believes that only a single sender deviated.
I will focus primarily on an equilibrium in which the Opposition Leader’s signal m conveys
information about S. In this equilibrium, the Opposition Leader chooses m = 1 if and only
if S is greater than a threshold S ⋆ ∈ (0, 1). The Regime Leader allows opposition media
(M = 1) if and only if m = 0. Define k ∈ {0, 1, ∅} as an index of the set of signals the
Supporters can observe. k = 1 denotes that the Supporters observe M = 1 and m = 1.
k = 0 denotes that they observe M = 1 and m = 0. k = ∅ denotes that they observe
M = 0 and thus do not see m. The Supporters protest (P = 1) if and only if C < Ck⋆ ,
where C0⋆ < C∅⋆ < C1⋆ .
First, let us derive the Supporters’ beliefs given the set of signals they observe. Define
f S [S|k] as the posterior density of S given the set of signals k. If they observe M = 1 and
m = 0, they infer that S ≤ S ⋆ . Thus, the posterior density of S is:
f S [S|0] =
fS [S]
F [S ⋆ ]
(2.1)
for S ≤ S ⋆ and 0 for S > S ⋆ . If they observe M = 1 and m = 1, which is off the
equilibrium path, they could believe either that the Opposition Leader and the Regime
Leader both deviated or that the Opposition Leader played her equilibrium strategy but the
Regime Leader deviated. By the principle of least deviation, the Supporters infer that the
Regime Leader deviated but not the Opposition Leader, so S > S ⋆ . Thus, the posterior
32
density of S is:
f S [S|1] =
fS [S]
1 − F [S ⋆ ]
(2.2)
for S > S ⋆ and 0 for S ≤ S ⋆ . Finally, if they observe M = 0, the posterior density of S is:
f S [S|∅] =
fS [S](1 − ρ)
1 − ρF [S ⋆ ]
(2.3)
f S [S|∅] =
fS [S]
1 − ρF [S ⋆ ]
(2.4)
for S ≤ S ⋆ and:
for S > S ⋆ .
Next, I will derive solutions for the thresholds Ck⋆ . The expected payoff of protesting given
signals k is:
S k − πsup (1 − S k ) − C
(2.5)
where S k is the expected probability of a successful protest given signals k. Specifically:
33
Sk =
Z
1
Sf S [S|k]dS
(2.6)
0
The payoff to not protesting is 0, so at the thresholds Ck⋆ , the expected payoff of protesting
is exactly 0:
Ck⋆ = −πsup + (1 + πsup )S k
Z 1
Sf S [S|k]dS
= −πsup + (1 + πsup )
(2.7)
(2.8)
0
Since Ck⋆ is increasing in S k , and it can be shown (see Appendix 1) that S 0 < S ∅ < S 1 , it
follows that C0⋆ < C∅⋆ < C1⋆ .
If the Supporters play such a threshold strategy, the Regime Leader will play the strategy
of choosing M = 1 if and only if m = 0. If the Regime Leader chooses M = 0, he gets
⋆
an expected payoff of −SF [C∅⋆ ], while if he chooses M = 1, he gets a payoff of −SF [Cm
].
Regardless of S: −SF [C0⋆ ] > −SF [C∅⋆ ] > −SF [C1⋆ ]. So if m = 1, the Regime Leader
chooses M = 0, getting a payoff of −SF [C∅⋆ ] rather than −SF [C1⋆ ]. If m = 0, the Regime
Leader chooses M = 1, getting a payoff of −SF [C0⋆ ] rather than −SF [C∅⋆ ].25
Finally, given the strategies of the Regime Leader and the Supporters, we can derive the
Opposition Leader’s threshold strategy. Given S and the set of signals k, the Opposition
Leader’s expected payoff is:
25
Note that the Regime Leader only has a move when R = 1.
34
F [Ck⋆ ] S − πOL (1 − S) = F [Ck⋆ ] S(πOL + 1) − πOL
(2.9)
Thus, the Opposition Leader prefers the Supporters to base their protest decision on a higher
threshold if and only if:
S > S⋆ =
πOL
πOL + 1
(2.10)
If the Opposition Leader chooses m = 1, the Supporters will see M = 0 (k = ∅), and the
probability that they protest will be F [C∅⋆ ]. If she chooses m = 0, the Supporters will see
M = 1 and m = 0 (k = 0), and the probability that they protest will be F [C0⋆ ]. Thus, if
and only if S > S ⋆ , the Activist chooses m = 1.
In addition to this equilibrium, there is also an equilibrium in which every player’s strategy
is reversed with respect to the signal m. In that equilibrium, if S is higher than S ⋆ , the
Opposition Leader sends m = 0 and otherwise sends m = 1. The Regime Leader chooses
M = 1 if and only if the Opposition Leader sends m = 1. And values of the thresholds C0⋆
and C1⋆ are reversed. However, this equilibrium does not differ substantively from the one
already discussed: m = 1 and m = 0 simply have opposite meanings.
There is also a set of babbling equilibria, in which the probability that the Opposition Leader
chooses m = 1 does not depend on S. In such equilibria, the Regime Leader’s choice of M
does not depend on m or S. Lastly, the Activist’s decision to protest is based on a single
threshold value that does not depend on M or m. This equilibrium is not particularly
plausible, however, since the equilibrium with communication that I have focused on always
35
exists.
2.5
Empirical Implications
Focusing on the communicative equilibrium, we can derive testable empirical predictions in
two ways: by looking at the range of outcomes that occur within the game while holding the
parameters fixed and looking at the effects of change in the equilibrium as the parameters
vary. The higher the likelihood that a protest would succeed (S), the lower the chance that
the opposition will be willing to send a demobilizing message and thus the lower the chance
that the regime will allow the opposition to have its own media. Thus, when citizens are
likely to have higher than usual grievances, the model predicts that a regime will be less
likely to allow opposition media, because the probability that protests would succeed (S) is
higher. The model also suggests that a regime will be more likely to allow opposition media
if there is a greater chance that security forces would crack down violently on protestors: the
greater the willingness and ability of the security forces to repress protestors, the less likely
protest is to topple the regime.
Furthermore, other factors that affect the likelihood of protest succeeding, but about which
elites would not have an informational advantage, would also affect the probability of the
government tolerating opposition media. Suppose higher values of a factor X made it more
likely that a protest would succeed in overthrowing a regime, at a given level of popular
grievances and repressive capacity, and that X was known by both elites and ordinary citizens.
For instance, a successful uprising against an authoritarian regime in a nearby country
could lead citizens and elites to revise up their estimates of the probability that collective
action could bring down their own regime. Such a factor would shift the distribution of the
36
probability of successful protest (FS ) to the right, making it more likely that the opposition
would be unwilling to send a demobilizing message (i.e. more likely that S > S ⋆ ). Thus, X
would be negatively correlated with a regime allowing opposition media.
Turning to comparative statics, the model predicts that, all else equal, opposition media is
more likely to be allowed when the expected cost to the opposition elite of a failed protest
is higher. The higher πOL , the higher is the threshold S ⋆ , and thus the more likely the
opposition is to send a demobilizing message (m = 0), which the regime would allow to be
disseminated by opposition media (M = 1).
Identifying empirical proxies for πOL is
complicated however, because factors that would increase the probability or severity of
punishment for opposition leaders after a failed protest would also likely be correlated with
factors that increase the probability or severity of repression of ordinary protestors, and
thus would affect the the likelihood of a protest succeeding (S). However, the net effect of
such factors is unambiguous. For instance, suppose a factor Y was positively associated
with both the probability that the security forces would be willing to arrest or kill
opposition leaders and the probability that they would use force against protestors. Such a
factor would both raise πOL (and thus S ⋆ ) and lower S, increasing the chances that the
opposition would send a demobilizing message and their media would be allowed. This
finding thus reinforces the prediction that the chances of the government allowing
opposition media are greater when it can rely on the security apparatus to carry out
repression.
The model assumes the existence of an organized opposition group and citizens who believe
their interests in political change are aligned with those of the group’s leaders. Each of the
following hypotheses I now derive is conditional on the existence of at least one such a group.
I will return to this point below.
37
Grievances are likely to be more widespread when national economic performance is weaker.
As noted, various studies suggest that negative economic shocks are correlated with antigovernment protests (Hendrix, Haggard, and Magaloni 2009; Caren, Gaby, and Herrold 2017;
Geddes, Wright, and Frantz 2018). Mass unemployment and declining incomes are likely to
be salient grievances, suggesting the first two hypotheses:
Hypothesis 1:
Higher unemployment is negatively correlated with the existence of
opposition media.
Hypothesis 2: Higher per capita GDP growth is positively correlated with the existence of
opposition media.
Wantchekon (2002) argues that natural resource rents allow regimes to buy public support,
and Cotet and Tsui (2013) find that greater non-tax-based revenue is associated with higher
government spending on social services. Thus, rents from natural resources should allow a
government to reduce citizens’ grievances against it. Higher oil revenue, in particular, is
also associated with greater military spending (Cotet and Tsui 2013; Wright, Frantz, and
Geddes 2013) and a greater likelihood of violence against civilians (Bove, Platteau, and
Sekeris 2017).26 Thus, it is not surprising that non-tax revenue is associated with greater
longevity of authoritarian regimes (Smith 2004; Cotet and Tsui 2013; Wright, Frantz, and
Geddes 2013). Bueno de Mesquita and Smith (2010) also find specifically that an increase in
protest is less likely to lead to democratization among dictatorships with greater oil revenue
(as a percent of GDP). By both lowering grievances and raising the chances that protest
will be met with repression, higher natural resource rents, particularly from oil and possibly
natural gas, should increase the probability that opposition organizations — if there are any
— are allowed to have their own media:
26
On the other hand, Smith (2004) finds that, while oil wealth reduces protest, this effect is not mediated
by repression.
38
Hypothesis 3: Higher per capita revenue from extractable fuel sources is positively
correlated with the existence of opposition media.27
As discussed above, personalization of a regime, specifically personal control by the dictator
of the security apparatus, should lead to a higher probability that security forces will use
violence against protestors (Bratton and van de Walle 1997; Bellin 2012; Brownlee, Masoud,
and Reynolds 2015), lowering the probability that protest would bring down the regime.
This observation leads to the fourth hypothesis:
Hypothesis 4: The degree to which control of the security forces is concentrated in the
hands of the dictator personally is positively correlated with the existence of opposition
media.
Following from the premise that the demonstration effect of a successful revolution abroad
can lead people to revise their beliefs about the chances for a successful revolt in their own
country:
Hypothesis 5: The occurrence of successful uprisings against nearby authoritarian regime
are negatively correlated with the existence of opposition media.
I believe that elites in authoritarian regimes generally have more information about the
variables mentioned in these hypotheses than ordinary citizens. However, as noted previously,
even if there were no information gap between elites and non-elites about a particular factor
that affects the chances for successful protest, it would still affect the likelihood that the
government would tolerate opposition media in the hypothesized direction, by shifting the
distribution of S. For example, if a publicly observable factor made protest more likely to
27
I follow Wright, Frantz, and Geddes (2013) in focusing on the effects of fuel revenue per capita, rather
than as a percent of GDP, to avoid confounding the effects of fuel rents with fluctuations in overall GDP.
39
succeed, then, holding all other factors equal, opposition leaders would be more likely to
conclude, based on their private and public information, that citizens should protest. And
therefore, the government would be more likely not to allow opposition media. Successful
uprisings against nearby regimes are probably close to being such a publicly observable
factor, since international media will tend to cover such dramatic events, and many citizens
in authoritarian regimes would observe the coverage by satellite television channels.28
The model’s assumption that an organized opposition group exists and a non-negligible
group of citizens believe the group shares their interests is an important scope condition. In
practice, the existence of organized opposition groups with significant number supporters is
not independent of the extent of grievances or the state’s repressive capacity. Regimes with
extremely low public discontent or extremely high repressive capacity may have no organized
opposition groups at all. Or they may have only very small opposition groups, which, as I
have argued, are beyond the scope of this theory. If we do not condition on the existence of
significant opposition groups, we might therefore find an inverted-U relationship between the
likelihood that protests could topple the regime — as proxied by economic performance, the
dictator’s personal control of the security apparatus, etc — and the likelihood that opposition
media exist. However, once we condition on the existence of significant organized opposition
groups — the set of cases to which the theory applies — we should find a strictly negative
relationship between the likelihood of protest succeeding and the likelihood that opposition
media will exist and be tolerated.29
In the next two chapters, I explore these hypotheses empirically. Chapter 3 illustrates the
28
The news would also be likely to proliferate to social media accounts that governments can’t selectively
block and to too many websites for the government to identify and block them all.
29
Defining the line between negligible and non-negligible movements is obviously fraught with subjectivity
and ambiguity. In the qualitative case study and the quantitative cross-national comparison, I base my
decisions to focus on some opposition groups and not others on indicators of their support such as the
numbers of followers they have been able to mobilize to protest or vote in the past. Identifying alternative
measures of opposition group strength is an important direction for future research, in order to test further
the robustness of my findings.
40
theory qualitatively, through a case study of Tunisia during the first half decade of Ben Ali’s
rule. Chapter 4 tests the theory quantitatively, through statistical analysis of data from ten
Arab authoritarian regimes since the end of the Cold War.
41
Chapter 3
“A Weird Period”: Opposition Media
in Tunisia, 1987-1992
When Zine el-Abidine Ben Ali seized power in Tunisia in November 1987, many observers
inside and outside the country believed he would be a political reformer. One of the reforms
he promised in the early morning hours after his coup was greater freedom of the press. And
indeed, during the first three years of his rule, he allowed many opposition groups to launch
new and previously banned newspapers, including the most popular opposition group, Hizb
en-Nahda. Yet within four years of the coup, Ben Ali had reversed his own press reforms.
He shut down Nahda’s newspaper in early 1991 (shortly before dismantling the group itself).
By 1998, Ben Ali had made it onto a list of the world’s top ten “Enemies of the Press,”
published by the Committee to Protect Journalists. Why did Ben Ali allow this opening
for the opposition press and why did he subsequently revert to suppressing the media of the
largest opposition group?
42
The early years of Ben Ali’s presidency were, as his Minister of Information put it at the
time, “a weird period” in Tunisian politics (Wright 1990a), marked by dramatic changes. In
this chapter, I seek to explain the variation in Ben Ali’s treatment of the opposition press,
over time as well as across groups, during the first half decade of his rule, from 1987 to 1992.
During this period, there was a great deal of variation in the key explanatory variables that
my theory suggests are important, including the performance of the national economy and
the distribution of power within the regime. I will show that the timing of changes in these
variables can explain the variation in which opposition groups were or were not allowed to
have their own media and when, in line with my hypotheses. For this case study I rely
largely on news reports from the 1987-1992 period, reports by human rights organizations,
and secondary sources. I also draw on interviews I carried out in Tunisia in 2016, with
journalists, human rights activists, former officials from Ben Ali’s government, and leaders
and members of the groups that opposed that government. Tunisia is particularly fruitful
setting for field research on authoritarian politics, because it combines a recent history under
authoritarian rule with a currently open and democratic polity. Thus, when I conducted my
fieldwork, I was able to meet with many individuals who had firsthand knowledge of events
under the former authoritarian regime, and they had the safety to discuss their experiences
frankly.
The theory suggests that dictators should be more likely to allow small, less popular
opposition groups to have media than larger, more popular groups, since the latter are
more likely to encourage their followers to protest. Across groups, Ben Ali was consistently
more willing to allow the weaker secular parties to have their own newspapers compared to
Nahda, which had greater potential to mobilize protests against his government.
The theory also suggests that dictators will be least likely to allow opposition groups to have
their own media when grievances are widespread and their ability to repress protests is low.
43
This implication helps us make sense of the Tunisian case over time. Ben Ali’s initial decision
not to allow Nahda to have any media during his first two years in power, his subsequent
decision to allow them to publish a newspaper in 1990, and finally his decision to shut
down that same newspaper in early 1991, all correspond to shifts in factors that indicated
the likelihood that Nahda could succeed in toppling Ben Ali’s government by mobilizing
protests. Similarly, these shifts help us understand why some secular opposition newspapers
were briefly shut down around the same time as Nahda’s but, unlike the latter, were allowed
to reemerge soon after.
Hypothesis 3 (Higher per capita revenue from extractable fuel sources is positively correlated
with the existence of opposition media) is not particularly relevant to the case. Tunisia did
not exhibit large variation in its modest per capita revenue from oil production during 19871992, and its natural gas production was negligible (BP 2019). Nor does Hypothesis 5
(The occurrence of successful uprisings against nearby authoritarian regime are negatively
correlated with the existence of opposition media) directly apply, since no Arab authoritarian
regime was overthrown in a mass uprising during this period. However, events in neighboring
Algeria at this time do support the more general claim that the strength of opposition groups
in other regimes can lead elites to update their beliefs about the prospects of opposition
groups in their own country.
Tunisia during this period did exhibit substantial variation in unemployment, economic
growth, and personalization of the regime, making the other three hypotheses relevant.
The case turns out to be consistent with each of them. Namely, we can better understand
this episode in light of Hypothesis 1 (Higher unemployment is negatively correlated with
the existence of opposition media), Hypothesis 2 (Higher per capita GDP growth is
positively correlated with the existence of opposition media), and Hypothesis 4 (The degree
to which control of the security forces is concentrated in the hands of the dictator
44
personally is positively correlated with the existence of opposition media).
In the next section I will provide some background information about the regime that Ben
Ali inherited when he came to power in 1987, about Ben Ali himself, and about the key
opposition groups in Tunisia at the time. In the succeeding section, I will describe in more
detail what is to be explained: the variation in the government’s toleration of opposition
media across groups and over time. In the section after that, I will argue that Ben Ali
allowed secular opposition groups to have media soon after he came to power but not Nahda
because of the greater mobilization threat that Nahda posed. Next, I will show that Ben
Ali’s decision to allow Nahda to have a newspaper in 1990, after initially forbidding it, is
explained by shifts in the economy and intra-regime politics that appeared to strengthen the
dictator. Following that, I will explain how events later in 1990 led Ben Ali and Nahda to
revaluate their relative strength and motivated Ben Ali to shut down Nahda’s newspaper.
Finally, I will argue that the changing relationship between the secular opposition groups
and Nahda explains variation over time in Ben Ali’s treatment of their media as well.
3.1
Background
Between independence from the French in 1956, and the “Jasmine Revolution” of 2011,
only two individuals ruled Tunisia: Habib Bourguiba and Zine el-Abidine Ben Ali. Having
led the fight for independence, Bourguiba ruled the country for three decades as the head
of the Destourian Socialist Party (Parti Socialiste Destourien or PSD), which monopolized
power over the state. During Bourguiba’s rule, Ben Ali rose through the military intelligence
service and the security apparatus to the post of interior minister. Just a month before Ben
Ali replaced Bourguiba in a bloodless coup, Bourguiba appointed Ben Ali as prime minister
45
(Erdle 2010: 196).
Tunisia had seen significant unrest since the late 1970s. Amid economic stagnation and the
Bourguiba government’s austerity measures, massive protests had broken out in 1978 and
1984. The former involved tens of thousands of participants, the latter half a million (Seddon
1986; Global Security 2020). Tunisia’s 1986 agreement with the IMF and the World Bank
to move further toward a market-based economy promised continued social tension (Murphy
1999: 104).
There were several small secular opposition parties in Tunisia when Ben Ali came to power,
the most prominent of which was the Movement of Socialist Democrats (Mouvement des
Démocrates Socialistes or MDS) (Bellin 1995: 131). There was also an opposition group
known as the Islamic Tendency Movement (Mouvement de la Tendance Islamique or MTI),
which would later rename itself the Party of the Renaissance (Hizb Nahda, often referred to
simply as Nahda) (Waltz and Kéchichian 2009). I will refer to this group as both the MTI
and Nahda, depending on the time period under discussion. Bourguiba arrested several of
the MTI’s leaders in 1987 and was on the verge of executing several of them when Ben Ali
overthrew him. Ben Ali may have timed the coup to prevent these executions, for fear that
they could provoke an uprising (Bellin 1995: 133).
3.2
Variation in state tolerance for opposition media
across groups and time
In his first year in power, Ben Ali allowed several opposition parties to relaunch papers that
had previously been banned. Within a few months of Ben Ali’s coup, his government ended
46
the bans on the party newspapers of two legal opposition parties: Al-Moustaqbal of the MDS
and Tariq el-Jedid of the Tunisian Communist Party (Parti Communiste Tunisien or PCT).
(Africa Report 1988: 8; Murphy 1999: 168). However, the new government maintained
tight controls on what these opposition papers could write. Khadija Cherif, a human rights
activist, described the liberalization of the press during the early years under Ben Ali as “very
controlled” (discussion with the author, May 2016). According to Amnesty International
(1994: 4): “Amendments to the existing Press Code [in 1988] only very slightly curtailed
the executive’s wide powers to control freedom of expression.” For example, the reform of
the press code’s provisions on defamation did not extend to criticism of the president or his
ministers (Harris 1989: 822). Samir Taieb, a member of the Movement of Renewal party
(Harakat Ettajdid, the successor to the Tunisian Communist Party), which opposed Ben
Ali, and an editor of its journal, confirmed that, even during relatively liberal first few years
under Ben Ali, criticism of the president and his family was not tolerated (discussion with
the author, May 2016). The journalist Slaheddine Jourchi of Er-Rai (Opinion), a journal
which was aligned with the MDS, agreed with this assessment, and added that criticism of the
defense ministry was also forbidden (discussion with the author, May 2016). Ahmed Mestiri,
the leader of the MDS when Ben Ali came to power, reported that Ben Ali;s government
did not interfere with the party’s weekly, Al-Moustaqbal during his first year in power, but
began to harass it in 1989 after the first legislative elections (discussion with the author,
May 2016). Alexander (2010: 55) writes of this period that:
While the government allowed a host of new newspapers and magazines to
appear, it also began signaling limits to writers and editors. In December 1988,
for example, the government seized copies of Al-Maouqif, the newspaper of the
Progressive Socialist Rally (Rassemblement Socialiste Progressiste or RSP) after
it published an article on Tunisia’s nonlegal movements and parties. A week
later, the government seized copies of Réalités, a popular news weekly, for an
47
editorial that criticized the lack of judicial independence. These seizures marked
the beginning of an insidious new form of press control.
The government
legalized many new publications and declared its respect for freedom of the
press. It even provided funding to help some newspapers. But then security
forces would jail a journalist or confiscate copies of publications because they
wrote or published an article that did not sit well with the president. Because
the government never laid out an explicit list of topics that journalists could not
touch, they never knew exactly when they would cross the line into forbidden
territory. This uncertainty forced members of the press to err on the side of
caution when making decisions about their stories. The result was a wooden,
self-censored press even though the government appeared to be relaxing the
formal restrictions on it.
Thus, as in the theory presented in the previous chapter, even when dictatorships allow
opposition media, they ensure that the content remains tame.
Only secular opposition parties were granted the right to publish newspapers during Ben Ali’s
first two years in power, while the MTI/Nahda remained excluded. It was not until January
1990 that Ben Ali’s government finally gave Nahda permission to publish a journal, El-Fajr,
which started appearing in April (Wright 1990b). However, the government prevented the
distribution of an issue of El-Fajr in June 1990 (Middle East Economic Digest 1990b), and
a week later a court suspended the journal for three months (Chouikha, Labidi, and Jouini
1992: 101). After El-Fajr began reappearing in the fall, the government stopped distribution
of two more issues (Amnesty International 1991). In February 1991, the government shut
down El-Fajr permanently (Donnadieu 1992: 39).
1990 also saw a crackdown on the secular opposition press. The government suspended the
48
journal Al-Badil, published by the Communist Party of the Workers of Tunisia (Parti
Communiste des Ouvriers de Tunisie or PCOT), for six months in October 1990 and
permanently suspended it the following April (Chouikha, Labidi, and Jouini 1992: 102.)
The MDS stopped publishing its weekly Al-Moustaqbal in June 1990. An editorial in the
final issue stated that the paper was closing because of financial difficulties, but also
because of the “refusal to let the opposition newspapers work freely” (Chouikha, Labidi,
and Jouini 1992: 103). Other secular opposition newspapers stopped publishing around
this time, including Al-Watan, of the Unionist Democratic Union (Union Démocratique
Unioniste or UDU), in August 1990 and the RSP’s Al-Maouqif in January 1991. While
both of these closures were ostensibly due to financial problems, government interference
may have exacerbated those problems: the government stopped distribution of two issues of
each newspaper in 1990 (Legum 1992: B 520; Chouikha, Labidi, and Jouini 1992: 102-103).
By 1991, the former (and by then exiled) prime minister Mohammed Mzali observed that
in Tunisia “there no longer exists today an organ of opposition press” (Murphy 1999: 203).
In April 1991, however, the government announced that it would provide subsidies to the
legal (all secular) opposition parties to publish their newspapers (Donnadieu 1992: 39).
Beginning in November 1991, secular opposition papers began to reappear, while El-Fajr
remained suppressed (Chouikha, Labidi, and Jouini 1992: 112).
The shutdown of El-Fajr would turn out to be the prelude to the dismantling of Nahda as
an organization. Several thousand Nahda activists were arrested over the course of 1991
(Europa Publications 1994:879; Alexander 2010: 60). In July 1992, 279 Islamists were tried
in the Tunis Military Tribunal, which ended in “heavy sentences for most of Nahda’s leading
figures,” including life sentences for some (Murphy 1999: 198). That same month the Middle
East Economic Digest (1992c) reported: “Police action has crushed Nahda, whose leadership
is either exiled, in prison or has dropped out of active politics.” The article quoted a “former
49
Nahda official” as saying: “We entered into direct confrontation wanting to take over and
we lost.” Europa Publications (1994: 880) observes regarding the July 1992 trials: “These
mass trials were seen as the culmination of the Tunisian Government’s long campaign against
al-Nahdah, whose organizational structures within the country were largely destroyed and
its leaders imprisoned or forced into exile.”1
3.3
Why the difference in media policies for the
secularist and Islamist opposition?
Although Ben Ali repressed both the Islamist and secular opposition press in late 1990 and
early 1991, overall he was more willing to tolerate secular opposition than the media of the
MTI/Nahda. For the first two years he was in power, Ben Ali allowed the secular groups to
publish newspapers, but not the MTI/Nahda. In late 1991, he allowed the secular opposition
press to reemerge — even subsidizing some of it — but kept Nahda silent. This differential
treatment of the MTI/Nahda and the other, secular, opposition groups was driven by the
greater strength of the Islamists relative to the secularists. Ben Ali had reason to think the
MTI posed the greatest threat to mobilize protestors, out of all the opposition groups, since
he first came to power. The MTI had been prominent in organizing the demonstrations
of 1984 (Shahin 1997: 88; Global Security 2020), after which Ghiles (1984) wrote that the
Islamists “provide the most serious challenge to the Tunisian regime.” In the last year of
Bourguiba’s rule: “The MTI proved its strength with its continued ability to preach against
1
Similarly, Willis (2012: 171-172) writes that “the Ben Ali regime rooted out what little remained of AnNahda, making Tunisia one of the only states in the Arab world in which Islamism had no real institutional
or organizational presence.” Murphy (1999: 199) writes: “Within two years, Nahda had virtually been
demolished in Tunisia, at least in terms of organizational and operational structures. With those of its
leaders still in Tunisia serving heavy prison sentences, and Ghannouchi exiled to London, what remained of
the Islamist movement was decapitated and fragmented.”
50
the regime in the mosques and organize public demonstrations despite the hunting down and
imprisonment of its leaders” (MacDonald 1988).
Zartman (1988: 82) writes that Tunisia’s secular parties in this period were organizations
representing “professionals and intellectuals, a safety valve for dissatisfaction rather than a
mass challenge to PSD.” By contrast: “The Islamic Tendency movement is another matter,
being neither a Destourian [ruling party] offshoot nor a negligible safety valve” (ibid ). Also
writing in 1988, Boulby argued that “it is possible to hazard the speculation that the MTI
has broader appeal than any of Tunisia’s other political movements (including, of course,
the PSD!)” (Boulby 1988: 607). When Ben Ali came to power “many feared that Al-Nahda
was about to take power by force” (Sagar 2009: 589). “Ben Ali did not want an all-out war
with the Islamists in 1987. The MTI had developed a large organization ... Even Ben Ali
could not be certain of its true capabilities. In light of the general disenchantment across
the country, a war with the Islamists might produce a groundswell of support for them and
overwhelm the new government” (Alexander 2010: 52-53).
3.4
Why Ben Ali first prohibited and later allowed
Islamist media
The model presented in the previous chapter suggests that when an opposition group has
significant capacity to mobilize protestors and the regime has little capacity to repress
protests, the opposition group will refuse to publish media that discourages protest, and so
the government will not allow them to publish at all. Ben Ali’s decision to deny the MTI
permission to produce any media during his first two years in power is consistent with this
prediction.
51
As noted previously, grievances were widespread in Tunisia in 1987 because of economic
stagnation and the austerity measures the government had put it place to try to spur growth.
This discontent had manifested in major incidents of collective action against the regime
in 1978 and 1984. “By the 1980s, Tunisia’s growing economic problems were acute and
were mirrored by deepening political tensions. . . . The January 1984 riots marked a
watershed for the Bourguiba regime. They signaled the onset of a major political role for
the Islamists, as well as warning of the dangers of economic austerity” (Mostyn and Hourani
1988: 431). The economic troubles underlying Tunisians’ dissatisfaction continued during
the first two years of Ben Ali’s rule. The economy shrank in 1988 and 1989, with per capita
GDP growth of -2.2% and -0.5%, respectively (World Bank 2019). The number of registered
unemployed Tunisians peaked in 1988 at approximately 237,000. As a percentage of the
labor force, the unemployment rate was 10.9% in 1988 (Rama 1998: 70). Although Ben Ali
signed a “National Pact” in the fall of 1988 with the major opposition parties, including
a representative from the MTI, in which he promised democratic reforms, the threat of
an uprising persisted into 1989. “The National Pact failed to curb political dissent in the
country, and tensions were exacerbated by the imposition of essential economic reforms”
(Europa Publications 1994: 878).
The risk posed by these widespread grievances was compounded by Ben Ali’s weak control
over the state itself. Many regime officials appointed by Bourguiba still held considerable
power in the early years after the coup. When he seized the presidency, Ben Ali found himself
in charge of a government in which he was, as Erdle (2010: 98) writes, an “outsider.” He
had spent his career in the military and the ministry of interior, in a regime dominated by
academics and party elites. Thus, when he became president, “the only real power bases he
had were in the interior ministry and security apparatus, but not in the ruling party, or in
the government bureaucracy as such” (ibid ). Moreover, Tunisia’s military had historically
played a very minimal role in politics (Barany 2011; Bellin 2012). As Alexander (1997: 37)
52
elaborates:
Many longtime party barons resented Ben Ali for preempting their own plans
for stepping into the presidency. From the beginning of his rule, Ben Ali feared
that one of these established politicians, or one of his own ministers, would use
their networks in the party, the state bureaucracy, and other organizations to
undermine him. As a relative newcomer to ruling party politics, Ben Ali lacked
the social bases and patronage networks so vital to Bourguiba’s style of political
management. He did not have the political resources to referee and manipulate
effectively an ongoing competition between powerful politicians and the social
actors they rallied to their camps.
Three months after the coup, a Washington Post foreign correspondent portrayed a leader
not fully in control of his regime: “Dominated for 30 years by the force of one man and the
political institutions he created in his image, Tunisia is now adjusting to a government headed
by a half-dozen strong personalities who are debating with each other over the directions
this North African nation should take” (Hoagland 1988).
Ben Ali’s weaknesses as a politician only made the challenge of asserting power over the
government and society more difficult. As the journalist Manoubi Marrouki put it, Ben
Ali was not “a political man”: he lacked Bourguiba’s skills for mobilizing the masses. He
entered office with a “deficit of popularity” according to Rachid Khechana, an opposition
journalist during the Ben Ali era. Even a high-level official who worked for both Bourguiba
and Ben Ali implied that Ben Ali lacked Bourguiba’s charisma (discussions with the author,
May-June 2016).2
2
When asked to contrast the two presidents’ leadership styles, the former official described Bourguiba’s
charisma at length, and then declined to comment on that of Ben Ali.
53
Thus, although Ben Ali’s background was in the security services, it was doubtful that the
leaders of the security services initially thought their privileges depended on keeping Ben
Ali in office, since he had to share power with many elites in the government appointed by
Bourguiba. If faced by mass protests, they might have refused to use violence to suppress
them. In addition, Ben Ali’s initially weak control of the ruling party meant its mobilization
capacity was not entirely in his hands. If, for instance, protests sparked a coup, party leaders
might have refused to mobilize members of the party in demonstrations to stop the coup.3
Moreover, the party might even have mobilized its members to join protests against Ben Ali.
Ben Ali was in such a weak political position when he first became president that Mestiri
(leader of the MDS at the time) said “We thought he had no card to play but democracy”
(discussion with the author, May 2016).
The relative strength of Nahda to mobilize protests and Ben Ali’s doubtful ability to repress
such protests explains why Ben Ali did not initially allow Nahda to have a newspaper. The
fact that he ultimately did allow them to publish a newspaper in 1990 can be explained
by the strengthening of his position over the intervening years. First of all, by 1990, he no
longer had much reason to think that Tunisians were inclined to rise up against him. “By
1990 the regime in Tunisia was benefiting from a degree of economic stability” (Murphy
1999: 201). Whereas per capita GDP growth had been negative in the previous two years,
it grew strongly in 1990, at 5.6% (World Bank 2019). The number of registered unemployed
individuals fell from 237,000 in 1988 to 152,000 in 1990. As a percentage of the labor force,
the unemployment rate fell from 10.9% in 1988 to 6.7% in 1990 (Rama 1998).
In addition, by 1990, Ben Ali had also solidified his control of the government and the ruling
party, replacing most of the political elite with his own allies. His attempts to supplant
Bourguiba’s allies in government with his own began on the very day of the coup, when
3
Geddes (2006) argues that ruling parties’ ability to mobilize protests against a military coup is an
important reason that many dictators create parties after they come to power.
54
Ben Ali appointed Hedi Baccouche, “a long-time ally,” as prime minister (Clayton 1987).
Frequent cabinet shuffles replaced “PSD barons” with Ben Ali loyalists (Erdle 2010: 100).
Military and security officials were brought into high positions in the government (Krieger
2003). Ben Ali also brought in “a younger generation of technocrats that answered to him”
(Perkins 2016a: 196). The frequent cabinet shuffles and Ben Ali’s close involvement in
the running of each ministry also helped him ensure that none of his ministers developed
independent power bases (Alexander 1997).
While Ben Ali took some immediate steps to consolidate power, the entire process was
gradual. On the eve of Ben Ali’s coup, there were twenty-one ministers in Bourguiba’s
cabinet, besides Ben Ali himself.
Only four were removed immediately.
Nevertheless,
through relentless reshuffling, Ben Ali ultimately excluded all but three of Bourguiba’s
ministers from his cabinet during the first three years of his presidency (Reuters 1987a;
Legum 1988: B 551-B 553; Middle East Economic Digest 1988; Reuters 1989b; Middle East
Economic Digest 1990c; Murphy 1999:194).
He took a crucial step in controlling the
security apparatus in early 1990, when he placed “key supporters” in “the most sensitive
security posts:
Abdelhamid Escheikh was appointed Minister of the Interior, while
Abdallah Kallel took office as Minister of Defence” (Murphy 1999:194). Not only did Ben
Ali manage to stack his cabinet with loyalists, he also managed to reduce its power over
time and take policy into his own hands.
By early 1991, Africa Confidential (1991b)
reported: “The cabinet as a whole is declining in influence as the presidency takes on new
powers. ... The [presidential] Palace increasingly houses a shadow government” of advisors.
Ben Ali was also surprisingly adept at tightening his grip on the ruling party, which he
quickly renamed the Democratic Constitutional Rally (Rassemblement Constitutionnel
Démocratique or RCD) (Dick 1988).
Within two weeks of the coup, he purged three
members of the party’s political bureau that had been particularly close to Bourguiba
55
(Reuters 1987b; Reuters 1987c). A month after the coup, Ben Ali reduced the political
bureau from twenty to twelve members, purging Bourguiba’s son among several other
individuals, leaving only three Bourguiba appointees. At the same time, he added several
new members, many of whom had backgrounds in the military or the Interior ministry,
including Habib Ammar, a former classmate of Ben Ali from the St. Cyr military academy
in France (Reuters 1987c; Keesing’s Record of World Events 1988; Africa Confidential
1988: 7). In July 1988, Ben Ali reduced the bureau to just six members, with Baccouche as
the sole remaining Bourguiba appointee. For good measure, Ben Ali also added another
classmate from St. Cyr to the bureau, former army chief-of-staff Abdelhamid Becheikh
(Reuters 1988). At a February 1988 meeting, the RCD central committee confirmed Ben
Ali as leader of the party (chairman), and his position was formally ratified at the party
congress in July of that year (Erdle 2010: 99-100). He also removed the Director of the
RCD Political Bureau from the cabinet that month (Europa Publications 1994: 878). That
summer, Ben Ali also appointed an Interior ministry veteran, Abderrahim Zouari, as
second-in-command (secretary-general) of the party (Africa Confidential 1988: 7; Erdle
2010: 100). The congress also selected a new central committee of the party, with Ben Ali
personally appointing 122 of the 200 members (Erdle 2010: 100). Several former Bourguiba
ministers were removed from the central committee, including Rachid Sfar, Ben Ali’s
predecessor as prime minister (Dick 1988). Nonetheless, Ben Ali’s purge of the central
committee was not yet complete, as “many other personalities linked with the Bourguiba
era remained in the committee” (ibid ). In 1989, the RCD renominated only 20 of the 125
incumbents in the National Assembly to run for reelection (Perkins 2016b: 218-220.). At
the level of mass mobilization, Ben Ali expanded RCD membership from 900,000 to 1.5
million between 1987 and 1989, creating a large cohort who owed their new opportunities
to him (Erdle 2010: 100)
Given the improved economy and his consolidation of personal power within the regime, Ben
56
Ali faced less risk of protest by 1990 and and could count on the security forces to repress any
that occurred. Thus, Ben Ali allowed Nahda to start publishing El-Fajr in 1990. Under these
circumstances, the model suggests that an opposition group will want to discourage protest
because it is too risky. It will agree to publish only mild criticism and so the government will
allow it to have its own media. On the eve of El-Fajr ’s first issue, its editor, Hamadi Jebali
sought to convey just how moderate the journal would be, telling a reporter from Reuters:
“I think it’s of a really high standard and not at all inflammatory” (Wright 1990b).
One might argue that the MTI’s behavior in 1988 and 1989 does not reflect the theory’s
prediction that, when an opposition group is strong enough that the government does not
allow it to have media, the group would prefer their supporters to protest against the regime.
The MTI initially displayed a cooperative attitude toward Ben Ali’s new government. In
April 1988, the jailed leaders of the MTI announced their support for the new government
in the newspaper As-Sabah (Garon 2003: 30). In August of that year, the MTI’s leader,
“Rached Ghannouchi had announced that the MTI accepted the law on political parties
and respected the Constitution” (Europa Publications 1994: 878).4 Even after the April
1989 legislative elections, which were manipulated in favor of the ruling party (Ghiles 1989;
Guardian 1989; Wright 1989), Nahda “did not take to the streets to protest the election
results” (Murphy 1999: 193). However, given the relative strength of Nahda, Ben Ali had
reason to suspect in the first two years that these conciliatory public statements by the MTI
were a ruse, and that, if allowed to produce a newspaper, they would have used it to incite
their followers to overthrow the regime.
Another possible anomaly for the theory is that members of the General Union of Tunisian
4
Shahin (1997: 100) observes: “Upon Ben Ali’s coming to power and the declaration of his new policies
on November 7 [1987], the MTI declared its full support for the removal of Bourguiba and its willingness to
assist the new president in achieving the objectives included in the November 7 statement.” Willis (2012:
166) writes: “Ben Ali’s constitutional coup was received positively by the MTI leadership, who believed that
the prime minister had not only acted to avoid a showdown with the movement but also wished to [develop]
a less confrontational relationship with it.”
57
Students (Union Générale des Etudiants Tunisiens or UGET), a group aligned with Nahda,
engaged in protests against the government in January and February 1990 (Wright 1990a;
Europa Publications 1994: 879). And yet, since Nahda had grown weaker relative to Ben Ali
by 1990, my theory suggests that they should not have been inclined to protest at this point.
However, it is not clear that these protests were supported by the Nahda leadership. In any
case, the first few months of 1990 may have been a transitional period in which the balance
of power between the regime and Nahda was somewhat uncertain. Indeed, although Ben Ali
granted legal authorization for Nahda’s journal, El-Fajr, in January 1990, the journal only
appeared in April of that year, as the group initially could not find a printing house willing
to publish it. Nahda’s leaders accused Ben Ali of discouraging the printers from working
with them (Reuters 1990g). If this accusation is true, it would indeed suggest that Ben Ali
was vacillating due to uncertainty about just how weak Nahda had become. The capstone
in Ben Ali’s consolidation of control of the security forces was set in March 1990, when
he placed Abdelhamid Escheikh, “a close ally,” at the head of the Ministry of the Interior
(Middle East Economic Digest 1990c), with Escheikh’s “military background and frequent
anti-Islamist statements suggesting a no-nonsense approach to Islamist agitation” (Legum
1992: B 519). Subsequently there was a hiatus in protests linked to Nahda and in April,
El-Fajr was able to start appearing on the streets.
3.5
Why Ben Ali cracked down on Islamist media
By early 1990 Ben Ali appeared to be in a strong enough position relative to Nahda to allow
them to have a newspaper. However, events later that year seem to have caused him to revise
this assessment. First, elections in neighboring Algeria suggested that Islamist movements
in the region could gain wide popularity and quickly. Algeria had been undergoing a period
58
of political liberalization since protests in October 1988. The Islamic Salvation Front (Front
Islamique du Salut or FIS) was founded in February 1989 and legalized by the Algerian
government in September of that year (Thielmann 2014). In March 1990, an article in
Reuters observed: “Algeria has already recognized its own Islamic movement, the Islamic
Salvation Front. The eyes of all Tunisian politicians are fixed on Algiers to see how the
experiment develops” (Wright 1990a). On June 12, 1990, the FIS won landslide victories in
municipal and regional elections (Ibrahim 1990). This event seems to have led both Nahda’s
leaders and Ben Ali to conclude that Nahda posed a greater threat to the Tunisian regime
than they had previously believed. Four days after the Algerian elections, an issue of ElFajr went to press with an article by Ghannouchi accusing Ben Ali’s government of being
undemocratic and calling for a “popular struggle” against the regime (Reuters 1990h). The
government seized the issue and a week later a court suspended El-Fajr for three months.
Murphy (1999: 194-195) writes: “Ben Ali became even more determined not to take the risks
associated with either legalizing the Islamist opposition or providing it with serious room
for political manoeuvres. The summer months of 1990 therefore witnessed a clear effort to
clamp down on the dissemination of Islamist ideas. The Tunisian media was ‘discouraged’
from reporting on the FIS victories in Algeria and technical failures were blamed for the
absence of radio and television coverage of the Algerian election results. Islamist criticism of
the Tunisian government led to arrests, imprisonments and the seizing of supposedly libelous
newspapers.” Similarly, Shahin (1997: 101) argues: “Confronted with a politically viable
Islamist opposition party and alarmed by the stunning victory of the Islamic Salvation front
(FIS) in the municipal and provincial elections in Algeria, Ben Ali decided to eliminate
al-Nahda as a potential threat and suppress the movement.”5
The Gulf War may also have revealed that Nahda’s ability to mobilize protestors was greater
than either side previously reckoned. Ben Ali was initially critical of Iraq when it invaded
5
Cowell (1990) observed at the time of the Algerian election: “The Algerian result has greatly sharpened
President Ben Ali’s problems over his country’s Islamic movement, En Nahda.”
59
Kuwait in August 1990 (Murphy 1999: 195). His position turned out to be very unpopular
with Tunisians. Nahda sensed an opportunity. “The Islamists began to take to the streets to
protest against both the government’s position and the lack of progress in domestic political
openings. ... Ghannouchi developed a suddenly more belligerent tone, hoping that support
for Iraq would bring anti-regime demonstrations out in the streets of Tunis” (ibid ). The
government cracked down on the protests violently, but this response initially provoked
more widespread unrest throughout the autumn (Middle East Economic Digest 1990d; Wolf
2017: 73). Arrests of Nahda members prompted further protests in January 1991 (Middle
East Economic Digest 1990f; Reuters 1991f). February protests by Islamists were “violently
suppressed” (Europa Publications 1994:879). As Nahda flexed its mobilization muscles,
Tunisia’s government seized individual issues of El-Fajr and in February 1991, shut it down
permanently. As Willis (2012: 168) argues: “An-Nahda members demonstrated against the
arrival of US troops in Saudi Arabia following Iraq’s invasion of Kuwait in August. In the
view of some, this persuaded the Tunisian regime to move decisively against An-Nahda. The
following months saw a gradual tightening against the movement, with restrictions on the
party’s activities, the closure of its newspaper and eventually the arrest of large numbers of
its activists and supporters.”
Thus, Ben Ali withdrew Nahda’s access to the media, I argue, because he concluded, from
the example of the FIS and Nahda’s ability to mobilize protests in the fall of 1990, that
the group posed a greater threat than it appeared to at the start of 1990. Although Ben
Ali seemed to have a firm grip on the regime and its repressive apparatus at this stage, if
Nahda could mobilize enough protestors, the security forces might become unwilling to use
enough violence to suppress them. Rached Ghannouchi, leader of Nahda, confirmed that
Ben Ali cracked down on the movement because he realized it was stronger than he had
thought: “He changed his mind [about tolerating Nahda] once he saw that the popularity of
this movement was very large” (discussion with the author, May 2016).
60
Why didn’t Ben Ali allow El-Fajr to return after crushing Nahda? Perhaps he calculated
that, while he had driven the movement underground, he had not eradicated its widespread
support, and that the group would seek to topple the regime if he ever allowed it a voice
again. Waltz and Kéchichian wrote as late as 2009 that Nahda “continues to enjoy popular
support—perhaps more than ever in the wake of disappointment with the Ben Ali
government. ... It remains the most significant opposition group in contemporary Tunisia.”
As further evidence that Nahda remained the most popular opposition group, after Ben Ali
was ousted in January 2011, Nahda came first in Tunisia’s first free and fair elections and
became the head of the governing coalition (EIU 2011X CR Tunisia December).
3.6
Why secular opposition newspapers were shut
down, but only briefly
Ben Ali also cracked down on the newspapers of secular opposition groups in 1990, not
because any one of them was individually strong enough to challenge the regime, but
because they showed signs of forming a broad alliance with Nahda. As Alexander (2010:
59) wrote of the secular parties: “Their poor showing in the April [1989] elections gave
them little bargaining power in their own right. But they could play on Ben Ali’s fear of a
united opposition front. By drawing closer to Ennahdha, they might force him to make
concessions that they could not win on their own. ... Iraq’s invasion of Kuwait reinforced
this opposition unity.” There was a precedent for this cooperation. Ali Laarayedh, a leader
in the MTI, remarked that ties between the Islamist and secular opposition groups grew
stronger in the 1980s (discussion with the author May 2016), an observation echoed by
numerous interviewees.6 As Perkins (2016a: 168) notes, discussing the years just before
6
The leftist Samir Taieb, MTI official Abdelhamid Jelassi, and Nahda member Saadok Sghiri confirmed
the existence of cooperation between the Islamist and secular opposition during the 1980s (discussions with
61
Ben Ali came to power: “The specter of an alliance, however unlikely, between the MTI
and one or another of the secular opposition parties alarmed the government.”
Admittedly, there were some signs of a secular-Islamist divide in early 1990. Reuters reported
in April 1990: “Three Tunisian leftist parties announced on Wednesday a common manifesto
which they hope will serve as the rallying point for a broad front against both the ruling
party and the powerful Islamic movement. ... The Tunisian left in general is worried that
the growth of the Islamic movement, now the second political force in the country, will leave
them permanently on the sidelines.” The joint statement said: “Our plan is ... to build a
democratic civilian state and to reject any plan for a religious state.” However, it also “called
for the recognition of all political parties”, implicitly criticizing the government’s refusal to
legalize Nahda (Wright 1990c). Later that year, the secular opposition took bolder steps
to coordinate with Nahda. All six legal opposition groups and Nahda boycotted the local
elections held in Tunisia on June 11 (Legum 1992a: B 520). After the government arrested
hundreds of Nahda supporters and leaders in the fall of 1990, MDS leaders condemned this
repression of the Islamists at their party congress. In January 1991, the six legal parties
issued a joint declaration, calling for a real democracy that would include secularists and
Islamists alike (Legum 1992a: B 520). In this context, with the secular opposition parties
lending their support to Nahda’s challenge to the state, Ben Ali closed down their newspapers
as well as Nahda’s.
However, over the course of 1991, as the government began successfully dismantling
Nahda’s organization, the secular parties realigned with the regime and their newspapers
were restored.
Whereas Nahda and the secular opposition had, in the words of
Ghannouchi, once acted in “coordination against our government,” Ben Ali convinced the
secularists to join “a sort of common war, total war against Nahda” (discussion with the
the author, May 2016).
62
author, May 2016). Early in 1991, the “government and the lay opposition parties which
they recognised” acted in “joint opposition to the Islamic movement” (Donnadieu 1992:
39). “Although the secular opposition was vocal in its denunciation of the most repressive
government measures, it was significantly ‘bought off’ by a coincidence of interests with the
ruling regime” (Murphy 1999: 201). In July of 1991 “ Ben Ali decorated [opposition] party
leaders for their services to the state” (Middle East Economic Digest 1991c). The secular
opposition papers began to reappear in November 1991.
The legal opposition parties ultimately resigned themselves to working within Ben Ali’s
democratic façade. The head of the MDS in 1992, Mohamed Moadda, publicly praised Ben
Ali’s policies and moved the MDS program closer to the RCD’s (Murphy 1999: 210; Perkins
2016a: 195).7 The Tunisian Communist Party abandoned Marxism for a moderate social
democratic program, renaming itself the Renewal Movement (Harakat Ettajdid) (Murphy
1999: 211). For the remainder of Ben Ali’s time in power, most of the legal opposition parties
offered little serious challenge to Ben Ali and the RCD. Two former members of the RCD
central committee, Abir Moussi and Raouf Khamassi, both referred to these parties as the
“cartooneeya opposition,” literally the “cartoon opposition” (discussions with the author,
May 2016).
3.7
Conclusion
The first half decade of Ben Ali’s dictatorship saw considerable variation in the government’s
treatment of opposition media, both over time and across groups. This variation largely
corresponded to patterns predicted by the model presented in the previous chapter. Ben Ali
7
In 1995, Moadda began taking a more confrontational approach toward the RCD and found himself in
jail within a few months (Murphy 1999: 215).
63
was more willing to allow the secular opposition to produce newspapers than the MTI/Nahda,
because the secular groups were weaker than the Islamists, and thus unwilling to challenge
his regime. Ben Ali briefly allowed Nahda to have a newspaper, but only after the country’s
economic performance had considerably improved and Ben Ali had consolidated his own
power over the government and the ruling party. When it appeared that Nahda might be
stronger than he had thought, and the group began consistently taking to the streets, he
took away their newspaper and drove the group underground. In the next chapter, I will
test how well these patterns generalize to countries across the Arab region and over multiple
decades.
64
Chapter 4
“Constructive Criticism”: A
Cross-National Study of Opposition
Media in Arab Authoritarian Regimes
In May of 1990, the Yemen Arab Republic (North Yemen) united with the People’s
Democratic Republic of Yemen (South Yemen) to form the Republic of Yemen (Europa
World 2020). The new state decreed that multiple political parties would be allowed, and
by the end of the year, more than 30 had formed (ibid ). Moreover, the government passed
a law allowing these new parties to publish their own newspapers (Rugh 2004: 105). But
the law also placed strict limits on freedom of expression, forbidding the media from
criticizing the head of state. The law made an exception only for what the state deemed to
be “constructive criticism” (Jones 2001: 2674).
This episode in Yemen’s history epitomizes a trend in post-Cold War Arab countries: even
65
those that have allowed opposition groups to publish media maintain tight restrictions on
the content of those media. As Sakr (2003: 35) remarks: “The authorities in a given
country may say the country has a ‘free press’, because opposition parties are permitted to
publish newspapers alongside the government-owned dailies. Yet analysis of the full range
of laws relating to freedom of expression reveals that the media are not free.”1 Opposition
publications do sometimes criticize government policies or officials, but there are lines they
generally do not cross. In particular, criticizing the regime leader is usually taboo, and
thus generally avoided. For example, in Mubarak’s Egypt: “The partisan media featured
critical coverage of prominent figures, though it faced more consistent repression when their
criticism reached the president himself and as such exercised self-censorship accordingly”
(Kamal 2018: 44).2
Governments often prevent prevent opposition groups from disseminating their media at
all, whether banning the distribution of newspapers (Motaouakal 2014: 68), jamming radio
signals (Dow Jones International News 2003), or blocking citizens’ access to websites (U.S.
Department of State 2011). So why do governments sometimes allow opposition groups to
publish media, if they have the power to prevent them? The tamest opposition media, after
all, is no opposition media. The answer, I have argued, is that tolerating opposition media
can be a tool for managing threats of unrest and perpetuating the survival of authoritarian
regimes. For a dictator, a dash of public criticism — as long as it is only a dash — can indeed
be constructive: opposition media that is only mildly critical signal to dissident citizens that
they are in no position to challenge their rulers.
1
Campagna (1998a) observed that, while in some parts of the Middle East and North Africa, no
independent media were tolerated, even in countries in the region “where the press enjoys a greater degree
of freedom, governments nevertheless use press laws and criminal defamation statutes to deter outspoken
journalists.”
2
Krämer (1992: 24) similarly observes that, even when Arab authoritarian regimes have allowed
opposition groups to have newspapers: “Some topics remain taboo—primarily God, army, king or president.”
These taboos have remained largely intact throughout the post-Cold War period in authoritarian Arab
countries. Korany (2010: 80) writes: “There is a, sometimes unwritten, code whereby residents of each
country know their red lines, and those who dare cross know they will have to bear the consequences.”
66
In this chapter, I present and analyze an original dataset on government toleration of
opposition media in Arab authoritarian regimes during the years 1992 through 2017. In the
next section, I discuss major trends in the region in this era that make it a fitting context
in which to test my theory. In the following section, I discuss the construction of my
dataset for measuring the environment for the opposition media in Arab authoritarian
regimes.
Next, I present the focal explanatory variables that I use to test the five
hypotheses derived from my theory, along with the control variables I use. I then describe
the empirical methodology and present the main results, using two dependent variables for
measuring toleration of opposition media, drawn from the original dataset. To test the
robustness of these findings, I also present results from alternative specifications of the
empirical models and an alternative version of my data. I conclude with a discussion of the
overall strength of the evidence these data provide for my theory.
I find that, in some times and places, Arab authoritarian governments since the early 1990s
silenced opposition voices, but in other cases they allowed a degree of media pluralism,
legalizing — or at least not preventing — the publication of opposition media.
This
bifurcation of approaches was manifest in regime’s treatment of opposition newspapers in
the early 1990s and continued through their responses to opposition satellite TV channels
and websites in the 2000s and 2010s.
As my data show, whether a regime tolerated
opposition media or not is largely consistent with the pattern my theory predicts. When
regimes are strong, they signal and reinforce their strength by tolerating opposition media.
When they are weak, they foster uncertainty by shutting out opposition voices.
67
4.1
The Arab Region Since the Cold War
The Arab region after the Cold War is a fruitful setting to test the theory laid out in Chapter
2. The region in this period has been characterized by both pervasive authoritarianism and
widespread challenges to governments by opposition groups. Amid varying economic and
political conditions, the region’s governments have responded to these challenges with a wide
range of repression and accommodation.
When the post-Cold War era began, there was not a single democracy among Arab countries
(Korn 1993: 22). The regimes’ hold on power was bolstered by a “vast expansion of the
state apparatus and its power to control and coerce, which had occurred in the 1960s and
early 1970s” across the region (Owen 1981). Yet many of these authoritarian governments
faced calls for political change from new opposition movements that had emerged across
the region in the 1970s and 1980s, particularly, though not exclusively, Islamist movements
(Brown, Hamzawy, and Ottaway 2006: 5). These new opposition groups quickly showed
their ability to mobilize protests and to win significant numbers of seats in elections (even
rigged ones), posing a new challenge to the survival of Arab authoritarian regimes. A
turning point came in December 1991, when the Islamic Salvation Front (Front Islamique du
Salut or FIS) won 41.5% of the votes in the first round of Algeria’s parliamentary election
(Economist Intelligence Unit 1996d: 4). The country’s panicked military leaders intervened,
ousting the reformist president Chadli Benjedid, canceling the second round of elections, and
initiating a devastating crackdown on the FIS that plunged the country into years of civil
war (Aghrout and Zoubir 2019). The FIS’s powerful showing in the election showed that
opposition groups, particularly Islamist ones, could muster massive public support. Regimes
would have to adapt to survive. This chapter will focus on the decades after this turning
point.
68
The rise of new movements (and the revitalization of some old ones), with their demands
for democracy, Islamic government, or both, grew in part out of the perceived economic and
political failures of the ideologies on which the newly independent Arab states had based their
popular appeal, in particular socialism and pan-Arab nationalism (Owen 1981; Haddad 1992;
Krämer 1992; Murphy 1998; Muslih and Browers 2014; Hamid 2017). Geopolitical events,
such as the breakup of the union of Egypt and Syria in 1961 (Europa 2019a), Israel’s decisive
defeat of its Arab neighbors in the Six Day War of 1967 (Hamid 2017), and the Israel invasion
of Lebanon in 1982 (Murphy 1998: 78) took the sheen off of Arab nationalism. By the
1980s, the region’s economic woes had become pronounced as well, with Arab states facing
crushing levels of debt, mounting corruption, swollen bureaucracies, and rapid population
growth (Murphy 1998: 78). Even oil revenue, which had boomed for the major producers
in the 1970s (Owen 1981), was flat-lining (Murphy 1998: 78). The “structural adjustment”
policies that governments implemented to try to solve their economic troubles by reducing
the role of the state in the economy did not endear them to their populations either (Murphy
1998: 79). Reduced subsidies for food provoked “bread riots” in Egypt in 1977, Morocco
and Tunisia in 1984, Sudan in 1985, Algeria in 1988, and Jordan in 1989 (Sadiki 2000: 80).3
The growing discontent in Arab countries fueled the growth of opposition movements. For
example, although Islamists did not spark the “bread riots” in Tunisia in 1984 and Algeria
in 1988, they used the unrest as an opportunity to organize protests of their own (Zhang
2019; Global Security 2020).
Opposition groups also benefitted indirectly from these
episodes of unrest, in those situations where governments responded with political
liberalization measures. For example, months after its own “bread riot” in 1989, Jordan
held its first parliamentary election in over two decades (Europa World 2019f). Candidates
affiliated with the Muslim Brotherhood became the largest bloc in parliament (Schwedler
3
Sadiki (2000: 82) argues that, while food price spikes triggered these protests, participants were driven
by a wide range of frustrations, including “social inequality, corruption, nepotism, authoritarianism, and the
regime’s incompetence.”
69
2015).4
In subsequent decades, continued mobilization reminded the region’s dictators that they
could not ignore the threat that opposition groups posed to their hold on power.
In
Mauritania, for example, the Front of Opposition Parties led demonstrations against the
government involving thousands of people in 1997 and 1999 (Reuters 1997b; BBC
Monitoring: Middle East 1999a).
The frequency of protest in the region picked up
considerably during the 2000s (Ottaway and Hamzawy 2011). Algeria’s Socialist Forces
Front and the Berber Cultural Movement led protests in favor of the rights of Berbers in
2001, some attracting tens of thousands of participants (Agence France Presse 2001a;
Drummond 2001; Reuters 2001b). Egypt’s Muslim Brotherhood led thousands into the
streets to demand political reforms on multiple occasions in 2005 (Abou El-Magd 2005;
Slackman 2005).
The “Arab Spring” uprisings of 2010 and 2011 posed the greatest threat to the region’s
authoritarian regimes in years, but they also raised doubts about the continued relevance of
traditional opposition groups. Howard and Hussain (2011: 48) argue that social media and
mobile phones enabled “distributed leadership” during the uprisings in Tunisia, Egypt, and
elsewhere. The new protests, they argued, were not spearheaded by centralized opposition
organizations with hierarchical leadership structures, but by networks of ordinary citizens.
And yet, many opposition groups with identifiable leaderships did play an important role
4
Sadiki (2000: 75) argues that “food protest was a leading factor in influencing government policy toward
democratic reform in Sudan and Jordan in 1986 and 1989, respectively.” Krämer (1992: 23-24) observes that
in numerous countries in the region during this period, “Socioeconomic tension, accumulating over time and
coinciding with reduced state-financed social services, erupted in urban riots which threatened the regimes’
legitimacy and survival and were quelled by the army. Yet the response also included important concessions,”
including “greater political freedom.” Nor were Islamists the only groups to benefit from liberalization. The
exiled former Algerian president Ahmed Ben Bella launched the Movement for Democracy in Algeria (MDA)
in 1984 (Naylor 2015: 117), but the Algerian state only legalized the group as a political party in March
1990, in the period of political opening triggered by the 1988 protests. A few months after it was legalized,
the MDA flexed its new political muscles by organizing an anti-government demonstration attended by over
20,000 (Reuters 1990c).
70
in organizing “Arab Spring” protests. The April 6 Youth Movement issued the call for
the massive January 25, 2011 protest against Hosni Mubarak’s government (Radio Free
Europe 2011; Public Broadcasting Service 2019). That same month, thousands joined a
demonstration in the Yemeni capital organized by the Joint Meeting Parties (Al-Masmari
and Coker 2011), a coalition led by the Islah party and the Yemeni Socialist Party (Durac
2011: 343). The Rally for Culture and Democracy, an Algerian party founded in 1989 (Naylor
2015: 439), led a protest in Algiers in February 2011 attended by over 10,000 (Faucon 2011).
Organized opposition groups have continued to mobilize their followers in the years since. In
2017, Morocco’s Justice and Charity, an Islamist group that has opposed the monarchy since
the 1980s (Laremont 2009), led more than 10,000 protestors in Rabat (Errazzouki 2017).
Over the decades, regimes have responded to these opposition challenge with both repression
and liberalization; in many cases, the same regime alternated between both (Buttorff 2019:
3). Generally speaking, both responses can be seen as strategies for “system maintenance”
(Krämer 1992: 23-24.) There were early hopes that political liberalization might lead to
full democratization across much of the region (Hudson 1991; Harik 1994). Indeed, a “wave
of democratization” (Huntington 1991: 12) had begun in the mid 1970s and spiked just
after the Cold War ended: numerous dictatorships gave way to new democracies in Latin
America, Eastern Europe, sub-Saharan Africa, and East Asia (Møller and Skaaning 2013:
99-103). Yet authoritarian government has remained remarkably prevalent in the region,
even after the hopes stirred by the revolutions of 2011 (Schmitter and Sika 2017; Freedom
House 2019). If both repression and liberalization can keep regimes in power, why does
a particular government in a particular time choose one strategy rather than the other?5
The theory presented in Chapter 2 offers an explanation of governments’ differing choices in
5
A regime can of course engage in (and even increase) repression in some domains while liberalizing or
making other concessions in other domains (Rasler 1996: 139; Inclán 2009: 795). Since I focus on the
particular domain of whether a regime tolerates opposition media, such mixed strategies are less relevant.
However, in the conclusion, I propose future research on why regimes sometimes tolerate one opposition
group’s media but not another group’s and why they even tolerate some media outlets of a particular group
while suppressing others produced by the same group.
71
either prohibiting opposition media or tolerating them.
Dramatic changes in media markets and communications technology in the post-Cold War
era have led some scholars to question whether governments in the Middle East and North
Africa can still control the flow of information to their citizens. Arab media have been
transformed in many ways by satellite television and the Internet (Gunter and Dickinson
2013: 1). When these trends were beginning to emerge, many authors expressed optimism
about their effects on freedom of expression in the Arab region. “In considering Arab
censorship, we must pay attention to the recent advent of satellite TV and the Internet,
which are paving the way for the easing of censorship, as they are proving impossible to
control” (Khazen 1999: 89).
Egypt began broadcasting television channels across the region by satellite in 1990, and
the privately-owned and London-based Middle East Broadcast Centre launched its satellite
channel across the region the following year (Berenger 2006: 207). In 1996, Qatar-based
Al-Jazeera became the first 24-hour Arab news channel (Korany 2010: 69). By the late
2000s, there were hundreds of satellite channels broadcasting across the Arab region and
tens of millions of viewers (Campagna 2009). In some ways, the satellite revolution undercut
censorship, as governments had to contend with an expanded volume of information flowing
in across their borders. “Arab governments could no longer hide information from their
publics, since the skies were by then virtually open, and the publics could receive the news via
satellite from anywhere around the world” (Korany 2010: 67). Some Arab satellite channels
have hosted talk shows that air discussions on topics like human rights and democracy (Ayish
2011: 97) and have had opposition leaders on as guests (Ghareeb 2000: 406).
Yet governments in the region have not been powerless to control the content that satellite
channels broadcast. In 2008 , information ministers from across the Arab region signed a
72
charter to regulate their satellite broadcasters, calling on states to take legal action against
broadcasters that had a “negative influence on social peace and national unity and public
order and decency” or that insulted “leaders or national and religious symbols” (Campagna
2009; Ayish 2011: 96; Korany 2010: 82). Arab governments have often interfered with
satellite broadcasters by shutting down their local bureaus or refusing to transmit them on
government-controlled satellites (Campagna 2009). Moreover, many Arab satellite channels
receive funding either directly from states or from close government allies (Korany 2010: 80).
Nonetheless, satellite broadcasters have no doubt made Arab citizens more aware of events
in other countries in the region, including anti-government protests. In 2005, Al-Jazeera
“reported extensively on ongoing political protests in countries like Lebanon and Egypt”
(Hafez 2010: 332). Al-Jazeera and Al-Arabiya brought news of the “Arab Spring” protests
to viewers across the region (Rosiny 2012: 4). Yet, as I will argue, this demonstration effect
of the “Arab Spring” protests actually led governments to reduce media freedom, increasing
the likelihood that they would not allow opposition media.
In the 2000s, access to the Internet spread rapidly in Arab countries, although with significant
variation from one country to another. Internet use in the region increased from one-tenth
of one percent in 2000 to nearly 20 percent by middle of the decade (El Gody 2007: 217). In
2017, an estimated 46.5% of individuals in all Arab countries used the Internet (International
Telecommunication Union 2020). Across all authoritarian Arab regimes with populations
over a million, the Internet use rate ranged from 21% in Mauritania to almost 100% in
Kuwait ibid ).
According to El Gody (2007: 218), the spread of Internet access enabled Arab citizens to
circumvent state controls on information. As Korany (2010: 76) puts it, the Internet gave
“users access to any news source they wanted, anywhere on the planet. These news sources
73
range from credible, brand-name news organizations (CNN, BBC, Reuters, and so on) to
news items written by individual citizens (blogs). The mere presence of the Internet as a
source of information therefore helps open up a freer space for public debate, and makes it
much more difficult for governments to censor information since the same information will
appear in a multitude of other sources in or out of the jurisdiction of the censoring country.”
El Gody (2007: 221) also argues that the Internet has enabled opposition groups to organize
anti-government protests more easily. Lim (2012), for example, traces Egyptian opposition
groups’ use of online tools to mobilize protestors, from Kefaya’s use of blogs beginning in
2004 (237) to the April 6 movement’s use of Facebook and Twitter in 2008 (240) and the
dissident Facebook group “We are all Khaled Said” that organized large protests in 2010
(241). In a survey of participants in the massive Tahrir Square protest in Cairo on January
25, 2011, Tufekci and Wilson (2012: 370) found that 28% of participants first learned the
protests would occur through Facebook; only “face-to-face communication” was a more
common way to learn of the protests, while less than 1% learned of the protests through
radio or newspapers.
Yet, like authoritarian governments elsewhere in the world, Arab dictatorships have adapted
to the growth of the Internet and found new ways to prevent citizens from using it against
them. Governments frequently resort to blocking access within their borders to dissident
websites (El Gody 2007: 225; World Association of Newspapers and News Publishers 2010:
669, 1059; OpenNet Initiative 2019a, 2019b, 2019c, 2019d, 2019e, 2019f). Some Arab citizens
use tools such as virtual private networks (VPNs) to circumvent filtering of websites (World
Association of Newspapers and News Publishers 2010: 697), but governments can also block
sites that provide VPN services, as Egypt began doing in 2017 (Rohan 2017). Thus, when
the Egyptian government later started blocking the news website Al Bedaiah, which often
covered human rights abuses by the government, traffic to the site fell by 98% (Reporters Sans
Frontières 2017). Some governments have even gone so far as to hack opposition websites
74
(BBC Monitoring: Middle East 2010a).
Arab governments also employ old-fashioned offline repression to control what dissidents post
online, frequently arresting dissident bloggers and other online activists (Korany 2010: 82).
Governments are unable to block opposition accounts on social media sites like Facebook
and Twitter, without blocking the platforms entirely. Instead, governments have used heavyhanded tactics to deter opposition groups from posting highly critical content on social media.
In 2013, for example, Kuwait “arrested at least a dozen persons for Twitter posts it deemed
to be defaming to the emir,” imposing multi-year sentences on some (U.S. Department of
State 2013). In 2014, Egypt’s government arrested the administrator of one of the Muslim
Brotherhood’s Facebook pages (Freedom House 2014b). In 2015, Bahrain imprisoned a
leading member of the Islamist group Al-Wefaq for criticizing the country’s elections on
Twitter (Amnesty International 2015; Bahrain Mirror 2015).6 Not only do the region’s
governments repress the producers of critical online content, they also intimidate would-be
consumers of that content by monitoring citizens’ Internet use, from installing cameras in
Internet cafes (World Association of Newspapers and News Publishers 2010: 638) to tracking
the websites users visit (Committee to Protect Journalists 2009).7
So while the Internet has provided Arab citizens with new tools to learn about the abuses
committed by their governments and to coordinate with their fellow citizens to protest
those abuses, the region’s authoritarian regimes maintain considerable control over the flow
of information. Rather than ending censorship in the region, the Internet has expanded the
technological terrain on which dissidents and states compete to control information.
Empirically, this means that, in order to measure government toleration of opposition
6
According to Algerian dissidents, “even the slightest misstep in a Facebook update could result in arrest
and questioning” (U.S. Department of State 2014).
7
In some countries in the region, all Internet traffic is routed through government-run servers (Committee
to Protect Journalists 2009), while in others governments force private Internet Service Providers (ISPs) to
implement controls (World Association of Newspapers and News Publishers 2010: 267).
75
media over the entirety of the post-Cold War era, we must be attentive to the new media
technologies by which opposition groups communicate with the public and the evolving
means by which governments have sought to regulate opposition media.
4.2
A Dataset of Arab Opposition Media
In order to test my theory, I constructed a cross-national dataset on the media of all
significant non-violent opposition groups under non-democratic Arab regimes (in countries
with populations over a million) during the years 1992-2017. I will discuss how I define
such groups and regimes below.
This dataset contains the two dependent variables I
examine in this chapter, each providing a measure of regimes’ tolerance or intolerance of
opposition media. Simply put, I aimed to measure when governments allowed opposition
media outlets to operate and citizens to access them, and when they did not. Coding
proceeded in three stages: 1) identifying the set of authoritarian regime-years in the region,
2) identifying all significant non-violent opposition groups in these regime-years, 3)
identifying, for each country-year and each group, whether or not the government allowed
the group to publish its own media. I considered both legal and illegal groups, some of
which participated in elections and others which did not. I also include a wide range of
media: print, broadcast, and online.
To code regimes, opposition groups, opposition media outlets, and whether or not regimes
tolerated those outlets, I consulted country case studies, historical dictionaries and other
encyclopedic sources, human rights reports from the U.S. State Department and
non-governmental groups such as Amnesty International, media freedom reports from
groups like the Committee to Protect Journalists and Freedom House, local and
76
international news media accounts, and archives of opposition websites, among other
sources. Each of these sources has potential blindspots, but by consulting a wide array of
sources, I aimed to minimize the chances of missing relevant evidence. For example, by
consulting many rather than few sources, each with different expected biases, I had a
better chance of coming across references to any opposition media outlets that existed in a
particular time and place, as well as references to state intervention against those outlets.
I define Arab countries as those countries that are or have been members of the League of
Arab States, more commonly known as the Arab League. There are 22 current or former
members (BBC News 2017d).8
To identify authoritarian regimes, I apply the criteria
developed by Geddes, Wright, and Frantz (2018). They define a regime in general as “the
set of very basic formal and informal rules for choosing leaders and policies” (5). An
authoritarian regime specifically is a non-democratic regime in an independent country, in
which the government controls most of the country’s territory.9 I use those authors’ regime
codings for Arab regime-years included in their dataset (Geddes, Wright, and Frantz
2014).10 I expanded the set of regime-years they coded in two ways. First, since their
dataset ends in 2010, I applied their rules to code additional authoritarian regime-years in
the region for 2011-2017. Second, two countries that had the authors excluded from their
dataset because their populations were too small — Bahrain and Qatar — had populations
over one million by 2018 (World Bank 2019)11 . Thus, I coded the regimes in both countries
8
Syria’s membership in the league was suspended in 2011 (Batty and Shenker 2011).
Authoritarian regimes begin when either 1) government leaders achieve power in some way other than
through free and fair elections or constitutional succession to an elected government; 2) an elected government
prevents subsequent elections from being free and fair; or 3) a government is chosen through an election but
the military intervenes to prevent popular candidates from running or dictates important areas of government
policy (Geddes, Wright, and Frantz 2018: 5). An authoritarian regime ends when one of the following occurs:
1) a competitive executive election — or an election to the body that chooses the executive, like a parliament
in some systems — is held, someone other than the incumbent or a close ally wins, and the winner is allowed
to take power; 2) the government is forcefully ousted (e.g. by a coup, mass uprising, or invasion) and a new
regime replaces it; or 3) regime elites substantially change the rules about who can become the leader or
who can choose policies (Geddes, Wright, and Frantz 2018: 6).
10
The “Autocratic Regimes Data” can be found at https://sites.psu.edu/dictators.
11
Data from the World Bank’s “World Development Indicators” can be found at:
https://databank.worldbank.org/source/world-development-indicators.
9
77
throughout 1992-2017. I did not include Comoros or Djibouti for any year, because the
population of each state remained under one million in 2018 (World Bank 2019). I also
modified the start and end years of some regimes. Whereas Geddes, Wright, and Frantz
(2014) code regime-years according to the regime that was in place on January 1 of that
year, I code regime-years according to the regime that was in place for the majority of the
year. Thus, for example, the regime I code for Algeria in 1992 is the regime that began on
January 11 of that year, whereas Geddes, Wright, and Frantz group Algeria in 1992 with
the regime that ended on that date. My approach enables me to consider the largest part
of a given year possible when coding how each regime treated opposition media in that
year. See Appendix 2 for a list of how I coded all regime-years across all Arab League
members and all years 1992-2017, along with short narratives on the events that began and
ended each authoritarian regime.12
Among the 20 Arab League members that had over a million inhabitants in 2018, numerous
years were excluded because they failed to meet at least one criterion of Geddes, Wright,
and Frantz’s definition of an authoritarian regime. Some countries are excluded in some
or all years because they did not have a de facto independent state, namely Iraq in 20032009, Lebanon in 1992-2004, and Palestine throughout the entirety of 1992-2017. Some
countries are excluded in some or all years because no single government clearly controlled a
majority of the country’s territory: Libya in 2012-2017, Somalia throughout 1992-2017, and
Yemen in 2015-2017. A few regime-years were excluded because a country had a democratic
government for the majority of the year: Egypt in 2013, Lebanon in 2005-2017, Mauritania
in 2007 and 2008, and Tunisia in 2012-2017. Finally, Tunisia in 2011 is excluded because
the government that ruled between the ouster of Ben Ali in January and the holding of
competitive elections in October meets the Geddes, Wright, and Frantz (2018) definition of
12
When available, the narratives are from the codebook for the Geddes, Wright, and Frantz (2014) dataset
(labeled “GWF”). For authoritarian regime start and end events not included in that dataset, I wrote
additional narratives.
78
a “transitional administration.”13 Altogether, among Arab countries with 2018 populations
over a million, I consider 416 regime-years as being authoritarian during this period.
I then identified all significant non-violent organized opposition groups that operated under
these authoritarian regimes during 1992-2017. In each regime-year, I coded whether there
was any major organized and nonviolent opposition group, and if so, whether there was one
that was clearly the most popular. By “major”, I mean a group that has enough support to
pose a non-negligible threat to the power of regime elites. If there was one major opposition
group that was clearly the most popular, I coded it as the only significant opposition group.
If there was no single clearly most popular group, then all major opposition groups were
considered significant and included in the dataset. I will now define these terms in detail.
A group is considered oppositional if it seeks substantial changes in some aspects of national
government policies and/or institutions.14 A group is not considered oppositional if any of its
leaders hold high executive office, e.g. cabinet ministers. (Membership in the legislature does
not preclude oppositional status.) A group is considered organized if it has an identifiable
leader or leaders. Groups are considered nonviolent if sources do not describe them as
insurgent, armed, etc and do not otherwise indicate that they persistently engage in violence
or advocate violence. Incidental acts of violence (e.g. clashes between protestors and police)
do not preclude coding the group as nonviolent overall. Nor do planned but isolated attacks
carried out by rank-and-file members exclude the group if the attacks do not appear to have
been directed by the group’s leaders.
13
Even if this short-lived government were coded as an authoritarian regime, it would have to be excluded
from the empirical analysis, since using regime fixed effects requires more than one observation per regime
(Allison 2009: 2).
14
Endorsing the incumbent executive for reelection (or endorsing the incumbent’s hand-picked successor)
is considered evidence that a group supports the regime leader’s agenda and that the group is therefore
non-oppositional. Separatist groups are excluded, since their goal is to withdraw from the jurisdiction of the
regime in question, rather than change its policies throughout its territory.
79
If, among the groups that fit the above criteria, there is a single group that most sources
describe as having more support than any other group, and the group had an active leadership
that was able to coordinate its members during the year in question, that group is coded as
the main opposition group and the only “significant” opposition group in that year. The idea
is to identify the focal opposition group if there is one, which would likely play a leading role in
coordinating any mass uprising against the regime and/or in governing a subsequent regime.15
If no such group can be identified, then I code a set of “significant” opposition groups. In
the absence of a “main” opposition, an opposition group is considered “significant” if, in the
prior seven years,16 the group fulfilled at least one of the following criteria: 1) participated
in a national (legislative or executive) election and its candidate/candidates won more than
10% of the vote,17 2) led an anti-government demonstration inside the country involving over
a thousand participants,18 3) was coded as the main opposition group, or 4) was the ruling
party.19 A group which fits at least one of these criteria has demonstrated that it likely has
15
Although a group can be coded as a major opposition group even if it is currently unable to mobilize
collective action due to state repression, there can only be a focal opposition group if that group is currently
able to mobilize collective action. Hence, a group must currently be able to mobilize protestors or voters
to be considered the “main” group, but it can continue to be coded as a significant opposition group for
another seven years (see description of “major” groups below) after losing the ability to mobilize, e.g. due
to a government crackdown.
16
The seven-year window is arbitrary, so in future research I will experiment with coding groups based
on shorter and longer windows. Shorter windows should tend to produce more false negatives — excluding
groups that have large followings — but fewer false positives — including groups that no longer have a large
following — and the converse holds for longer windows.
17
For legislatures with more than one chamber, the vote for the lower house will be used since such
chambers typically are more representative and have greater policy powers, so they are a better gauge
of voters’ support for different groups’ agendas. In multi-round elections, I use first-round results only if
subsequent rounds are runoffs, and include all rounds in legislative elections in which different seats are voted
on in different rounds.
18
Participation in the planning of a protest only qualifies the group if the sources identify it as either
the sole organizer, the leading organizer, or one of two leading organizers. The reason for this rule is that
if a group is one of several that organized a protest and did not play a prominent role, the regime learns
little from the demonstration about the group’s ability to mobilize. A demonstration that is described as
spontaneous can also qualify a group if the demonstrators are principally advocating the rights of the group
(e.g. the release of its leader from prison). Such a demonstration clearly indicates the group has support.
19
An election or demonstration that qualifies the group as significant may have occurred under a previous
regime, including democratic regimes. I consider opposition groups that used to be ruling parties as significant
on the assumption that ruling parties usually build up support networks while they are in power that survive
to some significant extent after their regime ends. Just two former ruling parties are subsequently coded as
opposition groups: the National Liberation Front (FLN) in Algeria in 1993-1996 and the Yemeni Socialist
Party in 1995-2000.
80
a non-negligible level of support. Even if the group subsequently decides not to mobilize
its followers into the streets, the regime’s leaders know the group probably has the capacity
to mobilize large numbers or that its supporters might mobilize in large numbers without
central direction.20 A group may be included in some years but not others depending on
its fulfillment of the above criteria in each year. In any year in which such changes occur, I
include the group if it fulfilled the criteria for more than half of the year.
I coded 39 significant opposition groups across the region during the covered period. See
Appendix 3 for a list of the significant opposition groups coded in each country and
explanations of their coding. To illustrate the coding of groups, consider Egypt. When the
dataset begins in 1992, I code the Muslim Brotherhood as Egypt’s main opposition group.
The Brotherhood “saw its membership soar” in the 1980s (Mattar 2004: 1622). The 1987
elections had made the Muslim Brotherhood the largest opposition group in Egypt’s
parliament (Ranko 2015: 109). The group also swept elections to the boards of many of the
country’s professional syndicates in the early 1990s (Goldschmidt 2013: 284; Ranko 2015:
109). Writing in 1996, the Economist Intelligence Unit (1996a: 8) observed: “The main
challenge to the regime comes from the Islamic trend, which can draw on a significant
groundswell of popular support. The largest, best-funded and organised Islamic group is
the Muslim Brotherhood.”
The Brotherhood remained Mubarak’s strongest opponents
until he was overthrown in February 2011.
Al-Anani (2016: 155) writes that “the
Brotherhood was the key opposition group in Egypt under Mubarak.”21 Given the expert
consensus on the Brotherhood’s preeminent support in the 1990s and 2000s, I code the
20
Thus a group which led a large protest in the last seven years, but which can no longer directly organize
protests because, for example, the government has driven its leaders into exile, can still be counted as
significant. I include such groups because, despite the leaders’ inability to directly organize collective action,
the government’s decision to tolerate or forbid the group’s media could still affect whether its supporters
protest.
21
Mattar (2004: 1622) refers to the Muslim Brotherhood during the 2000s as “the most important
representative of the Egyptian masses.” The Economist Intelligence Unit describes the group as “the best
supported of the opposition movements, with a national reach” (2008c: 7) and “the only opposition force
able to mobilise popular support” (2010X: 4).
81
group as Egypt’s main opposition in 1992-2010.
I continue to code the Muslim Brotherhood as the main opposition group in 2011 and 2012,
under the government led by the Supreme Council of the Armed Forces (SCAF) that took
power when Mubarak was ousted. Trager (2011) reports: “In the months since Mubarak’s
resignation, the Muslim Brotherhood has continued to demonstrate its unique capacity to
mobilize supporters. Protests continue to be held in Tahrir Square on most Fridays, and
those protests that are endorsed by the Muslim Brotherhood draw substantially larger crowds
than those that are not.” The group’s continuing popularity was confirmed in 2012 when its
Freedom and Justice Party won a large plurality of seats in the national assembly and its
candidate Mohammad Morsi won the presidential election (Europa World 2019a).
Egypt is not in the dataset for 2013, because I code it as a democracy for the majority of
that year. In July of 2013, a military coup toppled the Brotherhood-led government, and the
new military regime cracked down harshly on the group, imprisoning most of its leaders and
driving the rest into exile or hiding (Trager 2016: 230; Europa World 2019a).22 By the end
of 2013, the Muslim Brotherhood could no longer effectively organize its followers. Sherif
(2014: 7) reported that the Brotherhood “can no longer coordinate across Egypt.” Thus, in
2014-2017, under the regime led by Sisi, I do not code the Muslim Brotherhood as Egypt’s
main opposition.
In the absence of a main opposition group during this period, I code all Egyptian opposition
groups that meet one of the criteria as significant. This set includes the Muslim Brotherhood
itself, since it was the ruling party until 2013.23 The other significant opposition groups
during this period are the April 6 Youth Movement, No to Military Trials for Civilians, and
22
Thousands of regular members of the Muslim Brotherhood were arrested or killed in this crackdown as
well (France 24 2019).
23
It could also qualify as a significant group on the basis of having been the main opposition group until
2012.
82
(beginning in 2015) the Salafist Front. April 6 was founded in 2008, while the latter two
groups were founded after Mubarak’s ouster (Wedeman 2012; Al Arabiya 2013; Cousin 2014;
Public Broadcasting Service 2019). April 6 and No to Military Trials qualify as significant
opposition groups on the basis of large protests they organized in 2011 (Daily News Egypt
2011; Radio Free Europe 2011; Public Broadcasting Service 2019). April 6 continued to
reject the legitimacy of Sisi’s regime throughout the 2014-2017 period, while No to Military
Trials continued to advocate the end of military trials for civilians during this period (Al
Arabiya 2013; Guardian 2013; Cousin 2014). The Salafist Front organized a large protest
against the Sisi regime in November 2014 and remained critical of the regime throughout
the next three years (Shehata 2015; Salafist Front 2019a).
In the Egyptian case, the three new significant opposition groups in 2014-2017 all qualify
as significant on the basis of protests they organized. In other cases, groups are included as
significant because of their electoral performance. For instance, the Islamic Constitutional
Movement won more than 10% of the vote in Kuwait’s parliamentary elections in October
1996, qualifying it as significant during 1997-2003. In Morocco, the Democratic Bloc was a
coalition of five parties that won a combined 32.5% of the vote in the June 1993 parliamentary
elections, so it is coded as a significant opposition group during 1993-1997, until Justice and
Charity is coded as the main opposition group.
Some groups are coded as significant for more than seven years on the basis of multiple
qualifying events. For example, Algeria’s Socialist Forces Front (FFS) won 20.5% of the vote
in the first round of the December 1991 parliamentary elections and organized large antigovernment demonstrations in July 1998 and May 2001, qualifying as a significant opposition
group throughout 1992-2007.
There were groups that I coded as moving from non-oppositional status into opposition, and
83
vice versa. Yemen’s Islah party illustrates both possibilities. The ruling General People’s
Congress (GPC) party established Islah in 1990 (Durac 2019). Until 1997, Islah held seats
in Yemen’s cabinet (Hamzawy 2009: 5-7; Schmitz and Burrowes 2018: 257), and in 1999
Islah endorsed President Saleh for re-election (Economist Intelligence Unit 2008b: 5; Durac
2019). However, according to the Economist Intelligence Unit (2002c: 10), “In early 2001,
Islah underwent something of a renaissance: growing fiercely critical of the GPC-backed
constitutional reforms, and opposing the government on other key issues.” In response, the
government began arresting Islah members of parliament and some of its supporters (ibid ).
The following year, Islah and other parties formed the opposition Joint Meeting Parties
(JMP) coalition (Hamzawy 2009: 8; Durac 2011: 343-344). Starting in 2001, I therefore
count Islah as an opposition group (and in fact, the main opposition).24 However, after
Saleh transferred power to vice president Hadi in November 2011, the JMP endorsed the
confirmation of Hadi’s presidency in a February 2012 plebiscite and entered into a coalition
government with the GPC, with Islah taking seats in the cabinet (Reuters 2011c; Durac
2019). Thus, under the regime led by Hadi in 2012-2015, I once again code Islah as nonoppositional.
Other groups have left the dataset when they became violent or entered the dataset when
they abandoned violence. The FIS in Algeria illustrates the first pattern: they appear in the
dataset during 1992-1994, but they formed the insurgent Islamic Army of Salvation (AIS)
in July 1994, so they no longer appear in the dataset starting in 1995 (Willis 1996: 327;
Willis 2012: 174; Aghrout and Zoubir 2019).25 Sudan’s Umma Party (UP) went in the
other direction. When a new regime came to power in 1989, the UP formed an alliance
against the regime with armed groups such as the Sudan People’s Liberation Army (BBC
24
Al Jazeera (2011a) describes Islah as “the main opposition party in Yemen.” Schmitz and Burrowes
(2018: 256) argue that Islah was “the most formidable challenger to the Saleh regime.”
25
The FIS remains in the dataset for 1994 because it transitioned to violence after the mid point of the
year. Thus, for most of the year I code it as a non-violent group. In determining whether the group had any
media in 1994 and whether it was tolerated, I code on the basis only of the months before they formed the
AIS.
84
Monitoring: Middle East 1990a; New Sudanese rebel radio heard following announcement
by Radio SPLA; Berry 2015: 252; Woodside 2017). However, the UP left the alliance in
March 2000 and turned to peaceful activism, so it began appearing in the dataset (Pineau
2000; Woodside 2019).
In all, the dataset covers 202 regime-years, spanning ten countries and thirteen regimes. Some
authoritarian regimes in the region, such as Libya under Qaddafi and the Saudi monarchy,
are not included in the dataset at all, because I did not find any organizations that met
my definition of significant non-violent opposition groups. The panel is unbalanced for two
reasons. First, some regimes began or ended during the 1992-2017 timespan. Second, some
regimes entered the dataset after 1992 when they went from having no significant opposition
groups to having at least one, or left the dataset before 2017 when they ceased to have
any significant opposition groups. Table 1 lists the regimes that had at least one significant
opposition group and which therefore appear in the dataset. Each regime is denoted by
its beginning and end year. When a regime remained in power at the end of 2017, its
end year is listed as “NA.” The second column lists the years during which the regime
is included in the dataset. The third column lists the total number of opposition groups
that I coded as significant under the regime during 1992-2017 and the last column lists the
“main” opposition group associated with each regime, if any.26 Four regimes (e.g. Algeria
1992-NA, Egypt 2013-NA, Kuwait 1961-NA, and Mauritania 2008-NA) did not have any
identifiable “main” opposition group during the covered period. In some regimes that had
a main opposition group, the group was not considered the main opposition throughout the
entire period (e.g. Mauritania 1978-2005 and Tunisia 1956-2011).
There was a main opposition group in 57.4% of regime-years. In around 70% of regime26
In principle, a regime could have had more than one “main” opposition group in different years, but I
did not find any such cases during the 1992-2017 timeframe.
85
Table 4.1: Regimes included in the dataset
Regime
Years Included
Groups
Main
Algeria 1992-NA
1992-2017
9
Bahrain 1971-NA
2002-2017
1
Al-Wefaq
Egypt 1952-2011
1992-2010
1
Muslim Brotherhood
Egypt 2011-2012
2011-2012
1
Muslim Brotherhood
Egypt 2013-NA
2014-2017
4
Jordan 1946-NA
1992-2017
1
Kuwait 1961-NA
1997-2003, 2011-2017
6
Mauritania 1978-2005
1992-2005
2
Mauritania 2008-NA
2009-2017
5
Morocco 1956-NA
1992-2017
4
Justice and Charity (Ihsan)
Sudan 1989-NA
2000-2017
1
Umma Party
Tunisia 1956-2011
1992-1999
1
Nahda
Yemen 1978-2012
1992-2011
3
Yemeni Congregation for Reform (Islah)
Muslim Brotherhood
Union of Democratic Forces (UFD)
years, there was a single significant opposition group.27 The maximum number of significant
opposition groups in a single regime-year was five.28 Figure 1 shows the overall distribution
across regime-years of the number of groups (groups N ).
For each opposition group, I looked for evidence that the group had ever published a
newspaper, magazine, radio station, TV channel, or website during the 1992-2017 period. I
did not include pages on major social media sites based outside the region, such as
Facebook and Twitter, because in recent years it has been technically impossible for
governments to block individual pages without blocking an entire platform. Thus, the
27
The 26 observations in which there was a single significant opposition group but I did not code it as the
“main” opposition group fall into two sets. One set consists of cases like Nahda after 1993, in which a main
opposition group was driven underground. The other set consists of cases like the Coordination des Arush,
Dairat, et Communes (CADC) in Algeria in 2008-2010. The CADC was the only opposition group that met
the protest or electoral criterion for significance in those years, but there was not a consensus among sources
that the group was the most popular opposition movement in the country.
28
For example, in Mauritania in 2016, there was the Rally of Democratic Forces (formerly known as
the Union of Democratic Forces), the National Front for the Defense of Democracy, the Coordination of
Democratic Opposition, the National Rally for Reform and Development (Tewassoul), and the National
Forum for Democracy and Unity.
86
Figure 4.1: Significant opposition groups per regime-year
existence of opposition pages on these sites is not a clear indication that the government
has allowed the opposition to have media. I excluded any media outlet that sources did not
describe as being “owned” or “run” by the group or its top leader — leaving out
publications produced by sympathizers if its content is not ultimately controlled by the
group’s leaders — because I am interested in group leaders’ ability to communicate
publicly.
I consider an outlet not to be permitted if the government compels or tries to compel it
to to stop producing any content (for example, shutting down the outlet’s main office or
studio), prevents or tries to prevent the dissemination of content (for example, by blocking
access to or hacking a website, jamming a broadcast signal, or trying to seize all copies of a
newspaper),29 or transfers control over the outlet to the state or regime allies. I refer to the
29
I do not count quotas imposed on the number of copies of a print publication that can be distributed
as a banning, since the group is still allowed to distribute some copies. I did not find any cases in which a
quota was so stringent as to make distribution of a publication negligible.
87
event that marks the beginning of a period in which an outlet is not permitted as a banning.
A banning can happen to a media outlet that was previously permitted under the current
or a previous regime, but can also happen to an outlet that was never permitted in the first
place, if the government impedes it within two months after it begins publishing.30
When determining if a group had any permitted outlets and if any of them were banned, I
do not consider outlets that typically published less frequently than once every two weeks.
I will refer to outlets that published sufficiently frequently as publishing “actively.” In some
cases, evidence on frequency comes from references or other secondary sources, while in other
cases there is more direct evidence, e.g. dated posts on a website. If an outlet begins in a
given year or is banned for part of the year, I only base inclusion of the outlet in the dataset
on the rate of publication after the outlet was launched and when it was not banned.
Radio and TV almost invariably broadcast daily, and so sources typically only mention how
often the station broadcasts if it is less than daily. Thus, I assume that radio and TV outlets
broadcast more frequently than once every two weeks unless specific evidence is found to the
contrary.31
Print publications also tend to publish on a regular schedule and do not often change the
schedule (although such changes do occur occasionally), but schedules vary considerably
across publications, with dailies, annuals, and many types in between all common. Thus,
I only include a print publication if I find specific evidence that it was typically publishing
every two weeks or more often.32 If I find evidence that a print publication was publishing
30
If a radio or television channel initially broadcast from outside the country, I regard this as evidence
that it is not tolerated by the government. In such a case, I code the beginning of the channel’s broadcasting
as the banning event and code it as not permitted at the time it began.
31
In principle, I would include a station that regularly broadcast once every two weeks, even if skipped
one or two broadcasts during a year. In practice, I did not find any radio or TV stations that broadcast so
infrequently.
32
A print publication that skips one or two issues in a year but otherwise appears every two weeks is still
consider active.
88
actively in one year, I assume it was publishing frequently in previous and subsequent years,
as long as there is evidence that it was publishing at all, unless I find specific evidence that
the frequency changed before or after.
Websites often post irregularly within a given year and their frequency often varies
considerably over multiple years. If there is specific evidence on the frequency of posts in a
given year, a website is included if it posted at an average rate of at least once every 15
days, and excluded if it posted less frequently.33 To determine which posts had appeared
on a website’s homepage, I relied on the Internet Archive (the Wayback Machine).34 If no
evidence is available for a given year, the site is considered active during the year only if it
was posting frequently in the last year for which there was evidence and the first
subsequent year for which there was evidence. In all other years, a website is assumed to
33
I use fifteen days to make the standard equivalent to that applied to other media. If, for example, a
newspaper typically appeared once every two weeks, but skipped two of the possible 26 times it could have
published, then it published at rate of about once every 15 days, averaged over the year.
34
Available at http://web.archive.org. Because the Internet Archive records “captures” of sites irregularly
in most cases and often infrequently, it was not always possible to determine whether the site had posted at
an average rate of at least once every 15 days during that year (specifically the part of the year after the site
was launched and when it was not banned). In those cases, if site captures made it possible to determine the
number of posts during at least 90 days out of the year, then the number of posts during this period was used
to estimate the frequency of posting for the year. For example, suppose a site was only captured once in a
year, on June 30, and there were ten dated posts on the home page on that day, the oldest of which was from
Mar 1. (For simplicity, let’s assume the group was coded as significant for the entire year and the regime
did not change during the year.) Then we can infer that there were ten posts during the 122 days from
Mar 1 to June 30. Thus, the average posting rate was once every 12.2 days, which is sufficiently frequent to
qualify the site as active in that year. This estimate of the frequency of posting may be upwardly biased,
since users may record more site captures during periods in which the site is being updated more frequently.
If evidence was found that the site was down for part of the year, and there is no evidence that this was due
to government interference, then the average rate of posting will be adjusted to reflect the assumption that
there were no posts during that period. To continue the previous example, suppose captures on September
1 and November 30 both indicate that the site was down and there were either no captures in between these
dates or all captures also indicate that the site was down. We therefore assume that the site had no new
posts during the 91 days between September 1 and November 30, but was posting once every 12.2 days when
the site was not down. This frequency is applied to the 274 days that the site was assumed to be not down
(this may be an overestimate, if the site was down at other times that year that were not captured, or it may
be an underestimate if the site was not down at a time during September 1 to November 30 that was not
captured), minus any days before the site was launched and any days the site was banned. Call this number
of days in which the site had been launched, was not banned, and is assumed not to have been down X.
X
. In some cases, a site would
Thus, during the X + 91 days, the estimated posting frequency was: 12.2(X+91)
redirect to another, very similar site for part of the year. For the purpose of assessing posting frequency, I
treat the two sites in such a case as one media outlet, because the original site’s URL will lead a user to a
version of the site.
89
have posted infrequently.
I code a variable banned to record whether a banning, as defined above, occurred against at
least one media outlet of any of the significant opposition groups in a given regime-year.35
banned equals 1 if the government suspended or blocked any opposition outlet that either
was launched in that year or that the government previously allowed and that was actively
publishing up until the ban. banned also equals 1 if a new regime comes to power and
blocks or suspends an actively publishing outlet that began in a previous year.36 If the
government did not ban any opposition publication in the given year, then banned equals 0.
In order to identify years in which a ban could have taken place, can ban equals 1 if there was
any publication that could have been banned, i.e. if there had been an actively publishing
opposition outlet that was allowed as of the beginning of the year, if a new opposition outlet
was launched that year, or if a new regime came to power that year. If none of these
conditions holds, can ban equals 0.37
An opposition group’s media outlet is considered “suppressed” if, during the seven years
prior to the start of the year, it was banned and as of the beginning of the year it was still
not permitted. A group is considered “tolerated” in a given year if it actively published at
least one outlet, none of its outlets were suppressed at the start of the year, and none were
banned during the course of the year. tolerated equals 1 if at least one significant opposition
group was tolerated in the regime-year and equals 0 if none of the groups were tolerated.
tolerated all is more stringent: it equals 1 if every significant opposition group was tolerated
35
In years in which regimes changed, all variables in the dataset are coded only for the portion of the year
in which the regime that ruled for most of the year was in power.
36
The banning may last any amount of time. For instance, if the government prevented a newspaper from
distributing one issue, this would count as a banning. Censorship of specific articles or broadcast programs
does not count as a banning if the outlet is still able to publish or broadcast some content. Likewise, the
forced removal of a single post does not qualify as the banning of a website, but if a site is not allowed
to post any new articles for a period of time, or if access to the site is blocked within the country, that is
counted as a banning.
37
canban will not be used as a dependent variable, but rather to subset on cases in which banned could
equal 1.
90
and equals 0 if at least one group was not tolerated. When there is only one significant
opposition group, tolerated and tolerated all are necessarily equal. Among observations in
which there were two or more significant opposition groups, it turns out that tolerated all
equals 1 in only three regime-years, all from Yemen in the 1990s. Given this lack of variation,
I do not treat tolerated all as a dependent variable in the empirical analyses. Instead, I focus
on banned and tolerated.
tolerated may record some false negatives: it is possible that in a particular regime-year there
were no opposition media not because the government prevented groups from publishing, but
because groups lacked the resources to publish or chose not to do so. Nonetheless, we cannot
be sure that a regime is tolerant toward opposition media if there are no such media. For
example, Sudan’s Umma Party closed its weekly newspaper Sawt Al-Umma in 2012, due to
financial pressures. While the government did not directly shut the paper down, it seems
likely that its closure was a result of the government’s decision that year to impose a tax
on printing after the press widely reported on anti-government protests (Freedom House
2013a). The tolerated variable puts the burden of proof on declaring that the government
does tolerate opposition media. By contrast, because banned equals 1 only when I find direct
evidence of regime interference with the media, it puts the burden of proof on declaring that
the government does not tolerate opposition media.38 For this reason, I use both variables
to test my theory.
As an illustration of how I coded a wide range of media and varying levels of government
tolerance over time, consider Jordan, which had a main opposition throughout 1992-2017, the
Muslim Brotherhood. In 1991, the group launched a weekly newspaper, Ar-Ribat. Although
an issue was seized in that first year (which does not appear in the dataset), the government
did not interfere with it subsequently. In 1993, the Brotherhood replaced Ar-Ribat with a
38
In the Sudan example, banned equals 0 in 2012 and tolerated equals 1, but tolerated equals 0 in every
year after 2012.
91
new weekly, As-Sabeel, which became a daily in 2009. As-Sabeel was published through 2017
without interference. An associated website, AsSabeel.net, was never blocked in this period.
The Brotherhood launched www.IkhwanJo.com in 2002, and the group’s political party, the
Islamic Action Front, launched www.Jabha.net in 2004. From 1992-2006, tolerated is coded
as a 1, but in January 2007, the government blocked access to Jabha.net inside Jordan,
so tolerated equals 0 in that year. In 2008, 2009, and 2010, Jordan again gets a 1 for
tolerated, but in 2011, regime supporters hacked Jabha.net and IkhwanJo.com. Moreover,
the Muslim Brotherhood launched a satellite TV channel, Al-Yarmouk, that year, but the
government initially did not permit in Jordan, forcing the channel to broadcast from Bahrain.
tolerated is therefore 0 in 2011, as well as 2012, when Al-Yarmouk remained shut out of the
country. In 2013, the government allowed the channel to begin broadcasting from Jordan,
but later that year suspended the channel for five days, so I again code tolerated as a 0.
The government did not interfere with any of the Brotherhood’s media in 2014 (tolerated
equals 1), but in 2015, the government shut down the channel’s studios, temporarily forcing
the channel off the air. In that year, tolerated therefore equals 0. It remains 0 in 2016
and 2017. Although Al-Yarmouk began broadcasting again later in 2015 with the help of
some Jordanian companies, the following year the government pressured these companies to
stop working with Al-Yarmouk, forcing the channel to resume broadcasting from outside of
Jordan. banned equals 1 in 2007, 2011, 2013, 2015, and 2016. Although tolerated equals
0 in 2012 and 2017, banned does not equal 1 in these years, because Al-Yarmouk simply
remained not permitted.
In the Jordanian example, banned = 1 necessarily implies tolerated = 0, because there is only
one significant opposition group. In cases with more than one significant opposition group, it
is possible for both banned and tolerated to equal 1 in the same regime year, if the media of
one group is banned but the media of another group is tolerated. For instance, the Egyptian
government blocked April 6’s website, 6april.org in 2017, but continued to allow Egyptian
92
citizens to access the Salafist Front’s only outlet, gabhasalafia.com. However, there are only
three observations of this type, i.e. in which banned and tolerated both equal 1. There are,
by contrast, 85 observations in which both variables equal 0, so the two variables are far
from being perfectly negatively correlated. A regime-year in which there were no banning
events, but which is not coded as one of toleration can occur, for instance, if the government
banned all opposition media outlets in the prior year, and in the current year it does not
allow any of them to resume publishing. In the current year, banned may equal 0 (if no
opposition group tried to launch a new publication that the regime could have banned), but
tolerated also equals 0, because there were no opposition media currently publishing. It is
also possible for such cases to arise even if there has been no banning event in recent years,
but no opposition groups are currently publishing any media. Of course, it is possible that
all groups are refraining from publishing voluntarily, without any government pressure, and
that the regime would have tolerated their media if the groups had chosen to produce any.
It is also possible that the government would not have tolerated any attempts to launch
opposition media.39 I have chosen to handle such ambiguous observations by coding them
as a 0 for tolerated. In other words, my rules require that, in order to code a regime-year as
a 1 for tolerated, there must be tangible evidence of toleration, in the form of at least one
actively publishing opposition media outlet.
Appendix 4 contains narratives on the media of all opposition groups in the dataset. Table
2 provides summary statistics for banned and tolerated. The middle column shows the
percentage of all regime-years in which each variable equalled 1. In the case of banned, this
value is calculated only over the 130 observations for which can ban equals 1, while the share
for tolerated is calculated over all 202 observations in the dataset. In typical years, regimes
did not ban any opposition media: bans occurred in less than 17% of regime-years in which
39
The government may also have deterred opposition groups from trying to publish any media by privately
intimidating them, so that banned equals 0 only because the government’s actions to prevent opposition media
were not publicly observable.
93
Table 4.2: Bans and Toleration of Opposition Media
Variable = 1
Observations
banned
16.9%
130
tolerated
50.5%
202
there was any opposition media that could have been banned. In about half of regime-years
governments tolerated the media of at least one opposition group.
Figures 2-4 show how tolerated varies over time within regimes. In some regimes tolerated
remains constant at 0 or 1, but in most regimes there is notable variation. Figures 5-8
show the incidence of banning events over time with regimes. Blue bars denote years in
which a regime could have banned some opposition media (can ban=1) but did not
(banned =0). Red bars denote years in which the regime did ban at least one opposition
publication (can ban=banned =1). In years without a bar there was no opposition media to
ban (can ban=0). Mauritania 1978-2005 and Tunisia 1956-2011 are not pictured, because
can ban was 0 during every year that those regimes appeared in the dataset.
4.3
Explanatory Variables
In this section, I describe the focal explanatory variables I use in the empirical analyses, as
well as the control variables.
To test Hypothesis 1, I include the national unemployment rate (on a 0 to 100 scale), lagged
by one year, L unemployment. All variables that begin with “L ” are lagged by one year. I
lag the variables in order to avoid spurious findings due to reverse causality. Specifically,
I use the International Labor Organization modeled unemployment rate, from the World
94
Figure 4.2: tolerated over time: Algeria-Egypt
95
Figure 4.3: tolerated over time: Jordan-Morocco
96
Figure 4.4: tolerated over time: Sudan-Yemen
97
Figure 4.5: banned over time: Algeria-Egypt
98
Figure 4.6: banned over time: Egypt-Jordan
99
Figure 4.7: banned over time: Kuwait-Morocco
100
Figure 4.8: banned over time: Sudan-Yemen
101
Bank (2019).
I include annual growth in per capita GDP (lagged by one year), to test Hypothesis 2:
L GDP pc growth.
To test Hypothesis 3, I include L fuel, the per capita annual revenue from all oil and natural
gas production, in 2018 USD, lagged by one year. This variable is the sum of L oil and
L gas, the per capita revenue from oil and natural gas, respectively. Each of these variables
was constructed by multiplying total production from the respective fuel sources by their
prices at the time (obtained from BP 2019)40 and dividing by total population (World Bank
2019).41 Some Arab countries that have never produced more than negligible amounts of
oil or natural gas do not appear in the respective BP datasets. Countries that are in the
dataset have missing values for years in which their production was below certain very
small thresholds. For countries that do not appear in a dataset I assign values of 0 on the
corresponding variable (L oil or L gas) for all years , and I also assign zeros for missing values
in the dataset.42 Since I will use regime fixed-effects, regimes that have a 0 on L fuel for all
years do not contribute to the estimation of the coefficient on that variable, since they have
no within-unit variation on it (Mummolo and Peterson 2018: 829).
In order to test Hypothesis 4, I use two rough proxies for the regime leader’s personal
control over the security apparatus. Not all dictatorships become highly personalized, but
40
The data from BP 2019 are available at:
https://www.bp.com/en/global/corporate/energyeconomics/statistical-review-of-world-energy.html.
41
BP 2019 uses the Brent dated price of crude oil as its benchmark for world oil prices. Natural gas prices
vary more than oil prices across the globe (U.S. Energy Information Administration 2011), so the natural gas
component is probably noisier than the oil component. I use the U.S. Henry Hub price for natural gas. Since
BP (2019) records gas prices in terms of US dollars per British thermal unit (Btu) of gas, but records gas
production in terms of cubic feet, I used U.S. Energy Information Administration (2019) to convert between
the two measures of gas production. Also, the price was listed in constant dollars, so I converted to 2018
dollars using deflator data from U.S. Bureau of Economic Analysis (2019).
42
There are 95 regime-years in my dataset in which Lfuel equals 0, 111 in which Loil equals 0, and 121 in
which Lgas equals 0.
102
those that do tend to become more personalized the longer an individual leader has been
in power (Geddes, Wright, and Frantz 2018: 87).43 leader year office measures the number
of years that the de facto regime leader has been in power. It equals 1 during the year the
leader takes power, etc. In years in which there is a leader change, this variable refers to the
leader who was in power for most of the year.44 Founding leaders of regimes are also more
likely to personalize power than their successors (Geddes, Wright, and Frantz 2018: 86) and
so presumably likelier to have firm control of the security forces.45 regime first leader equals
1 if the regime leader was the first leader in the regime and 0 if he took power after the
regime began.46 These leader characteristic variables are based on my codings of de facto
regime leaders, which mostly match those of Goemans, Gleditsch, and Chiozza (2009)47 and
Geddes, Wright, and Frantz (2014), but depart from them in a few cases. See Appendix 5 for
a list of regime leaders. Where my coding decisions differ from either Goemans, Gleditsch,
and Chiozza (2009) or Geddes, Wright, and Frantz (2014), or for regime-years that were not
in either of their datasets, I provide the sources I used to code them.
I also include three period dummies — yrs00 05, yrs06 11, and yrs12 17 — each corresponding
to a six-year period, with the reference category being the first eight years, 1992-1999. These
dummies roughly capture the effects of time-varying factors that affected the whole region,
43
It could also be that more personalist dictators are less likely to lose power (Svolik 2012: 77-78) and so
the relationship between personalism and time in power could be in part a selection effect. However, Geddes,
Wright, and Frantz (2018: 87-88) still find that personalism rises the longer a leader has been in power when
they control for leader fixed effects. In any case, in selecting time in power as a proxy, the important thing
is that it is correlated with personalism.
44
There were no cases in the covered regimes and years in which no single leader was in power for the
majority of the year.
45
The Geddes, Wright, and Frantz (2018) measure of personalism combines indicators of personal control
of the security forces with control over the ruling party and high office generally.
46
The dataset of Geddes, Wright, and Frantz (2014) includes regime-year indicators that measure aspects
of personalism more directly, including whether the leader has direct control over the security apparatus.
I applied their coding rules for several of these indicators to the regime-years in my dataset that were not
included in theirs. However, I found almost no variation in these indicators within each regime over time,
during the period each regime appeared in my dataset, except for a couple regimes. Thus, it was infeasible to
use these measures of personalism in regime fixed-effects regressions, as any results would be driven almost
entirely by one or two regimes.
47
The
Archigos
dataset
of
government
leaders
can
be
found
at:
https://www.rochester.edu/college/faculty/hgoemans/data.htm.
103
including the launching of Al Jazeera, the invasion of Iraq, and the emergence of social
media.48 Hypothesis 5 predicts that governments will be less likely to tolerate opposition
media after a successful uprising against other authoritarian regimes in the same region.
Unarmed protestors brought down the regimes in Tunisia and Egypt in early 2011, so I will
test whether there is a post-“Arab Spring” effect using the yrs12 17 indicator. After all, the
apparent contagion of the protests in Tunisia and Egypt to other Arab countries suggests
that citizens across the region became more optimistic about their chances of bringing about
political change through protest, given the social, cultural, and political parallels between
those two countries and their own (Bellin 2012: 141-142).49
I also include other factors that are likely to affect government toleration of opposition media
as controls. Electoral authoritarian regimes often use control of the media to ensure their
victories at the polls (Schedler [2002: 43]).50 Also, elections can provide focal moments
around which to coordinate protests (Stein 2016). pres elec and leg elec are dummy variables
that equal 1 if there was an election for president or for the national legislature, respectively.51
I also include a dummy variable, leader fail, based on my coding of de facto regime leaders,
that equals 1 in years in which there was a change of regime leadership, since either new
leaders or leaders at acute risk of overthrow may behave differently than leaders in normal
years. Since Whitten-Woodring and Van Belle (2014) find that media freedom tends to
decline in countries experiencing civil wars, I include an indicator for civil war, lagged by
one year L civil war. This variable is derived from data produced by Gleditsch et al (2002)52 :
48
Inclusion of time period effects also helps to control for measurement error in my coding that is correlated
with time, e.g. if the accuracy of the sources I consulted improved over time.
49
The overthrow of Mubarak may also have inspired subsequent protests in Egypt against the SCAF and
Sisi.
50
On the other hand, governments sometimes let up on controls of the media in order to make elections
more credible. For example, the U.S. Department of State (1998) reports that, in the run-up to Mauritania’s
1997 presidential election, “the Government gave all five candidates equal access to its two newspapers and
to the electronic media; for the first time, citizens heard or read criticisms of both the Government and the
President in these media and not just in the private newspapers.”
51
These variables are coded based on Europa World (2019a, 2019b, 2019d, 2019e, 2019t, 2019u, 2019v,
2019w, 2020), Inter-Parliamentary Union (2019), and International Foundation for Electoral Systems (2019).
52
The UCDP/PRIO Armed Conflict Dataset is available at www.prio.org/Data/Armed-Conflict/UCDP-
104
Table 4.3: Explanatory Variables (continuous and integer)
Variable
Mean
Min
Max
Stand. Dev.
L unemployment
11.5
0.7
31.8
6.0
L GDP pc growth
1.4
-7.9
12.2
3.1
L fuel
1,726.5
0
38,079.4
5607.6
L oil
1,448.0
0
37,643.0
5493.5
L gas
278.6
0
4,279.3
672.5
leader year office
15.2
1
47
10.6
groups N
1.6
1
5
1.1
it equals 1 if there was either an “internal armed conflict” or an “internationalized internal
armed conflict” in which more than 1,000 people died in battle. Lastly, I include two variables
based on my coding of opposition groups. is main equals 1 if there was a “main” opposition
group and 0 otherwise. groups N equals the total number of significant opposition groups.53
Table 3 provides summary statistics for the non-binary explanatory variables and Table 4
gives the means of all binary explanatory variables. There are no missing values for any of
the explanatory variables.
As the histograms in Appendix 6 show, each of the continuous variables exhibits rightskewing.54 I therefore use the natural logarithm of each of the non-negative continuous
variables, adding a 1 to each of the fuel revenue variables, since these can take values of
0. Because income growth takes negative values for some observations, I opt for a different
transformation that also reduces the influence of extreme values. For positive growth values,
this transformed variable, sqrt L GDP pc growth equals the square root of L GDP pc growth .
√
For negative growth values, sqrt L GDP pc growth equals: − −L GDP pc growth.
PRIO and has been updated to cover the period through 2018.
53
groupsN is always greater than or equal to 1: if there were no significant opposition groups, the regimeyear was not included in the dataset. groupsN always equals 1 if ismain equals 1.
54
Appendix 6 also presents histograms for Lfuel, Loil, and Lgas with values of 0 dropped. Even among
these subsets, each variable is highly right-skewed.
105
Table 4.4: Explanatory Variables (binary)
4.4
Variable
Mean
regime first leader
0.18
yrs00 05
0.25
yrs06 11
0.23
yrs12 17
0.23
pres elec
0.11
leg elec
0.22
leader fail
0.04
L civil war
0.09
is main
0.57
Methodology
To test the effects of the focal variables on toleration of opposition media, I estimate linear
probability models in which tolerated and banned are the dependent variables, with fixed
effects by regime. There are many important differences between regimes that could confound
the estimation of the effects of the focal variables on toleration of opposition media, including
some I may not even be aware of and hence do not measure. Using fixed effects is useful
for making causal inferences from observational data (Brüderl and Ludwig 2014: 327). By
including regime fixed effects, specifically, we control for unmeasured variables that vary
across regimes but not within regimes over time (Allison 2009: 4). Such regime-specific
factors might include, for example, geography, colonial history, how the regime came to
power, or the group from which regime leaders are drawn55 (Derpanopoulos et al 2017: 2).
Such unobserved heterogeneity may be correlated with the explanatory variables (Allison
2009: 4) or with the selection of cases into and out of the dataset (Mummolo and Peterson
2018: 829), introducing bias if we did not use fixed effects.
55
If leaders start coming from a different group than before, a new regime has by definition begun (Geddes,
Wright, and Frantz 2018: 5).
106
Of course, using fixed regime effects cannot control for unobserved variables that change over
time (Allison 2009: 4). The period dummies help to control for unobserved time-varying
factors that affect countries across the region, as discussed in the previous section. Thus,
whether the results for the focal explanatory variables are unbiased depends on whether
I have included as controls all factors that vary from year to year within regime, which
substantially affect government toleration of opposition media, and are correlated with either
the focal variables or with the selection of regimes into and out of the dataset.
Although Horrace and Oaxaca (2006) argue that the linear probability model frequently
provides biased and inconsistent estimates of the effects of explanatory variables, Kwak,
Martin, and Woolridge (2018: 17) argue that a fixed-effects linear probability model can
produce less biased estimates than a conditional logit regression in the presence of serial
correlation, particularly when there are a small number of observations for each unit. Some
of the regimes do have only a few years in the dataset, and Breusch-Godfrey/Wooldridge
tests indicate serial correlation for both tolerated and banned.56 Nonetheless, I will compare
the results from the linear models with results from logit models. In order to account for this
non-independence of results within regimes, I cluster standard errors by regime (Cameron
and Miller 2015).57
56
I test for second-order serial correlation, since two years is the minimum length of time that any regime
appears in the dataset. The test of the null hypothesis of no serial correlation is rejected at the 1% level for
tolerated and at the 5% level for banned.
57
I use “HC2” type clustering which are unbiased when errors are homoskedastic (Cameron and Miller
2015: 341). Studentized Breusch-Pagan tests did not indicate evidence of heteroskedasticity for either
tolerated or banned.
107
4.5
Main Findings
Table 5 presents the main results for the focal variables. The findings in Table 5 are
consistent with Hypotheses 1-5 from Chapter 2. Supporting Hypothesis 1, the rate of
unemployment (L unemployment) is negatively correlated with tolerated, with the effect
significant at the 1% level . Providing partial support for Hypothesis 2, growth in per
capita GDP (sqrt L GDP pc growth) is positively correlated with tolerated. However, the
effect is only significant at the 10% level. Consistent with Hypothesis 3, per capita revenue
from oil and natural gas production (L fuel ) is positively correlated with the dependent
variable, with the coefficient significant at the 1% level. The number of years the regime
leader has been in office (leader year office) and whether the leader is the founding leader of
the regime (regime first leader ) are proxies for the regime leader’s personal control over the
security apparatus. Here we see partial support for Hypothesis 4: the leader’s time in office
is positively correlated with tolerated, albeit only at the 10% level of significance, while
being the first leader of the regime is not statistically significant.
Consistent with
Hypothesis 5, the period after the “Arab Spring” uprisings, 2012-2017 (yrs12 17 ), is
negatively correlated with tolerated (relative to the years 1992-1999), significant at the 1%
level. The period 2006-2011 is also negatively correlated with tolerated (significant at the
5% level), although the effect size is smaller.
Table 6 shows the results for the control variables. Presidential election years (pres elec) are
positively associated with toleration of opposition media, but the effect is only significant at
the 10% level. Legislative election years (leg elec), years in which the de facto regime leader
changes (leader fail ), and years following civil wars (L civil war ) do not have a significantly
higher or lower level probability of toleration of opposition media, all else equal.58 The
58
The null result for leaderfail does not necessarily mean that changes in leadership have no effect on the
existence of opposition media: leadership changes are generally associated with a decline in leaderyearoffice,
which predicts a decline in the probability of tolerance of opposition media. The null result simply indicates
108
government is also more likely to tolerate opposition media when there is a clearly identifiable
“main” opposition group, with is main positively correlated with the dependent variable at
the 1% level. Perhaps not surprisingly, the greater the number of significant opposition
groups, the more likely it is that at least one of them will be allowed to have media: groups N
is positively correlated with tolerated, significant at the 1% level.
Figure 9 shows the histograms of the residuals. The residuals appear roughly normal, albeit
with some right-skewing. Figure 10 displays the influence of each observation (regime-year)
on the model. Each circle represents one regime-year, with the size of each circle proportional
to the Cook’s distance, a measure of the observation’s influence. The five most influential
observations are labeled in the plot, and are all from either the 1992-NA regime in Algeria
or the 2011-2012 regime in Egypt. To see whether any single regime is driving the results,
I re-estimated the model dropping one regime at a time.59 Tables showing the results with
each regime excluded are shown in Appendix 7.
Some of the results that appear in Tables 5 and 6 disappear or weaken when specific regimes
are dropped. Consider first the focal explanatory variables. Dropping the Jordanian regime
(1946-NA) causes economic growth to lose significance. leader year office is not significant if
we exclude regimes from Egypt (1952-2011), Jordan (1946-NA), Kuwait (1961-NA), Morocco
(1956-NA), or Yemen (1978-2012). The effect for yrs12 17 is only significant at the 10% level
if the Kuwaiti regime is dropped (and the magnitude is reduced by more than half).
Turning to the non-focal variables, dropping certain regimes causes some unexpected results
to go away. Dropping the Jordanian or Kuwaiti regimes causes yrs06 11 to lose significance.
pres elec is not significant if we drop either Egypt 2011-2012, Jordan 1946-NA, Mauritania
that there is not a departure in this trend for the first or final year a leader is in office, on average.
59
In most cases, this amounted to dropping all observations from a single country. Recall, however, that
there are two Mauritanian regimes in the dataset and three Egyptian ones.
109
Table 4.5: Tolerated Opposition Media (Regime FE)
Dependent variable:
tolerated
log(L unemployment)
−0.984∗∗∗
(0.142)
sqrt L GDP pc growth
0.042∗
(0.021)
log(L fuel + 1)
0.366∗∗∗
(0.078)
log(leader year office)
0.084∗
(0.047)
regime first leader
−0.083
(0.309)
yrs00 05
−0.018
(0.084)
yrs06 11
−0.253∗∗
(0.099)
yrs12 17
−0.462∗∗∗
(0.097)
Observations
R
202
2
0.369
Adjusted R2
Note:
0.276
∗
p<0.1;
110
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.6: Tolerated Opposition Media (Regime FE), continued
Dependent variable:
tolerated
pres elec
0.159∗
(0.091)
leg elec
−0.080
(0.064)
leader fail
0.118
(0.148)
L civil war
0.225
(0.141)
is main
0.333∗∗∗
(0.122)
groups N
0.303∗∗∗
(0.052)
Observations
202
R2
0.369
Adjusted R2
0.276
Note:
∗
p<0.1;
∗∗
111
p<0.05;
∗∗∗
p<0.01
Figure 4.9: Residuals
Figure 4.10: Cook’s Distances
112
1978-2005, Mauritania 2008-NA, Sudan 1989-NA, Tunisia 1956-2011, or Yemen 1978-2012.
Finally, is main is not significant if we drop Yemen 1978-2012. The positive association of
groups N with tolerated remains significant at the 1% level with each regime excluded.
Thus, there are three results that are relatively robust and relate to the central theory
of this study. First, when unemployment is low or fuel revenue is high, governments are
more likely to allow opposition media, providing support for Hypothesis 1 and Hypothesis
3. It also appears that regimes were less likely to allow opposition media after 2011, all else
equal, providing support for Hypothesis 5.60 As predicted by the theory, regimes that can
be expected to face the greatest threat of unrest are least likely to allow opposition media,
while stronger regimes are more likely to allow opposition media.
Figures 11 and 12 show partial regression plots for unemployment and fuel revenue. These
plots highlight that no individual observation or cluster of observations seem to account for
these results.
The estimated effect sizes of unemployment, fuel revenue, and the post-2011 period are quite
large. Because unemployment and fuel revenue are logged, we must interpret the effects in
terms of increases in each predictor by a given ratio. For a point of reference, for each
predictor, I take the ratio of the maximum value to the minimum across all observations for
each regime, and then estimate the effect of increasing the predictor by the median of these
ratios. For unemployment, the median max:min ratio is 1.4. The median regime is the one
that existed in Egypt from 1952-2011, which is in the dataset for the years 1992-2010, and
since unemployment is lagged, has unemployment data from 1991-2009. During this time,
Egypt’s absolute level of unemployment varied between 7.9% and 11.2%.61 Within a given
60
Four of the regimes in the dataset ended in 2011 or 2012, but because regime fixed effects allow us to
examine changes within regimes over time, this “Arab Spring” effect cannot be driven by the selection of
these regimes out of the dataset.
61
The smallest ratio of maximum to minimum unemployment observed is for Tunisia, during the years
113
Figure 4.11: Partial regression plot: unemployment
Figure 4.12: Partial regression plot: fuel revenue per capita
114
Table 4.7: Effect sizes of focal variables
full sample effect
minimum effect
L unemployment
-33.1%
-17.2%
L fuel
35.0%
28.9%
yrs12 17
-46.2%
-20.6%
regime, the results predict that increasing the unemployment rate by this ratio will lead to a
33.1% decrease in the probability that any opposition media will be tolerated.62 Of course,
the estimated effect of unemployment varies notably when some regimes are excluded. The
smallest point estimate is observed when Kuwait is excluded, in which case the estimated
effect of increasing unemployment by a factor of 1.4 is a 17.2% decrease in the probability
that any opposition media are tolerated. Although this estimate is much smaller, it is still a
substantial effect. The median max:min ratio of fuel revenue per capita (plus $1) is 2.6. The
median regime is Tunisia, which had a max:min ratio of about 2.6. In absolute terms, Tunisia
in the 1991-1998 period saw a minimum per capita fuel revenue of $55 and a maximum of
$144.63 The estimated effect of increasing per capita fuel revenue by this ratio is a 35%
increase in the probability that a government tolerates some opposition media. The smallest
estimated effect of fuel on tolerated is 28.9%, obtained when dropping Egypt 1952-2011. The
probability that a regime would tolerate any opposition media is estimated to have fallen
by 46.2% after 2011, with a minimum estimate of 20.6% when Kuwait is excluded. Table 7
summarizes the estimated effects.
It is possible that, in some cases in which an opposition group did not produce any media, the
1991-1998. It has a max:min ratio of approximately 1.06. Its minimum unemployment rate was 15.1%
and the maximum was 16.0%. At the other end of the spectrum is Kuwait (over the years 1996-2002 and
2010-2016), with a max:min ratio of about 4.1. Kuwait’s absolute levels of unemployment are much lower
than Tunisia’s however, varying between 0.7% and 2.9%.
62
The coefficient on Lunemployment is -0.984: multiplying that coefficient by the natural logarithm of 1.4
yields approximately -0.331.
63
Five regimes, as noted above, did not have any recorded fuel revenues, so their max:min ratios (after
offsetting their observed values by $1) are all 1. The maximum ratio was 24.1, for Sudan (observed 19992016), which had a minimum per capita revenue of $20 and a maximum of $505.
115
government did not prevent them from doing so, nor were they deterred by the expectation
that the government would stop them. Instead, a group may have simply lacked the resources
or chose not to publish. To further test the theory that regimes threatened with unrest are
more likely to prevent opposition groups from publishing, we can use the variable banned
to examine when regimes actively shut down some opposition media. I subset on regimeyears in which there was some opposition media outlet — or an attempt to start one —
that the regime could have intervened to stop. That is, I focus on those regime-years in
which can ban equals 1. These include both years that began with some opposition group
publishing an outlet that the regime (or the previous regime) had tolerated up to that point
and those years in which an opposition group began publishing a new outlet. Subsetting
in this way shrinks the sample size by quite a bit, from 202 to 130 regime-years, reducing
the statistical power of the estimations. And whereas the tolerated variable runs the risk of
labelling regimes as not tolerant in cases when they would have allowed opposition media
if the opposition had tried to publish, the banned variables runs the opposite risk: labeling
cases in which the threat of interference deterred a group from even trying to publish media
as instances without government interference with opposition media. If we find support for
the theory using both measures, we have all the more reason to be confident in the results.
The explanatory variables are the same as in the previous models with one exception: I
break the indicator for the period 2006-2011 into two, one for 2006-2010 (yrs06 10 ) and one
just for the year 2011 (yr11 ). The reason is that banned measures a discrete event — a
regime receives a 1 only in the year in which a publication is first suspended or blocked —
and the “Arab Spring ” protests, which began in Tunisia in December 2010, may have led
governments to crack down on their opposition media before the end of 2011.
The results are consistent with the theory: governments are more likely to ban existing
opposition media when they face a greater threat of unrest. The higher the rate of per
116
capita GDP growth, the less likely governments are to ban opposition media, with
sqrt L GDP pc growth negatively correlated with banned and significant at the 1% level. This
finding is consistent with Hypothesis 2. Governments are also less likely to ban opposition
media when oil and gas revenues are higher: L fuel also has a negative effect, significant at
the 1% level, consistent with Hypothesis 3.
Governments were more likely to ban
opposition media in 2011, with the effect of yr2011 positive and significant at the 5% level,
supporting Hypothesis 5. However, unemployment does not have a significant effect on the
likelihood that governments ban opposition media, contrary to Hypothesis 1, nor did leader
characteristics (leader year office and regime first leader ) have any effect, contrary to
Hypothesis 4.
To test the robustness of these results, I again re-estimated the model dropping one regime
at a time. The full results with each regime dropped are shown in Appendix 8. The negative
effects of income growth and fuel revenue are highly robust, remaining statistically significant
at the 1% or 5% levels in all cases. However, the estimated effect of the year 2011 on bans,
while remaining positive, loses statistical significance if we drop either Jordan (1946-NA) or
Yemen (1978-2012). Thus, looking at cases in which governments banned existing opposition
media lends partial support to the theory: bans are less common when governments preside
over stronger economies.
4.6
Alternative Specifications
As further robustness checks, I also tested the hypotheses using different specifications of
the models, as well as an alternative version of the dataset baed on a different way of coding
“main” opposition groups. First, I re-estimated the models using country fixed-effects rather
117
Table 4.8: Opposition Media Bans (Regime FE)
Dependent variable:
banned
log(L unemployment)
0.378
(0.246)
sqrt L GDP pc growth
−0.083∗∗∗
(0.026)
log(L fuel + 1)
−0.473∗∗∗
(0.146)
log(leader year office)
−0.073
(0.055)
regime first leader
0.043
(0.319)
yrs00 05
0.042
(0.139)
yrs06 10
0.262∗
(0.150)
yr11
0.442∗∗
(0.198)
yrs12 17
0.114
(0.147)
Observations
R
130
2
Adjusted R
Note:
0.282
2
118
∗
p<0.1;
0.109
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.9: Opposition Media Bans (Regime FE), continued
Dependent variable:
banned
pres elec
−0.071
(0.127)
leg elec
0.072
(0.075)
leader fail
0.175
(0.179)
L civil war
−0.267
(0.195)
is main
−0.419∗∗
(0.206)
groups N
−0.226∗∗∗
(0.080)
Observations
130
R2
0.282
Adjusted R2
0.109
Note:
∗
p<0.1;
∗∗
119
p<0.05;
∗∗∗
p<0.01
than regime fixed effects. Most countries in the dataset are associated with a single regime,
but there are two regimes in the dataset from Mauritania (1978-2005 and 2008-NA) and
three from Egypt (1952-2011, 2011-2012, and 2013-NA). Nonetheless, using country fixed
effects does not change the main results, as can be seen in Tables 10 and 11 Unemployment
is negatively associated with tolerated, while fuel revenue is associated positively, and the
post-2011 period is associated negatively, all at the 1% level. Likewise, using country fixed
effects does not change the estimated effects of growth or fuel revenue on the probability
that governments will ban existing opposition media, as shown in Table 12.
Second, I disaggregate fuel revenue into its two component sources: oil and natural gas.
Cameron and Stanley (2017: 39) note that oil and natural gas differ in their transportation,
financial risks, and infrastructural requirements. So it is possible that their political effects
differ.64 Tables 14-17 present the results with lagged oil and natural gas revenue per capita
(L oil and L gas) disaggregated. Oil and natural gas both have positive effects on tolerated,
significant at the 5% and 1% levels. The estimated effects of unemployment and the post2011 period are not substantially changed by this alternative specification. The estimated
effects of oil and gas on banned are both negative. However, the coefficient on oil is not
significant, while the coefficient on natural gas is only significant at the 10% level. The effect
of growth on banned is not substantially altered.
I also estimated logistic regression models, with all the same explanatory variables, as well
as fixed regime effects and standard errors clustered by regime. As Beck (2020) notes, logit
models with fixed effects drop any units that have no variation on the dependent variable.
Thus, to compare a linear probability model with a logit model, the linear probability model
should also be estimated on the subset of units with variation on the dependent variable.
Tables 18 and 19 show results for linear and logit models of tolerated based only on the
64
I thank Yuree Noh for raising this point.
120
Table 4.10: Tolerated Opposition Media (Country FE)
Dependent variable:
tolerated
log(L unemployment)
−0.888∗∗∗
(0.142)
sqrt L GDP pc growth
0.037∗
(0.022)
log(L fuel + 1)
0.428∗∗∗
(0.078)
log(leader year office)
0.052
(0.048)
regime first leader
0.331∗
(0.177)
yrs00 05
−0.008
(0.086)
yrs06 11
−0.244∗∗
(0.099)
yrs12 17
−0.375∗∗∗
(0.094)
Observations
R
202
2
0.371
Adjusted R2
Note:
0.289
∗
p<0.1;
121
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.11: Tolerated Opposition Media (Country FE), continued
Dependent variable:
tolerated
pres elec
0.170∗
(0.093)
leg elec
−0.062
(0.065)
leader fail
0.156
(0.148)
L civil war
0.248∗
(0.142)
is main
0.193
(0.118)
groups N
0.259∗∗∗
(0.051)
Observations
202
R2
0.371
Adjusted R2
0.289
Note:
∗
p<0.1;
∗∗
122
p<0.05;
∗∗∗
p<0.01
Table 4.12: Opposition Media Bans (Country FE)
Dependent variable:
banned
log(L unemployment)
0.315
(0.242)
sqrt L GDP pc growth
−0.083∗∗∗
(0.026)
log(L fuel + 1)
−0.553∗∗∗
(0.140)
log(leader year office)
−0.057
(0.054)
regime first leader
−0.237
(0.238)
yrs00 05
0.062
(0.135)
yrs06 10
0.279∗
(0.144)
yr11
0.377∗∗
(0.189)
yrs12 17
0.087
(0.127)
Observations
R
130
2
Adjusted R
Note:
0.267
2
123
∗
p<0.1;
0.108
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.13: Opposition Media Bans (Country FE), continued
Dependent variable:
banned
pres elec
−0.116
(0.125)
leg elec
0.055
(0.075)
leader fail
0.139
(0.179)
L civil war
−0.304
(0.195)
is main
−0.400∗∗
(0.197)
groups N
−0.199∗∗
(0.079)
Observations
130
R2
0.267
Adjusted R2
0.108
Note:
∗
p<0.1;
∗∗
124
p<0.05;
∗∗∗
p<0.01
Table 4.14: Tolerated Opposition Media (Regime FE), disaggregated fuel
Dependent variable:
tolerated
log(L unemployment)
−0.964∗∗∗
(0.142)
sqrt L GDP pc growth
0.042∗
(0.021)
log(L oil + 1)
0.177∗∗
(0.086)
log(L gas + 1)
0.325∗∗∗
(0.080)
log(leader year office)
0.056
(0.048)
regime first leader
−0.126
(0.306)
yrs00 05
−0.115
(0.089)
yrs06 11
−0.312∗∗∗
(0.102)
yrs12 17
−0.476∗∗∗
(0.098)
Observations
R
202
2
Adjusted R
Note:
0.379
2
125
∗
p<0.1;
0.282
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.15: Tolerated Opposition Media (Regime FE), disaggregated fuel, continued
Dependent variable:
tolerated
pres elec
0.140
(0.091)
leg elec
−0.089
(0.064)
leader fail
0.080
(0.148)
L civil war
0.127
(0.145)
is main
0.371∗∗∗
(0.122)
groups N
0.319∗∗∗
(0.052)
Observations
202
R2
0.379
Adjusted R2
0.282
Note:
∗
p<0.1;
∗∗
126
p<0.05;
∗∗∗
p<0.01
Table 4.16: Opposition Media Bans (Regime FE), disaggregated fuel
Dependent variable:
banned
log(L unemployment)
0.491∗∗
(0.245)
sqrt L GDP pc growth
−0.077∗∗∗
(0.026)
log(L oil + 1)
−0.231
(0.196)
log(L gas + 1)
−0.275∗
(0.145)
log(leader year office)
−0.076
(0.056)
regime first leader
0.018
(0.326)
yrs00 05
0.042
(0.141)
yrs06 10
0.278∗
(0.154)
yr11
0.471∗∗
(0.202)
yrs12 17
0.155
(0.151)
127
Observations
R2
130
0.271
Table 4.17: Opposition Media Bans (Regime FE), disaggregated fuel, continued
Dependent variable:
banned
pres elec
−0.076
(0.128)
leg elec
0.073
(0.076)
leader fail
0.188
(0.181)
L civil war
−0.264
(0.200)
is main
−0.439∗∗
(0.211)
groups N
−0.226∗∗∗
(0.082)
Observations
130
R2
0.271
Adjusted R2
0.087
Note:
∗
p<0.1;
∗∗
128
p<0.05;
∗∗∗
p<0.01
regimes with changing values of tolerated.65 As we can see in the first column, subsetting the
data in this way does not notably change estimates of unemployment, fuel, and the post-2011
period, and all remain highly significant. Moving to the results from the logit model in the
second column, we see the signs and significance of the coefficients on these variables remain
the same. The logit results actually provide additional evidence in favor of the theory, with
the positive effect of economic growth significant at the 1% level, consistent with Hypothesis
2. The effect of the leader’s time in office is also positive and highly significant, supporting
Hypothesis 4. However, contrary to the expectation of Hypothesis 4, founding regime leaders
are less likely to tolerate opposition media, according to this model. Tables 20 and 21 show
the results of linear and logit models for banned, dropping the regimes that do not vary on
this measure.66 The results for growth and fuel revenue remain essentially the same in the
linear probability model when we subset the data, and moving to a logistic model yields
effects of the same sign and significance.
In the data presented thus far, a group is no longer coded as a main opposition group after
government repression substantially disrupts its ability to operate openly, as occurred with,
for example, the Muslim Brotherhood in Egypt in 2013. After such events, all significant
opposition groups are included in the coding of toleration of opposition media, including the
former main group. However, one could argue that we should still consider such groups to
be the main opposition groups even after they are driven underground, as long as there is no
evidence that their popularity has substantially declined. If the government were to relent
in its repression, such groups might reemerge as the largest and most popular opposition.
And if a mass uprising were to overthrow the regime, such groups might still be the best
placed to seize power or to win free and fair elections. If such a latent main opposition
65
The dropped regimes are Egypt 2011-2012, Egypt 2013-NA, Mauritania 1978-2005, and Tunisia 19562011.
66
The dropped regimes are Egypt 2011-2012, Kuwait 1961-NA, and Sudan 1989-NA. Recall that
Mauritania 1978-2005 and Tunisia 1956-2011 are also dropped because canban always equals 0 for these
regimes.
129
Table 4.18: Tolerated Opposition Media (Regime FE, OLS vs Logistic)
Dependent variable:
tolerated
panel
logistic
linear
log(L unemployment)
(1)
(2)
−1.043∗∗∗
−7.912∗∗∗
0.046∗
0.385∗∗∗
(0.025)
(0.019)
0.385∗∗∗
3.319∗∗∗
(0.087)
(0.099)
0.078
0.596∗∗∗
(0.053)
(0.047)
−0.139
−1.023∗∗∗
−0.038
−0.455∗∗∗
−0.282∗∗
−2.496∗∗∗
−0.494∗∗∗
−3.869∗∗∗
174
130
0.385
174
(0.156)
sqrt L GDP pc growth
log(L fuel + 1)
log(leader year office)
regime first leader
(0.334)
yrs00 05
(0.112)
yrs06 11
(0.121)
yrs12 17
(0.113)
Observations
R
2
Adjusted R2
Log Likelihood
(0.206)
(0.223)
(0.121)
(0.165)
(0.213)
0.295
−66.057
Table 4.19: Tolerated Opposition Media (Regime FE, OLS vs Logistic), continued
Dependent variable:
tolerated
panel
logistic
linear
(1)
(2)
0.135
0.492∗∗∗
(0.114)
(0.109)
−0.086
−0.753∗∗∗
0.144
1.091∗∗∗
(0.185)
(0.205)
0.238
2.421∗∗∗
(0.152)
(0.129)
0.444∗∗∗
3.008∗∗∗
(0.167)
(0.185)
0.329∗∗∗
2.665∗∗∗
(0.058)
(0.072)
174
174
pres elec
leg elec
(0.074)
leader fail
L civil war
is main
groups N
Observations
R2
0.385
Adjusted R2
0.295
(0.051)
Log Likelihood
−66.057
Akaike Inf. Crit.
Note:
178.113
∗
p<0.1;
131
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.20: Opposition Media Bans (Regime FE, OLS vs Logistic)
Dependent variable:
banned
panel
logistic
linear
log(L unemployment)
sqrt L GDP pc growth
(1)
(2)
0.379
9.793∗∗∗
(0.277)
(0.263)
−0.087∗∗∗
−1.374∗∗∗
−0.547∗∗∗
−10.658∗∗∗
−0.085
−3.078∗∗∗
−0.027
−2.839∗∗∗
0.071
−1.542∗∗∗
(0.028)
log(L fuel + 1)
(0.163)
log(leader year office)
(0.071)
regime first leader
(0.347)
yrs00 05
(0.189)
yrs06 10
yr11
yrs12 17
Observations
(0.013)
(0.133)
(0.077)
(0.194)
(0.152)
0.308
4.353∗∗∗
(0.187)
(0.143)
0.544∗∗
5.953∗∗∗
(0.245)
(0.259)
0.099
3.051∗∗∗
(0.180)
132
(0.116)
115
115
Table 4.21: Opposition Media Bans (Regime FE, OLS vs Logistic), continued
Dependent variable:
banned
panel
logistic
linear
pres elec
leg elec
leader fail
(1)
(2)
−0.050
0.167
(0.143)
(0.141)
0.111
2.106∗∗∗
(0.083)
(0.130)
0.208
−2.319∗∗∗
(0.264)
L civil war
−0.403
−24.925∗∗∗
−0.464∗
−5.032∗∗∗
−0.232∗∗∗
−2.791∗∗∗
115
115
(0.283)
is main
(0.235)
groups N
(0.084)
Observations
R2
0.316
Adjusted R2
0.152
Log Likelihood
(0.240)
(0.238)
(0.097)
−28.444
Akaike Inf. Crit.
Note:
(0.353)
102.889
∗
p<0.1;
133
∗∗
p<0.05;
∗∗∗
p<0.01
group exists, perhaps it remains unimportant whether or not other opposition groups are
allowed to have their own media. For this reason, I constructed an alternative dataset in
which any group that is coded as the main opposition group is still coded as such for seven
years after a crackdown drives it underground. Appendix 9 lists the regime-years and groups
(both the main group and other groups that would be excluded from the dataset if the first
group kept its “main” status) that are affected by this alternative coding rule. In addition
to the Egyptian Muslim Brotherhood, this alternative coding is affected by the crackdowns
on Nahda and on Algeria’s Islamic Salvation Front in 1992 and on Al-Wefaq in 2016. Table
22 shows that using this alternative coding rule does not substantially change the estimated
effects of unemployment, fuel, and the “Arab Spring” on tolerated. Table 24 shows that the
alternative coding does not substantially alter the estimated effects of growth and fuel on
banned.
4.7
Conclusion
The data on opposition media in Arab authoritarian regimes in the post-Cold War period
provide substantial support for my theory. In particular, regimes were more likely to allow
opposition media when they faced low unemployment and could draw on high revenues
from oil and natural gas. They were less likely to tolerate opposition media after the
revolutionary year of 2011. Among regimes where opposition groups decided to launch
their own publications, governments were less likely to ban the publications when per
capita GDP growth was high and their coffers were full of fuel revenue. With the partial
exception of the effect of fuel revenue on bans of existing media, these results were robust
to using country fixed effects rather than regime fixed effects, using a logit model rather
than a linear probability model, disaggregating fuel revenue by source, and coding main
134
Table 4.22: Tolerated Opposition Media (Regime FE, alternative coding of main groups)
Dependent variable:
tolerated
log(L unemployment)
−0.972∗∗∗
(0.139)
sqrt L GDP pc growth
0.041∗
(0.021)
log(L fuel + 1)
0.397∗∗∗
(0.078)
log(leader year office)
0.088∗
(0.047)
regime first leader
0.061
(0.307)
yrs00 05
−0.022
(0.082)
yrs06 11
−0.282∗∗∗
(0.098)
yrs12 17
−0.482∗∗∗
(0.097)
Observations
R
202
2
0.391
Adjusted R2
Note:
0.300
∗
p<0.1;
135
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.23: Tolerated Opposition Media (Regime FE, alternative coding of main groups),
continued
Dependent variable:
tolerated
pres elec
0.157∗
(0.089)
leg elec
−0.074
(0.063)
leader fail
0.136
(0.147)
L civil war
0.238∗
(0.139)
is main
0.397∗∗∗
(0.134)
groups N
0.290∗∗∗
(0.055)
Observations
202
R2
Adjusted R
Note:
0.391
2
0.300
∗
p<0.1;
∗∗
136
p<0.05;
∗∗∗
p<0.01
Table 4.24: Opposition Media Bans (Regime FE, alternative coding of main groups)
Dependent variable:
banned
log(L unemployment)
0.377
(0.244)
sqrt L GDP pc growth
−0.070∗∗∗
(0.026)
log(L fuel + 1)
−0.428∗∗∗
(0.146)
log(leader year office)
−0.039
(0.058)
regime first leader
0.457
(0.496)
yrs00 05
0.086
(0.142)
yrs06 10
0.279∗
(0.151)
yr11
0.469∗∗
(0.196)
yrs12 17
0.117
(0.146)
Observations
R
128
2
Adjusted R
Note:
0.279
2
137
∗
p<0.1;
0.102
∗∗
p<0.05;
∗∗∗
p<0.01
Table 4.25: Opposition Media Bans (Regime FE, alternative coding of main groups),
continued
Dependent variable:
banned
pres elec
−0.052
(0.126)
leg elec
0.076
(0.074)
leader fail
−0.030
(0.202)
L civil war
−0.168
(0.203)
is main
−0.473∗∗
(0.212)
groups N
−0.258∗∗∗
(0.084)
Observations
128
R2
Adjusted R
Note:
0.279
2
0.102
∗
p<0.1;
∗∗
138
p<0.05;
∗∗∗
p<0.01
opposition groups differently. Some results were only weakly supported. The correlations
between toleration of opposition media and economic growth, leader tenure in office, and
whether the leader was the regime founder were sensitive to the particular regimes included
in the data. Unemployment was not correlated with bans on existing media specifically,
while the effect of the “Arab Spring” on bans may have been driven by a couple regimes.
The effects of unemployment and growth suggest that these proxies for grievance levels do
affect toleration of opposition media in the way the theory predicts: regimes are less likely
to allow opposition media in years when their economic performance is weaker.
The
positive relationship between fuel revenues and toleration of opposition media (measured
by both dependent variables) may also reflect this negative effect of grievances on
toleration, although an alternative mechanism is that fuel revenues increase the ability and
willingness of the security apparatus to engage in violence. Finally, the lower probability of
toleration after the Arab Spring supports the idea that shocks to the probability of
successful protest that elites and ordinary citizens alike see can also affect toleration of
opposition media.
139
Chapter 5
Conclusion
At the start of this study, I raised a question that has previously received little scholarly
attention, although it relates to the ongoing debates about media freedom in authoritarian
regimes: Why do dictators sometimes allow opposition groups to produce mass media and
sometimes not? In Chapter 2, I proposed a theory that provides an answer to that question:
dictators allow opposition media when their regimes are strong, in order to convince citizens
not to protest. I formalized my claims in a signaling model. From that model, I derived
empirical implications about the relationship between certain economic, institutional, and
international variables, on the one hand, and on the other hand, the likelihood of observing
tolerated opposition media outlets in an authoritarian country. In Chapter 3, I illustrated
some of these implications by focusing on the details of one regime in a brief period —
that of Tunisia in 1987 to 1992. I discussed the opposition groups that it faced, and how
its treatment of their newspapers varied across groups and time, in response to changing
economic conditions, international events, and Ben Ali’s consolidation of personal power over
the regime. In Chapter 4, I described the construction of a new dataset measuring whether
140
governments tolerated opposition media. This dataset included all countries and years which
had authoritarian regimes and in which I could identify any significant opposition groups,
out of all Arab countries during the years 1992-2017. I used this dataset to test the model’s
implications quantitatively. For some hypotheses, some tests did not yield confirmatory
evidence. However, I found some evidence in favor of all but one of them, the hypothesis
about personalization of dictatorships. Overall, the quantitative investigation provides ample
evidence for the overarching implication of the model, that regimes are most likely to allow
opposition groups to produce media when the regimes are least at risk of being overthrown
by mass protests.
5.1
Summary of Findings
In this section I summarize how the hypotheses derived from the model helped to explain
the Tunisian case and were supported by the cross-national evidence.
Hypothesis 1:
Higher unemployment is negatively correlated with the existence of
opposition media.
This hypothesis was consistent with both the case study and the cross-national study. In
Tunisia, Ben Ali did not allow Nahda, the strongest opposition group, to publish any media
when he first came to power and the unemployment rate was nearly eleven percent. Two
years later, with unemployment below seven percent, he allowed Nahda to begin publishing
the newspaper El-Fajr. He did suppress El-Fajr from February 1991 on, but apparently
because Nahda turned out to be stronger than they had appeared a year earlier, despite the
improved economic situation. Unemployment, like all the other factors in these hypotheses,
141
is only one factor among many that affects elites’ assessments of the relative strength of the
opposition and the regime. These hypothesized patterns will not be born out in all times
and places, but rather hold on average, controlling for other factors. When we turned to
the cross-national analysis, and employed regression models that controlled for several key
variables, we found that, although the unemployment rate was not significantly associated
with the variable banned, it was negatively correlated with the variable tolerated. Consistent
with Hypothesis 1, a given regime was less likely to allow any opposition groups to have media
when the unemployment rate was higher than the average for that regime, all else equal.
This effect was substantively large and robust to all the alternative model specifications and
data coding.
Hypothesis 2: Higher per capita GDP growth is positively correlated with the existence of
opposition media.
This hypothesis was also consistent with both the single-country case study and the crossnational data. During Ben Ali’s first two years in power, per capita income in Tunisia
declined. But economic growth was positive and strong in 1990, when Ben Ali decided to
authorize El-Fajr. Of course, as noted, it also matched the improvement in the job market, so
from this single-country study we cannot disentangle the effects of unemployment and income
growth and be sure that — as the theory suggests — both contributed to his decision to allow
Nahda’s newspaper. In the cross-national analysis, we could estimate each effect separately.
Per capita GDP growth is positively associated with tolerated, as predicted, but the effect
is not very robust. Economic growth is negatively correlated with banned, in regime-years
in which there was any opposition media to ban, as predicted. That is, in years when either
some opposition media outlet was initially permitted or in which some opposition group tried
to launch a new media outlet, a given regime was less likely to ban the outlet if economic
growth was higher than that regime’s average. This finding is robust to all the tests I tried.
142
Hypothesis 3: Higher per capita revenue from extractable fuel sources is positively
correlated with the existence of opposition media.
Tunisia was not a substantial producer of oil or natural gas during the late 1980s and early
1990s, so this hypothesis could not contribute to our understanding of that case. But the
cross-national evidence supported this hypothesis. Fuel revenues were positively correlated
with tolerated and negatively correlated with banned, as predicted. The finding for tolerated
is robust to all tests and the effect is substantively large. The finding for banned is robust
with one exception: when fuel revenue is disaggregated into revenue from oil and natural
gas, only gas has a statistically significant effect, and only at the 10% level.
Hypothesis 4: The degree to which control of the security forces is concentrated in the
hands of the dictator personally is positively correlated with the existence of opposition
media.
Ben Ali only allowed Nahda to publish El-Fajr after he had consolidated personal power
within the Tunisian regime, consistent with this hypothesis. However, the cross-national
results provide no clear evidence in favor of this hypothesis. The number of years the regime
leader has been in power, which I use as a proxy for personalization, is positively correlated
with tolerated, but this effect is only significant at the 10% level and is not robust to dropping
certain regimes. Whether the leader founded the regime, another proxy for personalization,
is not significantly correlated with tolerated, and neither measure is significantly associated
with banned. However, leaders’ time in power and whether they founded their regimes are
very rough proxies for personalization, so the findings for this hypothesis are quite tentative.
Hypothesis 5: The occurrence of successful uprisings against nearby authoritarian regime
are negatively correlated with the existence of opposition media.
143
This hypothesis does not directly help us understand the case study of Tunisia, since no Arab
authoritarian regimes were overthrown by protests in 1987-1992. However, the sweeping
electoral success of the FIS in Algeria in June 1990 did seem to lead to a crackdown on
opposition media in Tunisia, consistent with the more general idea that when events in other
authoritarian countries reveal that opposition groups there are popular and able to mobilize
many citizens, dictators will be less likely to tolerate the media of similar opposition groups
in their own countries. The cross-national analysis tests Hypothesis 5 using one particular
historical juncture: the wave of uprisings that overthrew dictatorships across the Middle East
and North Africa in 2011. The period after 2011 was negatively correlated with tolerated
and the year 2011 is positively correlated with banned. The latter effect was not robust to
dropping certain regimes, but the former effect was robust to this and all other tests I tried
(albeit only significant at the 10% level when the Kuwaiti regime was dropped).
In summary, these hypothesis provide substantial insights into the Tunisian case and find
much support in the cross-national data. There is particularly strong evidence for the effects
of the economic variables: unemployment appears to negatively affect toleration of opposition
media, while economic growth and fuel revenues affect it positively.
5.2
Contributions of this Study
This study contributes to the literature on variation in media freedom in authoritarian
regimes by focusing on a previously understudied aspect of the phenomenon: when opposition
groups are and are not allowed to produce their own media. Previous studies have either
looked at media freedom very broadly — using indices such as Freedom House’s “Freedom
of the Press” scores that combine many different aspects of media freedom (e.g. Egorov,
144
Guriev, and Sonin 2009) — or have looked at specific aspects of media freedom other than
the one I study here, such as censorship of citizens’ social media posts (King, Pan, and
Roberts 2013). This study proposes a new theory and analyzes new data that specifically
address the determinants of toleration for opposition media.
This study has contributed to the theoretical literature on this topic. Previous studies have
suggested various motivations that sometimes lead authoritarian leaders to allow a degree
of media pluralism. My theory proposes a new motivation: in some circumstances, dictators
can signal their strength by allowing opposition groups to publish media. Empirically, this
study has documented that, firstly, authoritarian regimes do often tolerate opposition media.
Secondly, it has uncovered a number of factors that affect whether regimes tolerate opposition
media or not. In particular, stronger economic growth and greater fuel rents make them more
likely to tolerate opposition media, while higher unemployment and successful uprisings
against other authoritarian regimes make them less tolerant.
This study sheds new light on certain empirical findings from previous research. King, Pan,
and Roberts (2013) find evidence that even China, despite its high capacity to censor
information, sometimes chooses not to.
This suggests that variation in the extent of
censorship is not driven only by variation in state capacity to censor, but is also driven by
variation in the political benefits of allowing relative freedom of expression. My analysis of
cross-national, multi-decade data provides evidence that, indeed, the existence of
opposition media under some authoritarian regimes cannot be explained by a lack of state
capacity.
For it is precisely in years when countries experience strong economic
performance and have access to high revenues from extracting oil and gas that we are most
likely to see opposition media publishing openly.
My theory provides a possible explanation for the finding of Whitten-Woodring and Van Belle
145
(2014) that protests are associated with declines in media freedom. Organized protests can
indicate that opposition elites believe their chances of success are high, while the outbreak
of spontaneous protests may cause elites to think citizens are more willing to protest than
they thought. In these circumstances, my theory suggests that opposition leaders will be
unwilling to produce tame media content and dictators will be inclined to prohibit opposition
media.
The results showing that higher fuel rents increase the probability that governments will
tolerate opposition media complicates the finding of Egorov, Guriev, and Sonin (2009) that
oil revenue is negatively correlated with media freedom. The difference in our findings is
likely due to the fact that we measure media freedom differently. Those authors use a broad
index of media freedom produced by Freedom House, which is based on questions about how
countries’ laws, extra-legal actions, economic conditions, and even non-state actors affect
the ability of all types of media outlets to operate (Freedom House 2017c). By contrast,
I measure the more specific phenomenon of whether governments allow opposition groups
to operate media outlets and citizens to access them, or not. This study suggest that the
logic of their theory, which addresses when and why dictators allow investigative journalists
to report on the actions of low-ranking government officials, does not apply to the separate
question of why dictators do or do not allow opposition groups to produce their own media.
As I have argued, attempts by opposition media (as opposed to less partisan media) to expose
the failings of authoritarian regimes are likely to be discounted as non-credible by citizens,
whereas a message from an opposition media outlet that protest is unlikely to succeed is
more credible, because it would not always be in the opposition’s interest to send it. My
findings do not invalidate those of Egorov, Guriev, and Sonin, but rather suggest that their
findings do not generalize to all aspects of media freedom. My study also underscores the
need for fine-grained measures of the distinct dimensions of media freedom.
146
5.3
Limitations of this Study and Suggestions for
Future Research
As any model must, the one I have presented makes many simplifying assumptions that limit
its realism. Extended versions of the model can determine whether the results depend on such
strong assumptions, as well as yielding additional and more nuanced empirical predictions.
The following are some extensions worth pursuing:
• Instead of assuming that the regime leader and opposition leader have the same
information, assume that they see private but correlated signals of the probability of
protest success.
Such a model would yield insight into how the regime and the
opposition might behave when the opposition believes protest is likely to succeed but
the regime does not, and vice versa.
• In constructing the dataset, I came across many cases in which there were multiple
significant opposition groups in a single regime-year. Often these groups differed
significantly from each other in terms of their ideologies or their willingness to
cooperate with the regime. An extended model could include multiple opposition
leaders, each of whom has some chance of coming to power if the regime is
overthrown. Some opposition players may prefer the continuation of the regime to a
transition to rule by certain other opposition groups and their groups may have
varying degrees of public support. In which cases will each opposition leader want to
encourage or discourage protest, and when will the regime allow allow some of them
to have media but not others? What inferences do citizens make if they see that some
opposition groups have media and others do not?
• Lastly, the game could be repeated over multiple periods. We could assume that
147
whether the opposition has the resources to produce media in a given period is
positively correlated with whether they had the resources in the prior period. In such
a model, if the opposition have media in one period and not the next, citizens would
put more weight in the later period on the probability that the government actively
prevented the opposition from producing media, as opposed to the opposition lacking
the resources, compared to a situation in which the opposition did not have media in
either period. This effect may reduce the dictator’s willingness to allow opposition
media in the first place. Alternatively, allowing an opposition media outlet in one
period might enable the regime leader to more credibly commit to allowing it in the
future.
Further research is needed to address some limitations of my case study of Tunisia under
Ben Ali. I was not able to obtain interviews with as many individuals who held high-ranking
positions in the regime during 1987-1992 as I was with leaders of opposition groups from that
period. More interviews with former regime officials could reveal whether their beliefs and
goals at the time were consistent with those in my model. Content analysis of opposition
media from this period would also be illuminating. When the government was tolerating
certain opposition media outlets, did those outlets call only for limited policy reforms in the
short-run (even if the groups’ long-term goals included a regime transition or other sweeping
change), as my theory predicts? Did they discourage or at least not actively call for antigovernment protests? On occasions when opposition media did call for protests or immediate
regime change, did Ben Ali’s government invariably shut them down, as he did with El-Fajr
in June 1990?
In order to estimate causal effects, the cross-national analysis made use of regime fixed
effects and period dummies. However, this is not a perfect identification strategy, as it
does not control for possible omitted variables that vary over time within regimes. Future
148
research should look for exogenous shocks to unemployment, growth, fuel revenue, and other
explanatory variables in order to identify their causal effects on toleration of opposition
media with greater confidence. By obviating the need for regime fixed effects, this approach
would also make it possible to test the effects of time-invariant regime characteristics, such
as how the ruling group came to power.
Since the unit of observation in the cross-national analysis was the regime-year, it did not
measure variation in governments’ treatment of media across different opposition groups in
a given year, and how that such differential treatment is affected by group-specific variables,
such as how recently the group was established, its ideological orientation (e.g. Islamist or
secular), or whether the group advocated an eventual regime transition or only reform within
the existing regime. Because I collected detailed information on the media of each of the
opposition groups on which the dataset was based, it would be straightforward in future to
expand the analysis to the regime-group-year level.
Again, content analysis is another promising avenue for research on this topic.
For
contemporary regimes, it would be instructive to analyze opposition groups’ accounts on
international social media platforms like Facebook and Twitter, which are difficult for
regimes to control. A testable implication of my theory is that, if we compare groups that
are allowed to have newspapers, unblocked websites, or other media that governments can
cut off, with groups that are not allowed any such media, we would find the former posting
relatively mild criticism on their social media accounts and not calling for protest, while
the latter groups post more critical content and are more likely to call for protest. We can
also compare the same group over time when it was and was not allowed to have media
outlets. Automated methods would enable content analysis of large volumes of posts across
many nations and years.
149
To further test the robustness of my findings from the cross-national chapter, it would also
be useful to produce alternative versions of my dataset based on different coding rules. For
instance, do the results change if we include all opposition groups that meet the electoral
or protest criterion for significance, even in regime-years when there is also an identifiable
“main” opposition group? Does it matter if we use lower thresholds to code groups as
significant (e.g. winning 5% of votes rather than 10% or organizing protests with over 100
participants rather than 1,000)?
The cross-national evidence was limited to Arab countries. Do the findings from that region
generalize to other regions? Eventually the dataset I collected for Arab countries should be
expanded to include authoritarian regimes around the world.
With the theory, the case study, and the cross-national analysis, I deliberately focused on
regimes in which there were opposition groups with non-negligible ability to mobilize citizens.
I suspect that regime leaders’ incentives to allow or prohibit opposition media might differ
when they face the smallest opposition groups, with almost no mobilization capacity. The
potential benefit of giving such a group a public platform to discourage protest by its few
followers may be small compared to the risk that allowing its media would make more
citizens aware of the group’s existence, helping it to grow. Thus, when we consider the
entire universe of authoritarian regimes and opposition groups, the relationship between an
opposition group’s mobilization capability (relative to the regime’s repression capability)
and the likelihood that the regime will let them have a media outlet may be an inverted U
curve. This argument might explain the apparent absence of opposition media in certain
stable states with only minuscule opposition groups, such as the United Arab Emirates.
Social scientists must remain humble about the longevity of the patterns they uncover.
Technological changes — particularly in how people communicate and process information
150
— may have profound and rapid effects on the nature of opposition media in authoritarian
regimes. As these changes unfold, they will raise new questions for political scientists on
that topic and possibly change the answers to current questions. In the long-run, who will be
most empowered by technological developments: opposition leaders, authoritarian regimes,
or ordinary citizens? Will opposition groups gain increasing ability to communicate with
citizens in ways that regimes cannot stop? Will citizens gain greater access to alternative
sources of information — beyond those produced by governments or opposition groups —
as well as becoming more able to communicate directly with each other? Or will regimes
become increasingly adept at filtering their citizens’ access to information from all sources?
How will these changes in the distribution of technical capacities affect governments’ policies
regarding opposition media, opposition groups’ strategies for promoting political change,
and citizens’ decisions to oppose, acquiesce to, or actively support their rulers?
5.4
Conclusion
From the perspective of those who value freedom of expression and democracy, my
conclusions are in some ways pessimistic. They suggest that the existence of opposition
media under an authoritarian regime does not imply a lively public sphere in which
opposition groups fiercely challenge the regime. Nor is the toleration of opposition media
necessarily a harbinger of further political liberalization: in fact, it can be a means for
preventing political change.
But what I have found also has at least one hopeful
implication. Censorship, specifically governments’ efforts to stop opposition media outlets
from functioning or to block the public’s access to them, is not completely effective at
hiding a regime’s weakness. The very absence of opposition media can provide a strong
clue to citizens that their rulers fear their collective power.
151
Appendix 1: Proof of result in model
Proof that S 0 < S ∅ < S 1
S0 =
S∅ =
(1 − ρ)
R S⋆
0
S1 =
R S⋆
0
SfS [S]dS
F [S ⋆ ]
(5.1)
R1
SfS [S]dS + S ⋆ SfS [S]dS
1 − ρF [S ⋆ ]
(5.2)
R1
SfS [S]dS
1 − F [S ⋆ ]
(5.3)
S − ρF [S ⋆ ]S 0
1 − ρF [S ⋆ ]
(5.4)
S⋆
The equation for S ∅ can be rewritten as:
S∅ =
152
where S is the prior expectation of S. It can be shown that S 0 < S ∅ by contradiction.
Suppose that S 0 ≥ S ∅ :
S0 ≥
S − ρF [S ⋆ ]S 0
1 − ρF [S ⋆ ]
(5.5)
This would imply that: S 0 ≥ S, which cannot be the case, since S 0 is the expectation of S
when fS is truncated to (0, S ⋆ ].
It can also be shown by contradiction that S ∅ < S 1 . Suppose that: S ∅ ≥ S 1 :
(1 − ρ)
R S⋆
0
R1
SfS [S]dS + S ⋆ SfS [S]dS
≥ S1
1 − ρF [S ⋆ ]
(5.6)
This inequality can be expressed as:
(1 − ρ)F [S ⋆ ]S 0 + (1 − F [S ⋆ ])S 1
≥ S1
1 − ρF [S ⋆ ]
(5.7)
which implies that: S 0 ≥ S 1 . But this cannot be the case, since, as noted, S 0 is the
expectation of S when fS is truncated to (0, S ⋆ ], while S 1 is the expectation of S when fS
is truncated to (S ⋆ , 1).
153
Appendix 2: Regime Narratives
Algeria
1992-2017: authoritarian (1992-NA)
Start: 1/11/1992 “Military coup. Benjedid was replaced by the High Council of State, and
the constitution was suspended. Although the military had been an important pillar of the
pre-1992 regime, the post-1992 period is considered a different regime because for the first
several years after the coup the FLN, which had been the ruling party, was excluded from
influence and office.” [GWF]
End: Regime continued in power as of December 31, 2017 (Economist Intelligence Unit
2018a; Aghrout and Zoubir 2019).
154
Bahrain
1992-2017: authoritarian (1971-NA)
Start: 8/15/1971 The Al Khalifa dynasty has ruled Bahrain as a monarchy since conquering
the island in 1796 (with the exception of 1800-1802). Bahrain became independent from the
United Kingdom in 1971 (Economist Intelligence Unit 2018b; Jones 2018; Lansford 2017).
End: Regime continued in power as of December 31, 2017 (Economist Intelligence Unit
2018b; Jones 2019).
Comoros
Less than 1 million population in 2018
Djibouti
Less than 1 million population in 2018
155
Egypt
1992-2012: authoritarian (1952-NA and 2011-2012)
2013: democracy
2013-2017: authoritarian (2013-NA)
Start: 7/23/1952 “Coup led by Lt Col Nasser and the Free Officers, an organization of junior
and mid-level officers. They established the Revolutionary Command Council made up of
members of the Free Officers’ executive committee to rule.” [GWF]
End/Start: 2/11/2011 Mass protests drove President Mubarak to resign, handing power to
the Supreme Council of the Armed Forces (SCAF). The SCAF is coded as a new regime,
because it was completely dominated by the military leadership as a group, whereas they
had a lesser political role under the somewhat personalist rulers of the previous regime
(Economist Intelligence Unit 1996a, Lansford 2017, Europa World 2019a. “Contemporary
Political History (Egypt).” ).
End: 8/12/2012 Morsi forced the head of the SCAF (Tantawi) and several other top military
leaders in the SCAF to resign and revoked the SCAF’s June decree expanding its powers.
(After dissolving the parliament in June 2012, the SCAF had placed restrictions on the
power of the president and assumed authority over law-making, the military budget, and the
156
drafting of a new constitution.) The dismissals were not challenged by the SCAF and are
seen as the moment when the balance of power shifted from the SCAF to Morsi’s government.
Morsi had been elected president in elections considered mostly free and fair in June, so this
event is coded as a transition to democracy (Freedom House 2013 “Freedom in the World”
, Lansford 2017, Europa World 2019a. “Contemporary Political History (Egypt).” ).
Start: 7/3/2013 The military ousted Morsi in a coup and Egypt was thereafter ruled by
General Abd al Fatah al Sisi, the Commander in Chief of the armed forces and Minister of
Defense and Military Production. Al Sisi initially was a de facto ruler, having appointed a
civilian president. After violently suppressing the Muslim Brotherhood (whose Freedom and
Justice Party had won competitive legislative and presidential elections in 2012), Al Sisi was
elected president in 2014 (Lansford 2017, Europa World 2019a. “Contemporary Political
History (Egypt).” ).
End: The regime remained in power as of December 31, 2017 (Europa World 2019a.
“Contemporary Political History (Egypt).” ).
Iraq
1992-2003: authoritarian (1979-2003)
2004-2009: foreign occupation
157
2010- 2017: authoritarian (2010-NA)
Start: 7/16/1979 “The formal transfer of power from Field Marshal al-Bakr to Saddam
Hussein completed a gradual shift from a regime based mostly on Ba’thist military officers
and the Ba’th party to one in which the group from which leaders could be chosen included
few outside Saddam’s family and home region. Saddam had built his power base in the
party and security service, not the officer corps. He exerted significant party control within
the military through his control of party networks prior to al-Bakr’s retirement, but he was
not a career officer. After al-Bakr’s retirement Saddam purged the party of anyone with an
independent base of support and then reduced the party’s power and relevance. He executed
several members of the ruling Revolutionary Command Council a few days after his accession
to full power. Beginning in 1980, the party was subordinated to the military and security
services. The party’s Regional Command was stacked with Saddam’s ministerial and security
service subordinates, his advisors, and his relatives.” [GWF]
End: 4/7/2003 “The fall of Baghdad to invading U.S. and coalition forces. U.S. forces took
control of presidential palace on April 7.” [GWF]
Start: 3/7/2010 The first parliamentary elections since the US transferred authority over
security to Iraqi forces mark the emergence of a new authoritarian regime.
Over 500
candidates (about 1 in 12) were banned from running because of alleged Baathist ties.
Many prominent Sunni and secular Shia politicians were targeted, particularly from
Iraqiya, a rival party to the State of Law Party of the incumbent prime minister Nuri
al-Maliki. He was selected by the new parliament to remain as prime minister (Al-Jazeera
158
2010, Anderson 2019, Human Rights Watch 2010a. “Iraq’s 2010 National Elections: A
Human Rights Platform for Candidates.” , Reuters 2010).
End: The regime remained in power as of December 31, 2017 (Anderson 2019, Dodge 2013).
Jordan
1992-2017: authoritarian (1946-NA)
Start: 5/25/1946 “Independence under a monarchy established by the British.
King
Adbullah of the Hashemite family was the son of the Ottoman amir of Mecca, who claimed
a hereditary right to rule in the Hijaz. He was a leader of the Arab nationalist movement
against Ottoman rule and sided with the British during World Wars I and II.” [GWF]
End: Regime continued in power as of December 31, 2017 (Economist Intelligence Unit
2018c; Freer 2019).
Kuwait
1992-2017: authoritarian (1961-NA)
159
Start: 6/19/1961 “Independence under the traditional al Sabah emirate.” [GWF]
End: Regime continued in power as of December 31, 2017 (Crystal 2019; Economist
Intelligence Unit 2018d).
Lebanon
1992-2004: warlord/foreign-occupied
2005-2017: democracy
Sources on 2011-2017: Economist Intelligence Unit 2013a, Economist Intelligence Unit 2013b;
Lansford 2017, Saouli 2019
Libya
1992-2011: authoritarian (1969-2011)
2012-2017: warlord
160
Start: 9/1/1969 “Coup by a small group of junior officers led by Capt Qadhaffi deposed the
monarch and established the Revolutionary Command Council to rule.” [GWF]
End: 8/22/2011 The regime ends with the fall of the capital, Tripoli, to rebel forces. Qaddafi
was captured and killed October 20, 2011. Libya is subsequently coded as being in a condition
of warlordism, as the transitional government was unable to wrest control of the country from
militias, and by 2016, the three governments were competing for control of Libya (Erdbrink
2011; Kirkpatrick 2012; Recknagel 2012; U.S. Department of State 2014; Freedom House
2017a. “Freedom in the World” ; St John 2019.
Mauritania
1992-2006: authoritarian (1978-2005 and 2005-2007)
2007-2008: democracy
2009-2017: authoritarian (2008-NA)
Start: 7/10/1978 “Coup by junior officers led by army commander Col Salek ousted the
Daddah government and formed the Military Committee for National Recovery (CMRN) of
20 officers to rule. The regime was significantly civilianized under Col Taya, who succeeded
the officer who succeeded Salek.” [GWF]
161
End/Start: 8/3/2005 “Coup by members of the presidential guard and the military ousted
the Taya government.” [GWF]
End: 3/25/2007 “Competitive presidential election considered free and fair transferred power
to civilians.” [GWF]
Start: 8/6/2008 “Coup led by active duty and recently dismissed senior officers from the
President’s Security Battalion ousted the elected government and established the all-military
High Council of State.” [GWF]
End: Regime remained in power as of December 31, 2017 (Economist Intelligence Unit
2018e).
Morocco
1992-2017: authoritarian (1956-NA)
Start: Start: 3/2/1956 “Independence under the rule of the traditional Sultan, Mohammed
V.” [GWF]
162
End: Regime remained in power as of December 31, 2017 (Lansford 2017; Partrick 2019a).
Oman
1992-2017: authoritarian (1941-NA)
Start: 12/31/1741 “We date the start of the Al Said dynasty to 1741. With the previous
dynasty weakened by civil war over the succession and poor leadership, in 1741 Ahmed bin
Said al Busaidi, governor of Sohar on the coast of what is now of Oman, led the city’s defense
against a Persian invasion. Although he did not become the formal leader of Oman until
1744 (probably—date of formal election is disputed) when he was named imam, Ahmed bin
Said seems to have been the most powerful leader during a very chaotic time beginning in
1741. The Al Said have remained in power as traditional sultans since then.” [GWF]
End: Regime remained in power as of December 31, 2017 (Lansford 2017; Valeri 2019).
Palestine
1992-2017: not independent (Parsons 2019)
163
Qatar
1992-2017: authoritarian (1971-NA)
Start: 9/1/1971 With British support, the Al Thani dynasty achieved de facto control over
Qatar in 1868 (Peck 2008). Qatar became independent from the UK under the Al Thanis
on September 1, 1971 (Lansford 2017).
End:
Regime remained in power as of December 31, 2017 (Europa World 2019c.
“Contemporary Political History (Qatar)” ).
Saudi Arabia
1992-2017: authoritarian (1927-NA)
Start: 5/20/1927 “The Treaty of Jeddah in which Great Britain recognized the independence
of the Kingdoms of Hijaz and Najd was signed on this date. They were ruled by the monarchy
that had been established during the previous 25 years through conquest by Adb al Aziz al
Saud, leader of the al Saud family. In September 1932, the kingdoms were formally united
under the name Saudi Arabia.” [GWF]
164
End: Regime remained in power as of December 31, 2017 (Peterson and Nonneman 2019).
Somalia
1992-2017: warlord
Sources on 2011-2017: Bacon 2018, BBC News 2017b, Danish Immigration Service 2017,
Lewis and Janzen 2018, Reuters 2013; Freedom House 2018a. “Freedom in the World”
Sudan
1992-2017: authoritarian (1989-NA)
Start: 6/30/1989 “Coup led by Col al-Bashir and an Islamist faction of the military ousted
the elected government, imposed a state of emergency, dissolved parties and unions,
banned demonstrations, and established the 15 member, all military Revolutionary
Command Council for National Salvation to rule. The new rulers dismissed much of the
officer corps.” [GWF]
165
End: Regime remained in power as of December 31, 2017 (Woodside 2019).
Syria
1992-2017: authoritarian (1963-NA)
Start: “3/8/1963 Coup led by pro-Ba’thist officers ousted Gen Zahr al-Din’s government
and its civilian allies. The March 1963 coup was led by officers sympathetic to the Ba’th
but not actually members. The National Council of the Revolutionary Command set up
immediately after the coup contained a minority of Ba’th members, and the first cabinet
they chose was half Ba’th. Over the next few months of factional struggle within the new
government and within the military, however, most non-Ba’th officers were excluded from
leadership and many non-Ba’thist officers and NCOs were purged from the army, leaving the
regime dominated by the Ba’thist faction of the military.” [GWF]
End: Regime remained in power as of December 31, 2017 (United States Central Intelligence
Agency 2019d. “Syria” , Zisser 2019).
166
Tunisia
1992-2010: authoritarian (1956-2011)
2011: transitional
2012-2017: democracy
Start:
“3/20/1956 Independence under the control of Neo-Destour and Bourguiba.
Bourguiba and other Neo-Destour party insiders chose electoral rules without consultation,
effectively eliminating opposition from the government and Constituent Assembly, which
was elected March 25, 1956. Although the government was nominally a constitutional
monarchy, Bourguiba and Neo-Destour had full control of decision making from the
beginning. The Bey was formally deposed the following year.” [GWF]
End/Start: 1/14/2011 Mass protests forced Ben Ali to resign and flee the country. The
new interim administration was headed by long-serving officials from the previous regime
but the cabinet combined both former officials from the ruling RCD party and opposition
and human rights leaders. On January 19, 2011, the former RCD members of the cabinet
resigned from the party. The RCD itself was abolished by a court order on March 9, 2011
(Lansford 2017, Partrick 2019b).
End: 10/23/2011 Tunisia transitioned to democracy with the holding of elections for a
167
Constituent Assembly, which were widely regarded as free and fair by international observers.
The Constituent Assembly chose a new prime minister and president (Freedom House 2018a
“Freedom in the World” , Levinson 2011, Murphy 2013, National Democratic Institute 2011,
Partrick 2019).
United Arab Emirates
1992-2017: authoritarian (1971-NA)
Start: 12/2/1971 “Formed when several sheikdoms along the Gulf united to create the
independent UAE. The president is chosen by the Supreme Council of the Union, which is
made up of the hereditary rulers of the initially six, now seven, emirates, from among its
members. He serves as head of state, chair of the SCU, and commander of the military.”
[GWF]
End: Regime remained in power as of December 31, 2017 (Coates Ulrichsen 2019).
168
Yemen
1992-2014: authoritarian (1978-2012, 2012-2015)
2015-2017: warlord
Start: In 7/17/1978 “The four-man Presidential Council that led the 1974-78 regime briefly
after earlier leaders’ assassinations chose Lt-Col Salih as president and commander-in-chief
of the armed forces. The post-1978 regime is considered different from the 1974-78 regime
because Salih began almost immediately to change the identity of those who could influence
policy, reducing the military’s role and incorporating sheikhs. Beginning in late 1978, he
purged important officers, narrowingthe faction of the military included in the inner circle.
He brought back to influence sheikhs who had been excluded since 1974 and gave his family
and tribe a privileged place in decision making, distribution, and command positions in the
military. Beginning in 1979, Salih began developing institutions to allow some participation
in politics by ordinary citizens.” [GWF]
End/Start: 2/27/2012 Saleh resigns and formally transfers power to his vice-president, Abd
Rabo Mansour Hadi in a plebiscite, who had won a plebiscite on February 21. Hadi led a
coalition government in which the formerly dominant General People’s Congress party shared
power with Joint Meeting Parties, a coalition of opposition parties. Hadi’s presidency marks
the beginning of a new regime, as Hadi immediately begins systematically removing Saleh’s
family members from the many leadership posts they held in the military and security
apparatus (Mounassar 2012, Lansford 2017, Schmitz and Burrowes 2018, Durac 2019).
169
End 1/22/2015: President Hadi submitted his resignation, leaving the Houthi rebels in
control of the capital, Sanaa (Agence France Presse 2015, Reuters 2015). However, the
Houthis had not achieved control of most of Yemen’s territory by the end of 2017, so Yemen
is coded as in a state of warlordism in 2015-2017 (Baron 2016; Al Jazeera 2019; European
Council on Foreign Relations 2019).
170
Appendix 3: Opposition Group
Narratives
ALGERIA
The Front Islamique du Salut is coded as a significant opposition group during 19921994, having won 41.5% of the vote in the first round of the December 1991 parliamentary
elections, a large plurality (Economist Intelligence Unit 1996d: 4). (The military mounted a
coup in January 1992 and cancelled the second round of the elections [Aghrout and Zoubir
2019].)1 During the years 1995-1999, the FIS is coded as an insurgent group, having formed
an “armed wing” known as the Armée Islamique du Salut (Islamic Army of Salvation—AIS)
1
Although the FIS was clearly the most popular opposition group during 1990-1991, the new military
regime cracked down on the group in 1992 to the extent that it could not function openly as an unarmed
movement, so it is not coded as the main opposition group during the years in the dataset. After the ruling
Front de libération nationale (National Liberation Front—FLN) allowed other parties to legally form for the
first time in 1989, the FIS “quickly emerged as the only serious nationwide competitor to the FLN” (Aghrout
and Zoubir 2019). Similarly, Entelis and Arone (1995: 416-417) write that “the FIS emerged as the only
national contender to the political hegemony of the FLN” in 1989. The Economist Intelligence Unit (2002d:
12) observes: “After its establishment in 1989, the FIS quickly became the most potent opposition force in
the country.” “The status of the FIS as the main opposition party was confirmed emphatically by the local
elections held on 12 June 1990, the first free elections since independence” (Evans and Phillips 2007: 157),
when the FIS won the majority of votes (Aghrout and Zoubir 2019) and control of the councils of Algeria’s
three largest cities (Economist Intelligence Unit 1996d: 4). After the January 1992 coup, the new military
systematically crushed the FIS (Aghrout and Zoubir 2019). Most of the FIS’s national leadership, hundreds
of its local officials, and over 30,000 of its rank-and-file supporters were arrested (CQ Press 2006: 76-77).
171
in July 1994 (Willis 1996: 327; Willis 2012: 174; Aghrout and Zoubir 2019).2 The FIS is
therefore coded as one of several significant opposition groups in 1992-1994.3
The Mouvement pour la Démocratie en Algérie (Movement for Democracy in
Algeria—MDA) was founded by former president Ahmed Ben Bella in 1984 to advocate for
“a pluralistic and democratic Algerian government” (Naylor 2015: 117) The MDA is coded
as a significant opposition group for the years 1992-1996, on the basis of anti-government
protests it organized in June 1990 (Reuters 1990c).
The Front des Forces Socialistes (Socialist Forces Front—FFS) was founded in 1963 to
advocate for the Berbers of the Kabylia region (Naylor 2015: 268). The FFS is coded as
a significant opposition during 1992-2007, based on winning more than 10% of the vote in
the first round of the December 1991 parliamentary elections (Economist Intelligence Unit
1996d: 4) and on organizing anti-government demonstrations in July 1998 (BBC Monitoring:
Middle East 1998c. Thousands march in Algiers in protest at Berber singer’s death) and
May 2001 (Reuters 2001b. Manifestation à Alger, troubles en petite Kabylie).
The Mouvement Culturel Berbère (Berber Cultural Movement—MCB) was founded
in the 1980s to advocate for the official recognition and teaching of the Berber language,
Amazigh, and the protection of the civil rights of the Berber population (Reuters 1989a.
2
Aghrout and Zoubir (2019) and Smith (2004) also refer to the AIS as the “armed wing” of the FIS.
Although other Islamist groups had taken up arms against the Algerian state before July 1994, the connection
of these groups to the FIS was tenuous. Willis (2012: 173) notes that: “The nature and extent of links
between the armed groups and the FIS itself were initially unclear, mostly because the FIS itself was in
turmoil, its leadership broken up and divided between exile, hiding and imprisonment.” Willis (1996: 328)
therefore argues that “The creation of the AIS differed from past attempts at uniting and unifying the
Islamist struggle in that, for the first time, the armed struggle within Algeria was explicitly subordinated to
the political leadership of the FIS.”
3
One of the FIS founders, Ali Benhadj, participated in protests in January 2011 in Algiers (Naylor 2015:
271), but contemporaneous news accounts indicate that neither he nor the FIS as a group organized the
event (Agence France Presse 2011a.201 Algerian Islamists’ former number two indicted). The former leader
of the AIS and another high-ranking former FIS official endorsed the reelection of the incumbent president,
Abdelaziz Bouteflika, in 2004, but the FIS leadership did not endorse Bouteflika, so I continue to code the
FIS as an opposition group in this period (Agence France Presse 2004).
172
Algerian Berbers Take First Step Toward Political Party; BBC Monitoring: Middle East
1994b. “Shops Close in Tizi Ouzou in Response to Berber Cultural Movement’s Strike
Call.” ; Chtatou 2019). The group is coded as significant in 1992-2007, having organized
anti-government demonstrations in January 1990 (Reuters 1990a. About 50,000 Berbers
Demonstrated in Central Algiers), April 1994 (Reuters. 1994b. “Berbers march in Algeria,
towns paralysed.” ), and April 2001 (Agence France Presse 2001a).
The FLN, despite being Algeria’s ruling party from independence until the January 1992
coup and part of the ruling coalition under Bouteflika, was effectively an opposition group
for the years 1993-1996.
The secretary-general of the FLN immediately declared the
January 1992 coup to be unconstitutional (Middle East Economic Digest 1992d. “Front de
Liberation Nationale deeply split over cancellation of elections” ).
The military
marginalized the FLN from power during this period, even ordering the party to turn its
headquarters over to the state (Reuters 1992). However, the Prime Minister, Sid Ahmed
Ghozali, was a member of the FLN central committee until he resigned from both the
committee and the premiership in July 1992. He was replaced by Belaid Abdessalem, who
had “been in the political wilderness” since 1979.
So I code the FLN as becoming
oppositional only when Ghozali left the government. Entelis and Arone (1995: 416) write:
“Now barred from political activity by military decree following the coup, the FLN became
one of the regime’s strongest opposition groups.” Likewise, Naylor (2015: 267) observes
that, after the coup “the FLN became part of the opposition to the growing authority of
the Pouvoir” (the “power,” a common term in Algeria for regime elites). Likewise, Legum
(1996: B 468) describes the FLN as one of the two “main opposition parties.”
4
The FLN
qualifies as a significant group, having been the ruling party until January 1992 (Baggaley
1992c; Aghrout and Zoubir 2019). I cease coding the FLN as an opposition party in 1997,
4
Lansford (2017) argues: “By late 1994 the FLN was firmly in the opposition camp.” I begin coding the
FLN as oppositional in 1992 because of the party secretary-general’s early public opposition to the coup.
173
because the party has been part of each Algerian cabinet since June 1997.5 Thus the FLN
is considered a significant opposition group in the years 1993-1996.
The Coordination des Aarchs, Dairas et Communes (CADC6 ) is a pro-democracy
group that advocates for the Berbers of Kabylia (Naylor 2015: 192), specifically for the
removal of gendarmes from the region (Aghrout and Zoubir 2019). The CADC organized
large anti-government demonstrations in June 2001 (Aghrout and Zoubir 2019) and April
2004 (All Africa 2004) and thus is coded as a significant opposition group for 2001-2010.
The Rassemblement pour la Culture et la Démocratie (Rally for Culture and
Democracy—RCD) was legally established in 1989 and represents the Berber minority
(Naylor 2015: 439).7
The party organized a large anti-government demonstration in
February 2011 (Faucon 2011) and thus is coded as a significant opposition party in
2011-2017.
The Coordination Nationale pour le Changement et la Démocratie (National
Coordination for Change and Democracy—CNCD) is an umbrella group that began in
5
Parties’ presence or absence in the cabinet are based on the following articles: BBC Monitoring: Middle
East 1996c. HAMAS and the Algerian Renewal Party Given Portfolios in New Government; Middle East
Economic Digest 1997; All Africa 1999; Touati 1999; Agence France Presse 2002; de Bendern 2004; BBC
Monitoring: Middle East 2006; Agence France Presse 2008a Algerian president brings back Ouyahia for third
stint as PM; BBC Monitoring: Middle East 2012; Economist Intelligence Unit 2012a; Aflou 2014; Asharq
Al-Awsat 2017; Economist Intelligence Unit 2017a.
6
“Aarch,” “daira,” and “commune” denote different levels of administrative units in Algeria (Agenzia
Nazionale Stampa Associata 2004; Naylor 2015: 192).
7
The RCD had an ambiguous stance toward the regime in the 1990s (see: Reuters 1995a; Financial Times
1997b; Khalaf 1997; Reuters 1997; Economist Intelligence Unit 2000: 15; Naylor 2015: 225, 452). However,
since 2001, the party has been clearly oppositional. In May 2001, the RCD withdrew from the cabinet to
protest the government’s use of violence against Berber demonstrators (Reuters 2001a). The party boycotted
elections in 2002, 2009, and 2014, and accused the government of rigging a 2005 referendum (Lansford 2012;
Lansford 2017). In 2008, when President Bouteflika announced his intention to stand for a third term,
the RCD accused him of trying to be a “president for life” and of having made “slaves of all Algerians”
(Saleh 2008). Lansford (2017) writes that “the RCD was one of the most vocal opposition groups calling
for constitutional overhaul and the ouster of the Bouteflika government in concert with the Arab Spring in
the first half of 2011.” In 2014, the RCD’s leader declared that the “regime no longer has a future” (BBC
Monitoring: Middle East 2014).
174
2011, which called for the constitutional removal from office of Bouteflika (Naylor 2015:
192-193). The CNCD organized a large anti-government protest in February 2011 (BBC
Monitoring: Newsfile 2011), and so it is coded as a significant opposition group for
2011-2017.
The Coordination Nationale pour les Libertés et la Transition Démocratique
(National Coordination for Liberties and Democratic Transition—CNLTD) was formed
shortly after the April 2014 presidential election, calling on Bouteflika to resign and
advocating “democratization and political inclusion” more generally. Its members included
MSP, Ennahda, RCD and leading figures from Jil Jadid, the Front de la Justice et le
Développment (Justice and Development Front—FJD), and FIS (Naylor 2015: 193). The
CNLTD organized a series of large anti-government demonstrations in early 2015
(Economist Intelligence Unit 2015a) and thus is coded as a significant opposition
organization for 2015-2017.
BAHRAIN
Bahrain lacked a significant organized opposition in 1992-2001. Formed in November 2001
as a political bloc to compete in parliamentary elections, Jam‘iyyat al-Wefaq al-Watani
al-Islamiyya (The Islamic National Accord Society, or al-Wefaq for short) called for a
constitutional monarchy and an end to discrimination against the Shiite Muslims who make
up the majority of Bahrain’s population, while most high-ranking government officials are
Sunni (Lulu 2011; Valeri 2018: 168; U.S. Central Intelligence Agency 2019a). Al-Wefaq
almost immediately emerged as the biggest opposition group. In 2002, the new group led
175
“widespread protest” (Zweiri 2007: 5). Menon (2002) observes that “the Shia political
body, Wefaq, has emerged by far as the strongest political organisation in the kingdom.”
Throughout the 2000s and 2010s, observers continued to report that Al-Wefaq was the largest
opposition group. Abouyoub (2009) refers to it as “the most influential political society”
and “the largest political society both in terms of its membership and its results at the
polls.”
9
8
Lansford (2017) describes Al-Wefaq as “the country’s largest opposition group.”
In July 2016, a court ordered Al-Wefaq dissolved (Agence France Presse 2016a) and the
group ceased to operate thereafter (EIU CR Jan 2018: 20). Al-Wefaq is thus coded as the
main opposition group in Bahrain in 2002-2016 and a significant opposition group in 2017.
EGYPT
Under the rule of Hosni Mubarak (1981-2011), the Ikhwan al-Muslimoun (Muslim
Brotherhood—MB) constituted the main opposition to the Egyptian government. The MB
continued to be the most popular opposition organization under the transitional military
government (led by the Supreme Council of the Armed Forces or SCAF) that followed
Mubarak’s ouster in 2011. Egypt briefly transitioned to democracy in 2012, with MB
victories in the legislative and presidential elections. After the elected government was
ousted by the military the following year, the coup leader, General Sisi, quickly crushed the
Brotherhood, effectively ending its ability to organize collective action nationally.
8
Likewise, Crystal (2014: 174) reports that “Al-Wifaq is the largest society with some 65,000 members.”
The Economist Intelligence Unit (2002a: 8-9) reports: “Al-Wefaq, which has links with the BFM, has
emerged as the main opposition group.” BBC News (2006) refers to Al-Wefaq as “the main Shia opposition
group,” in a country where Shia are a large majority of the population. Economist Intelligence Unit (2006)
refers to Al-Wefaq as “the main opposition group” (3) and “the biggest political group in Bahrain” (12).
Agence France Presse (2016a) describes Al-Wefaq as the “main opposition group” in Bahrain. Valeri (2018:
166) describes Al-Wefaq as “the most important Bahraini political society.”
9
176
The Muslim Brotherhood was founded in Egypt in 1928 and banned by Nasser in 1954
(Economist Intelligence Unit 1996a: 8). Nonetheless, the group was generally tolerated by
Mubarak. Al-Anani (2016: 155) writes that “the Brotherhood was the key opposition group
in Egypt under Mubarak.” The MB “saw its membership soar” in the 1980s, as Mubarak
adopted a more tolerant stance toward the group than his predecessors had (Mattar 2004:
1622). Technically running on the tickets of the Socialist Labor and Socialist Liberal parties
in the 1987 elections to the People’s Assembly (the national legislature), the MB won 35 out
of 448 elected seats (Sela 2002: 624; Inter-Parliamentary Union 2019), making it “the biggest
opposition in the new parliament” (Ranko 2015: 109). Furthermore, the organization won
elections to control the boards of many of the country’s professional syndicates in the early
1990s (Goldschmidt 2013: 284; Ranko 2015: 109).10 Thus, by the 1990s, the Economist
Intelligence Unit (1996a: 8) could write: “The main challenge to the regime comes from the
Islamic trend, which can draw on a significant groundswell of popular support. The largest,
best-funded and organised Islamic group is the Muslim Brotherhood.”
The MB won 17 seats in the 2000 elections and 88 seats in the 2005 elections (Goldschmidt
2013: 284).11 Thus, Mattar (2004: 1622) refers to the MB during the 2000s as “the most
important representative of the Egyptian masses.” The Economist Intelligence Unit describes
the group as “the best supported of the opposition movements, with a national reach” (2008c:
7) and “the only opposition force able to mobilise popular support” (2010b: 4).
After the uprising that overthrew Mubarak, Egypt was ruled by the Supreme Council of the
Armed Forces (Sullivan and Verticelli 2017). Trager (2011) notes that, under the interim
SCAF administration, the MB remained “Egypt’s most cohesive political movement, with
10
Sela (2002: 624) writes that “the Brotherhood won almost complete control” of the “professional and
worker’s syndicates.”
11
On the Brotherhood’s 2005 result, Ayubi, Kéchichian, Sullivan, Lawson, and Boulby (2009) point out:
“What was even more remarkable was that this gain was made despite clear violations of process, including
the physical prevention of citizens from casting their ballots, and the arrest of thousands.”
177
an unparalleled ability to mobilize its followers,” noting that the group had approximately
600,000 members despite its “highly selective membership process.”
12
The MB launched a
formal political party, the Freedom and Justice Party, which quickly became “Egypt’s most
popular political party” (Goldschmidt 2013: 284).13 This status was confirmed in 2012,
when the Brotherhood won a large plurality of seats in the People’s Assembly and the MB
candidate, Mohammad Morsi, won the presidential election (Sullivan and Verticelli 2017).
However, a military coup in July 2013, led by General Abd al-Fatah al-Sisi, removed Morsi
from power (ibid ). The military swiftly cracked down on the MB, imprisoning most of the
top tiers of its leaders and driving the rest into exile by the end of 2013 (Trager 2016: 230).14
Thus, the new military government “shattered the group’s hierarchy” (Kirkpatrick 2014)
and “effectively decapitated the Brotherhood” (Trager 2016: 230).15
Although local protests by Brotherhood supporters continued in the face of this heavy
repression in the years after the coup (Abdelaziz 2014, Benhaida 2014, Al-Anani 2016), the
removal of the group’s leadership ended the MB’s ability to coordinate protests nationally.
El-Sherif (2014: 7) reports that “the regime’s ongoing suppression of the group has forced
the Brotherhood to decentralize,”
which “limits the Brotherhood’s ability to mobilize
resources nationwide and maintain cohesion since the group can no longer coordinate
across Egypt.” Al-Anani (2017) reports that “the Brotherhood has been largely paralysed
and lost much of its organisational and mobilising capabilities.”
16
12
Trager (2011) elaborates: “In the months since Mubarak’s resignation, the Muslim Brotherhood has
continued to demonstrate its unique capacity to mobilize supporters. Protests continue to be held in Tahrir
Square on most Fridays, and those protests that are endorsed by the Muslim Brotherhood draw substantially
larger crowds than those that are not.”
13
The Freedom and Justice Party is treated here as an extension of the Muslim Brotherhood, rather than
a separate organization.
14
Specifically, Trager (2016: 230) writes:By the end of 2013, most of the Brotherhood’s Guidance Office
was in prison, and its other top leaders were either in exile or hiding. The regime similarly arrested many
of the Brotherhood’s General Shura Committee and provincial leaders, thereby shutting down the top two
tiers of the Brotherhood’s chain of command and disrupting its ability to make and execute decisions.
15
In addition to decimating the MB’s leadership, Sisi’s government also “imprisoned more than 16,000 of
its supporters and killed more than a thousand others at street protests” (Kirkpatrick 2014).
16
Multiple sources argue that the combination of the MB’s record in office and the Sisi regime’s post-coup
media campaign against the group have substantially reduced public support for the Brotherhood (Economist
178
Thus, the Brotherhood’s political influence was essentially destroyed by the end of 2013, and
no other opposition group took on a comparable role: “The Muslim Brotherhood is crushed
and largely discredited, while secular opposition forces are marginalised and in disarray”
(Economist 2014). Trager (2016: 234) writes: “The Brotherhood no longer represents a
political challenge for al-Sisi’s regime.” Brennan (2018) argues that the Muslim Brotherhood
is “effectively vanquished” and more generally, “the opposition has been silenced.”
17
Egypt is excluded from the opposition dataset for 2013, because the country is coded as a
democracy.
In 1992-2012, the Muslim Brotherhood is the main opposition group.
In
2014-2017, under the regime headed by Sisi, the Muslim Brotherhood is one of several
significant opposition groups.
The following organizations constitute the rest of the
significant opposition.
Harakat Shabab 6 Abril (April 6 Youth Movement) was founded in 2008 to advocate for
workers’ rights (Public Broadcasting Service 2019). The group has consistently opposed the
post-2013 military regime (Guardian 2013). Having led the organization of the January 25,
2011 protest against Mubarak (Radio Free Europe 2011; Public Broadcasting Service 2019),
April 6 is coded as a significant opposition group during 2014-2017.
La lil-Muhakamat Al-Askeriyya lil-Madaniyin (No to Military Trials for Civilians) is
a human rights organization founded in 2011 to oppose the prosecution of civilians through
military courts and continued to advocate for this cause after the coup (Al Arabiya 2013;
Cousin 2014). The group staged a protest against the SCAF government in July 2011 (Daily
Intelligence Unit (2014a: 3; Al-Anani 2016: 161).
17
Lynch (2016) writes of the Muslim Brotherhood:Its organization has been shattered, with its leadership
either in prison, exiled or dead and the survivors divided between multiple power centers inside Egypt
and abroad. It is no longer deeply embedded in society. . . It no longer has a robust internal organization,
vast financial resources, a clearly defined ideology, or a tightly disciplined membership.Although the
Economist Intelligence Unit (2018f: 4) counts the Muslim Brotherhood among “the government’s most
active opponents,” it acknowledges that it has been “much-weakened” by the government’s crackdown.
179
News Egypt 2011), so it is coded as a significant opposition group during 2014-2017.
Al-Gabhat As-Salafiyya (Salafist Front) is a fundamentalist Islamist group founded after
Mubarak’s ouster (Wedeman 2012). The Salafist Front organized an anti-government protest
in November 2014 (Shehata 2015) and remained critical of Sisi’s government through 2017
(Salafist Front 2019a), so it is coded as a significant opposition group during 2015-2017.18
JORDAN
The Muslim Brotherhood, the dominant organization within the Islamic Action Front
coalition, was the main opposition group in Jordan throughout 1992-2017.19 Although the
group has often sought tactical cooperation with the Hashemite monarchy,20 Schwedler
(2015) observes: “Since the late 1980s, the Islamic Action Front has been the leading
opposition movement in Jordan, often boldly challenging the monarchy’s domestic and
foreign policies.” The group has sought democratic reforms, an overhaul of the education
system along Islamic lines, and a more interventionist economic policy, and they have been
particularly critical of what they see as the state’s overly-accommodating relationship with
Israel (Hamid 2013; Suwaed 2016: 260).
The Jordanian Muslim Brotherhood was founded in 1946 and through the 1970s was highly
18
The group has remained opposed to the post-coup regime since. For example, in an April 6, 2019 post
to its Facebook page, an administrator of the page writes (translation by Facebook): “I am here to stress
that the # front continues to reject and fight the regime, within its revolutionary framework and the
well-known attitude that has not changed . . . and that we will be at the forefront of the nation and the #
Egyptianpeople if they move to bring it down” (Salafist Front 2019c).
19
On the dominance of the Muslim Brotherhood within the Islamic Action Front, Schwedler (2015) writes:
“A year after its creation the Front was regarded as the de facto political party of the Muslim Brotherhood.”
20
As Schwedler (2015) notes, the IAF has referred to itself as a “loyal opposition.”
180
supportive of the monarchy (Tachau 1994: 290). However, after a stunning electoral victory
in 1989 that made them the largest group in parliament (Tachau 1994: 290; Seddon 2004:
382; Hiro 2013: 446), the MB shifted toward a more oppositional stance. After that election,
“the Muslim Brotherhood formed the largest bloc and the real opposition in parliament—
their allies controlled more than one-third of the legislative body—and disagreed with the
government on a number of important policies” (Seddon 2004: 382).21 From this new position
of strength, “in the 1990s, the party became a menace to the regime since it adopted the
coloring of a political movement and provided a nonleftist alternative for opposition politics”
(Tachau 1994: 290). Although five Muslim Brotherhood politicians were given cabinet
positions in January 1991, they refused to join the new cabinet formed in June 1991, because
of their disagreement with the king’s rapprochement with Israel (Brinkely 1991; Tachau 1994:
290) and subsequently they have rejected all offers to join the government (Hamid 2013).
Shortly after political parties were legalized in late 1992, the Islamic Action Front was
founded as an official political party, with the Muslim Brotherhood at the helm (Anzalone
2009).
The Muslim Brotherhood/Islamic Action Front has remained the most popular Jordanian
opposition group since the early 1990s. The Economist Intelligence Unit (1996a) describes
the MB as “the largest political movement in the country” (8) with “a large grass-roots
following” (7). This assessment continued in the 2000s. The Economist Intelligence Unit
(2002d: 9; 2008d: 13), describes the MB as the “only group with significant popular support.”
Likewise, Anzalone (2009) refers to the IAF as “Jordan’s most influential political opposition
movement.”22 The Brotherhood’s political sway has persisted into the 2010s. In 2011, the
21
The Economist Intelligence Unit (1996e: 4) writes: “The election for the National Assembly in November
1989, which was regarded as free and fair, saw the emergence of Islamists as a significant and well-organised
body in Jordanian politics. With around 34 deputies out of 80 either aligned with, or sympathetic to, the
Islamists, the group had eclipsed radical nationalists and leftists as the chief opposition to the regime and
its traditional supporters.”
22
Although the IAF’s parliamentary representation declined to six seats in the 2007 elections, Hamdi
(2013) observes that this setback was likely the result of manipulation by the regime: “The 2007 elections,
arguably the most fraudulent in Jordan’s history, marked the culmination of regime efforts to marginalize
181
group appears to have pressured King Abdullah II into calling for Syria’s President Assad
to step down (Hiro 2013: 447). Hamid (2013) describes the IAF as “Jordan’s largest and
most organized political party.” The Economist Intelligence Unit (2016: 3) calls the MB
“the dominant Islamist opposition group” and (2018c: 5) claims that “the IAF is the most
successful party” in Jordan. In December 2017, the MB led “the largest protest Jordan has
seen in nearly a decade” (Luck 2017).
Thus, I code the Muslim Brotherhood as the primary opposition movement in Jordan
throughout 1990-2017. (Although the group participated in the cabinet in 1991, they left
before half a year had elapsed, so I do not code that year as an interruption in the MB’s
opposition status.)
KUWAIT
In Kuwait in 1992-2017, there was no one opposition group that was clearly the most
popular.
However, the country did have several groups that are coded as significant
opposition organizations.
The Muslim Brotherhood has been active in Kuwait at least since the 1960s, when the
Brotherhood-affiliated Social Reform Society (Jam‘iyya al-Islah al Ijtima‘iyya) was founded
(Tachau 1994: 91). In 1991, Kuwait’s Muslim Brothers founded the Islamic Constitutional
Movement (al-Haraka ad-Dusturiyya al-Islamiyya or Hadas [Schmidmayr 2010: 159]).23
the Islamic movement.”
23
Although Kuwait does not officially allow political parties, HADAS and other political coalitions function
effectively as parliamentary parties (Lansford 2017).
182
Hadas advocates both Islamic cultural values, the implementation of Shari‘a as the basis of
the legal system, and electoral reform (Tachau 1994: 77; Swiney 2009). Hadas candidates
won more than 10% of the vote in parliamentary elections October 1996 (Herb 2019). Thus,
Hadas is coded as a significant opposition organization for the years 1997-2003.
Kafi (Enough) is a youth group that formed in early 2011 to call for the resignation of the
prime minister and for greater political freedoms (Murphy 2011; Reuters 2011b. Kuwaiti
protests on Tuesday aim to remove PM). In March 2011, Kafi co-led a protest against the
government with as-Soor al-Khames (Fifth Wall), another youth group that sought the
prime minister’s resignation and that started in early 2011 (The Gulf 2011). Each group is
coded as a significant opposition group in 2011-2017.
Nahj (Path) is a coalition of young Islamists, formed in 2011, that advocates constitutional
monarchy (Agence France Presse 2012; Ghabra 2014; Dazi-Heni 2015). The group organized
an anti-government rally in August 2012 (Agence France Presse 2012), and thus is coded as
a significant opposition group during 2011-2017.
Karamat Watan (A Nation’s Pride or A Nation’s Dignity) was formed in October 2012
to oppose changes by the emir to the electoral system and to call for parliament to have
the right to appoint the prime minister (Albloshi and Herb 2018: 414). That same month
the group organized an anti-government demonstration (Albloshi and Herb 2018: 418), so
Karamat Watan is coded as a significant opposition group during 2013-2017.
Kutlat Al-Amal Ash-Shabii (Popular Action Bloc) is a “populist” and “nationalist”
electoral coalition (Lansford 2017). A large protest against the conviction of a leading
Popular Action Bloc figure, Musallam al-Barrak, occurred in April 2013 (Ingram 2013a).
Thus the bloc qualifies as a significant opposition group during 2013-2017.
183
MAURITANIA
The Union des Forces Démocratiques (Union of Democratic Forces—UFD) is coded
as the main opposition party for the years 1992-1996. UFD was authorized in October
1991, and in its first public statement it called on Taya to resign so that a government
of national unity could oversee truly democratic elections (Pazzanita 2008: 511).
24
In the
January 1992 presidential election, the UFD candidate, Ahmed Ould Daddah won almost
33% of the vote, despite alleged electoral interference by the ruling party (Pazzanita 2008:
512). “Before long, it was apparent that the UFD was the most significant opposition group
facing the incumbent Mauritanian president’s” party (Pazzanita 2008: 511) and it remained
“Mauritania’s most prominent opposition party during much of the 1990s” (Pazzanita 2008:
409-410). In its 1996-1997 Country Profile of Mauritania, the Economist Intelligence Unit
(1997: 29) observed: “The Union des forces démocratiques (UFD) is generally considered to
be the second political force in Mauritania” after the ruling party. Both BBC Monitoring:
Middle East (1996a) and Phythian (1996) refer to UFD as the “main opposition.”
However, a new party, Action pour le Changement (Action for Change—AC) was founded
in August 1995 and “relatively quickly surpassed the UFD’s prior level of backing among
the politically aware populace” (Pazzanita 2008: 513). The Economist Intelligence Unit
(1997: 32) speculated that AC leader Messaoud Ould Boulkheir would pose “a more serious
challenge” to Taya in the upcoming presidential election. However, some sources (BBC
Monitoring: Middle East 1998b; BBC Monitoring: Middle East 2000a) continue to refer to
the UFD as the “main opposition.” Given this mixed evidence, I stop coding the UFD as the
single most popular opposition group after 1996 and code all significant opposition groups
in subsequent years.25
25
Pazzanita (2008: 187) argues that AC “evolved into Mauritania’s most significant and most broadly
based political party, which opposed President Ould Taya,” but other sources do not confirm this, so I do
184
The UFD — which was reconstituted as the Rassemblement des Forces Démocratiques (Rally
of Democratic Forces—RFD) in 200126 — is coded as one of these significant opposition
groups throughout 1997-2004 and 2009-2016, based on having been the main opposition
group until 1997, co-organizing an anti-government demonstration in May 2009 (Agence
France Presse 2009d), and its presidential candidate, Ahmed Ould Daddah, winning more
than 10% of the vote in July 2009 (International Foundation for Electoral Systems 2019).27
(Mauritania is excluded from the dataset in 2007 and 2008, because it is coded as a democracy
for those years.)
Front des Partis de l’Opposition (Opposition Parties Front—FPO) is a coalition of five
opposition parties established in March 1997 (BBC Monitoring: Africa 1997). The coalition
organized anti-government demonstrations in October 1997 (Reuters 1997b. Mauritanian
opposition stages protest march) and January 1999 (BBC Monitoring: Middle East 1999).
The FPO is therefore coded as a significant opposition group during 1998-2005.
Front National pour la Défense de la Démocratie (National Front for the Defense
of Democracy—FNDD) is a coalition of parties and civil society organizations that formed
to oppose the 2008 coup (Agence France Presse 2008e) The coalition organized a protest in
the capital against the coup in August 2008 (Agence France Presse 2008d). The coalition’s
candidate for president in the July 2009 election won over 10% of the vote (Agence France
Presse 2009c; Economist Intelligence Unit 2009b). The coalition is thus coded as a significant
opposition group in 2009-2016.
not count the AC as the main opposition party at any point.
26
The Mauritanian government banned the UFD in 2000. The following year, the former leaders of the
UFD established the RFD as a “successor party” (Pazzanita 2008: 426).
27
Although the RFD was supportive of the military’s ouster of the elected government in 2008 (Maimone
2019), the party opposed the plans of members of the junta to run in subsequent elections (Agence France
Presse 2008b) and by February 2009, the RFD was calling for the junta to resign in favor of a transitional
government (All Africa 2009a). Thus, I still code the RDF as oppositional in 2009-2017. Note that Ould
Daddah also won more than 10% of the vote in the first round of the March 2007 presidential election
(International Foundation for Electoral Systems 2019).
185
Coordination de l’Opposition Démocratique (Coordination of the Democratic
Opposition—COD) began in December 2009, when it organized a large protest over price
increases and the authoritarian nature of the regime (Lansford 2017). The coalition also
organized a large anti-government demonstration in July 2012 (Agence France Presse
2014). The coalition of numerous parties seeks to promote a “peaceful struggle” to force
Abdelaziz from power (BBC Monitoring: Middle East 2013). The coalition is coded as a
significant opposition group for 2010-2017.
Rassemblement Nationale pour la Réforme et le Développement (National Rally
for Reform and Development—RNRD, known in Arabic as at-Tajmua‘ al-Watani lil-Islah
wa at-Tanmiya, or Tewassoul for short) is an Islamist party formed in 2007 (Lansford 2017).
Although Tewassoul allied itself with Abdel Aziz after he was elected president in 2009 (All
Africa 2009b), the party joined the COD in 2011 (All Africa 2011d). The party organized
an anti-government demonstration in February 2013 (BBC Monitoring: Middle East 2013b).
Tewassoul is therefore coded as a significant opposition group during 2013-2017.
Forum National pour la Démocratie et l’Unité (National Forum for Democracy and
Unity—FNDU) is an umbrella coalition of parties and trade unions that was formed in
early 2014 to call for a boycott of the upcoming presidential election (Lansford 2017) and
has advocated for electoral reforms in general (PANAPRESS 2014b.“Mauritania: Radical
opposition, civil society demand direct dialogue with gov’t.” ). The coalition staged large
demonstrations calling for greater electoral transparency in June 2014 (PANAPRESS 2014a.
Mauritania: Opposition protests ahead of 21 June election). The FNDU is therefore coded
as a significant opposition group for 2014-2017.
186
MOROCCO
The banned Islamist group al-Adl wal-Ihsan (often shortened to Ihsan, the group’s name
is variously translated as Justice and Charity, Justice and Benevolence, and Justice and
Spirituality) has consistently opposed the Moroccan government since before the 1990s.
There is some ambiguity about whether, in the early 1990s, Ihsan had a greater following
than two legal, secular parties: the Parti Istiqlal (PI, the Independence Party, usually
referred to as Istiqlal) and the Union Socialiste des Forces Populaires (USFP, Socialist
Union of Popular Forces). However, these two parties entered the government in 1998,
making Ihsan the clearly most significant opposition group subsequently. Although both
Istiqlal and the USFP returned to opposition status at various points in the 2010s, sources
generally agree that the Ihsan remained the most popular opposition group in these years.
Sources mention two major Islamist opposition groups that were active in Morocco in the
1990s: Ihsan and the Parti de la Justice et du Développement (PJD: Party of Justice and
Development). Ihsan was founded in 1987 (Laremont 2009), but its leader, Abdesselam
Yassine, had been politically active since the 1970s (Boum and Park 2016: 25). The PJD
was founded in 1998 but it was the successor to at-Tawhid wal-Islah (Unity and Reform),
which was founded in 1992 and was itself the successor to al-Jama’a al-Islamiyya (Islamic
Group), which began in the early 1980s (Willis 1999: 45-47). Ihsan has consistently been seen
as the more radical of the two groups (Financial Times 1997a. Moroccans Learn Algeria’s
Lessons). Ihsan rejects the legitimacy of the monarchy and thus has always refused to take
part in elections (Varadarajan 1996; Laremont 2009; Partrick 2019a), while the PJD has
consistently participated since 1997 (Partrick 2019a).28
28
In 1997, the activists who would found the PJD ran under the banner of the Mouvement Populaire
Constitutionelle et Démocratique (MPCD, Constitutional and Democratic Popular Movement [Partrick
2019a]).
187
Some sources refer to these groups jointly as Morocco’s main Islamist opposition groups in
the 1990s (Reuters 1994d. “Islamic fundamentalist groups active in Morocco” ; Willis 2009).
However, most sources assert that Ihsan has always had a larger following than the PJD or
its predecessors. (No sources found made the opposite claim.) For instance, Layachi (2009)
relates that, of all the Islamist groups in Morocco in the early 1990s, “the most popular one
was that of the al-’Adl wa al-Ihsan.” This view is not merely one made in hindsight, but
was voiced by contemporary news accounts. For instance, (Africa Confidential 1991a: 2-3),
refers to Ihsan as the “largest movement” among Islamist groups29 and the Financial Times
(1995) also refers to Ihsan as “Morocco’s largest Islamic movement.”
30
In absolute terms, there is evidence that Moroccan Islamists had amassed considerable
support by the 1990s. Africa Confidential (1992: 5) reports that “the Islamist movement is
giving cause for concern” to the Moroccan monarchy. The Economist Intelligence Unit
(1996b: 6) says: “In recent years the inability of the moderate opposition parties to win
genuine participation in government has fostered the emergence of illegal opposition
movements. The most feared of these are the Islamic dissidents.”
31
And there is evidence
that Ihsan in particular had a large support base by the 1990s. Yassine had recruited
“thousands” of followers by the mid 1980s and during that decade “Yassin became a
popular symbol of antiestablishment Islam” (Tachau 1994: 407). Willis (2007, 156) argues
that, by the 1990s, “it was clear to the regime that by far the biggest and best organized
movement in the country was of Yassine’s al-Adl wal-Ihsan.” Ihsan’s strength in this period
was reflected in its ability to organize mass collective action. For example, to protest the
trial of 15 activists in 1991, Ihsan organized a sit-in at the courthouse in Casablanca by
29
This assertion is repeated in other Africa Confidential articles over the next few years, including in 1993
(5) and 1994 (7).
30
Middle East Economic Digest (1992e ) describes Ihsan’s leader, Yassine, as “the kingdom’s best-known
Islamist leader” and the Economist Intelligence Unit (1996b: 7) mentions Yassine among Morocco’s “main
political figures.”
31
Africa Confidential (1990: 6) predicts that, if Morocco held free elections, “Islamists would win
substantial votes.” Africa Confidential (1991a: 2-3) reports that 30,000 Islamists took part in a demonstration
in February 1991 against the Western intervention against Iraq.
188
about 2,000 participants, followed by a march which drew 5,000 (Reuters 1991d. Morocco
Offers to Legalise Moslem Party on Certain Conditions).32
However, some sources claim that Moroccan Islamists generally, and Ihsan in particular, were
not yet a significant force in the early 1990s. The Middle East Economic Digest (1991a. A
critical year for the kingdom) and Africa Confidential (1993) describe Islamist opposition
in Morocco as “limited.”
33
Tessler, Entelis, and White (1995: 380) argue that “Islamist
opposition does not enjoy mass support” (although “its presence is growing” ). Regarding
Ihsan specifically, Tachau (1994: 407) writes: “Observers have differed in their sense of
the ability of movements such as Yassin’s to attract widespread support given the plethora
of legal Moroccan parties and the real Islamic legitimacy King Hassan has enjoyed among
important segments of the Moroccan population.” Waltz (1995: 125) argues that by the
mid-1990s no Moroccan Islamist movement, including Ihsan, had “grown to the proportions
known by the FIS in Algeria or al-Nahda in Tunisia.”
34
At the same time, some sources point to Istiqlal and the USFP as the main opposition
forces in the early 1990s.
The Economist Intelligence Unit (1996b: 6) writes: “The
opposition is dominated by two moderate left-wing parties, the Union socialiste des forces
populaires (USFP) and the Parti Istiqlal (PI).”
35
The USFP and Istiqlal accounted for
almost all the seats won by opposition parties in the 1984 and 1993 (consecutive)
parliamentary elections (Tessler, Entelis, and White 1995: 381).36
32
Both parties have
When six of Ihsan’s leaders were arrested in 1990, about 2,000 supporters demonstrated in Rabat (Boum
and Park 2016: 25).
33
Africa Confidential (1994: 6) also refers to a “lack of a visible Islamist challenge.”
34
Waltz (1995: 126) notes that Ihsan’s influence was hindered in the early 1990s by the imprisonment for
two years of Ihsan’s “complete executive bureau.”
35
Legum (1992a: B487) also describes the USFP and Istiqlal as the “main opposition” parties.
36
Sources disagree about which of the two parties was more popular in the 1990s. Tachau (1994: 402)
claims: “The Istiqlal is historically Morocco’s most important party, and it remains, with the Socialist Union
of Popular Forces (USFP), a vital political unit for the legal articulation of opposition views.” Lust-Okar
(2005: 183) by contrast, argues that the USFP was “considered the most important legal opposition party”
around this time.
189
tended to support social-democratic policies, the Istiqlal with a nationalist and moderately
Islamist bent (Economist Intelligence Unit 1996b: 6; Lust-Okar 2005: 179), and both have
advocated gradual democratic reforms (Maghraoui 2001: 81). The USFP and Istiqlal have
their own trade unions (Economist Intelligence Unit 1996b:
anti-government collective action.
6) and have supported
For example, both parties’ unions were involved in
general strikes in 1990 (Tachau 1994: 420; Boum and Park 2016: 392).
However, in February 1998, the USFP and Istiqlal entered a coalition government, led by the
USFP’s Abd al-Rahman al-Youssoufi as prime minister and alongside other smaller parties
(Partrick 2019a). Meanwhile, the influence of Ihsan only continued to grow during the 1990s
and early 2000s. “From 1989 until 2000, Yassine was under house arrest but his movement
grew bigger and his reputation became global” (Bouasria 2015: 104). The Bouzerda (2000b.
“Morocco’s Islamist chief free, group still shackled.” ) describes Ihsan as “Morocco’s main
Moslem fundamentalist group” and the Economist Intelligence Unit (2002b: 8) describes
Ihsan as “Morocco’s largest Islamist group.” Reuters (2000) describes it as not only “the
country’s biggest Islamist group,” but “the main opposition movement” more generally.37
In 2002, the PJD performed surprisingly well in the parliamentary elections, increasing its
seats from 9 to 42, making it the third largest party in parliament (Inter-Parliamentary Union
2019).38 In 2007, the PJD increased its representation to 46 seats (ibid ). Nonetheless, Ihsan
appears to have remained the largest opposition movement throughout the 2000s. Karam
(2003) describes Ihsan as “the main source of popular opposition to Morocco’s monarchical
government.” Multiple sources refer to Ihsan, not the PJD, as the most popular Islamist
group in this decade (Fakihani and Pettit 2003; Ghanmi 2004; Thorne 2006; Charlton 2007;
Economist Intelligence Unit 2008a: 12; Laremont 2009). (No sources I found make the
37
Bouzerda (1999) describes Yassine as Morocco’s “most prominent Islamist leader” and Legum (2000:
B639) writes: “Yassine commands respect and support in a significant portion of Moroccan society.”
38
Moreover, it was not far behind Istiqlal — 48 seats — and the USFP — 50 seats (Inter-Parliamentary
Union 2019).
190
opposite claim or report any non-Islamist opposition group being more popular than Ihsan.)
From 2011-2017, the PJD led coalition governments, while the USFP and Istiqlal were each
out of government for part of this period, the former exiting in 2011 and returning in 2017,
the latter exiting in 2013 and not returning by the end of 2017 (Economist Intelligence Unit
2008a: 9-10; Middle East Reporter 2012; Benmehdi 2013; Al Jazeera 2017c. Morocco’s king
names new coalition government; Partrick 2019a). Yet the USFP had suffered major losses
in parliament by the time it left government in 2011, slumping from first place in 2002 to
fifth place in 2007 and 2011 and sixth place in 2016 (Inter-Parliamentary Union 2019). The
Parti Authenticité et Modernité (PAM, Authenticity and Modernity Party), founded in 2008
(Boum and Park 2016, 389), became a major electoral player in the 2010s, winning the fourth
largest bloc in the 2011 elections and reaching second place in 2016 (Inter-Parliamentary
Union 2019). Some sources describe the PAM as the “main opposition” party during this
period (BBC Monitoring: Middle East 2016; Agence France Presse 2016b. Opposition vows
to rid Morocco of Islamists in Oct vote), yet this designation seems to refer to the PAM’s
electoral rivalry with the PJD rather than any reformist agenda. The PAM was founded
by an advisor to King Mohammed VI (Boum and Park 2016: 389) and seems designed
to advance royal interests, with sources describing it as “loyalist” (Masbah 2014), “palacealigned” (Economist Intelligence Unit 2017a.: 3), and “close to the palace” (Naoum 2017).
Istiqlal, which has been more successful than the USFP — it came in second in 2011 and
third in 2016 (Inter-Parliamentary Union 2019) — while retaining more distance from the
palace than the PAM,39 seems to have the best claim to lead the legal opposition during this
period, after it left the governing coalition in 2013.40
Nonetheless, Ihsan remained the largest opposition group throughout the 2000s and 2010s.
39
For instance, the Economist Intelligence Unit (2017b: 17) lists the PAM among the “loyalist” parties,
but not Istiqlal.
40
On the other hand, Boum and Park (2016, 392) argue that the Istiqlal has been characterized throughout
its history by “a continued opposition to any major political change.”
191
At the time of his death in 2012, Yassine was still regarded as “the country’s biggest
opposition figure” and tens of thousands attended his funeral (Schemm 2012).
The
movement persisted after its founder’s passing, remaining “the most popular and powerful
Islamist movement in Morocco” (Bouasria 2015: 100). Throughout the 2010s, sources
continued to describe Ihsan as the largest Islamist movement (Economist Intelligence Unit
2010a: 25; Schemm 2013; Sakthivel 2014; Economist Intelligence Unit 2016a: 17). Multiple
sources also describe Ihsan as the main opposition group overall in this period. Schemm
(2013) argues that Ihsan “has largely become the voice of the opposition.” Motaouakal
(2014: 101) describes the group as “the best organised social and political movement in
Morocco.” Reuters (2016) describes Ihsan as “the main opposition.” Bouasria (2015: 100)
estimated Ihsan’s membership to be in the tens of thousands by the 2010s, while others
sources estimate it to have had hundreds of thousands of supporters since as early as the
mid 2000s (Thorne 2006; Economist Intelligence Unit 2008a: 12; Schemm 2013) or even
nearly a million by the end of that decade (BBC Monitoring: Media 2009a “Morocco:
Websites of Islamic movement inaccessible since 17 January.” ).41 The group was “a major
player in the 2011 protests” (Errazzouki 2017) and in 2017 led the “largest protests since
2011” in Morocco’s capital, with at least tens of thousands of participants (BBC
Monitoring: Middle East 2017).
Given the ambiguous evidence for the early 1990s, there is no single main opposition group
at the start of the coding period. However, sources point to Istiqlal, the USFP, and Ihsan as
the main opposition forces in the 1990s and the first two entered government in late 1997.
Among the remaining opposition groups, there was a clear consensus that Ihsan was the
most popular by this time. By the time the USFP and Istiqlal left the governing coalition
in the 2010s, they clearly had less support than Ihsan. Thus Ihsan is coded as the main
41
The Economist Intelligence Unit (2008a: 12) specifically addresses the group’s capacity for collective
action, relating: “It has hundreds of local branches and can mobilise hundreds of thousands of supporters.”
Schemm (2013) indicates the geographic and socioeconomic breadth of its support, noting that Ihsan “can
be found across the country, with members ranging from blue collar workers to doctors and engineers.”
192
opposition organization for the years 1998-2017.
Throughout 1992-1997, each of these three parties is coded as a significant opposition
group. The USFP and Istiqlal co-led a large anti-government demonstration in February
1991 (Reuters 1991c.
Morocco Bans Mass Pro-Iraqi Demonstration in Casablanca).42
Ihsan’s supporters engaged in large demonstrations in May 1990 in response to the trial of
six of the group’s leaders (BBC Monitoring: Middle East 1990; Boum and Park 2016: 25)
and the group organized another anti-government protest in August 1991 (Reuters 1991e.
Moslem Fundamentalists March Through Casablanca Streets).
Bloc Democratique (Democratic Bloc, known in Arabic as Al-Koutlat At-Tajammu’a AdDimuqrati or Koutla, for short) was an electoral coalition formed in May 1992 by Istiqlal,
USFP, and three smaller parties. It also called for reduction in minimum ages to vote and
run for office (Daadaoui 2010: 197; Boum and Park 2016: 385). The parties of the coalition
won a combined 32.5% of the vote in the June 1993 election (Montabes Pereira and Parejo
Fernández 2003; Interparliamentary Union 2019). Thus the Koutla is coded as a significant
opposition group for 1993-1997.
SUDAN
The Umma (Community) Party (UP, later known as the Umma National Party), was the
main opposition group in Sudan during the period 2000-2017. The UP is an Islamist party.
The UP and another Islamist party, the Democratic Unionist Party (DUP) are typically
42
Each party also won over 10% of the vote in the 1993 parliamentary elections (Montabes Pereira and
Parejo Fernández 2003).
193
described as having been the largest two parties in recent decades, other than the ruling
National Islamic Front (NIF, later known as the National People’s Congress). For instance,
Tachau (1994: 488) writes: “Both the Umma and PDP/DUP have been able to draw upon
the mass following and hierarchy of their associated sects.” The Economist Intelligence Unit
(1996c: 6), notes that: prior to the NIF regime, “Sudan’s periods of parliamentary rule have
been dominated by two political parties: the Umma Party and the Democratic Unionist
Party (DUP).”
43
Elbagir and Karimi (2012) refer to the UP and DUP as “Sudan’s two
largest opposition parties.”
44
The UP won the most seats in the 1986 parliamentary elections, while the DUP finished in
second place (Tachau 1996: 486-487). The leader of the UP, Mahdi, became prime minister,
forming a coalition government with the DUP; in 1989, the government was overthrown in
a military coup led by Omar al-Bashir (Woodside 2019).45
Soon after the coup, the UP, DUP, and other opposition groups formed an umbrella
organization called the National Democratic Alliance (NDA), in order to resist Bashir’s
government. The NDA also included armed groups, such as the Sudan People’s Liberation
Army (BBC Monitoring: Middle East 1990a; Berry 2015: 252; Woodside 2019).46 So
although the NDA was “the main opposition group” in the 1990s (Kramer, Lobban, and
43
On the ideological outlook of the UP and DUP, see for example Cudsi and Voll (2009) and the Economist
Intelligence Unit (1996c: 6)
44
However, the Economist Intelligence Unit (2009a: 6) observes that the two parties “have been weakened
and divided by the long rule of the NCP and its efforts to co-opt them, which led to the formation of
pro-government splinter factions. . . ” Another opposition party frequently mentioned in news reports is the
Popular National Congress, founded by Turabi, the former head of the ruling party, after he fell out with
President Bashir. However, the Economist Intelligence Unit (2002b: 6) says that “the group enjoyed little
influence” and “as a political player, [Turabi’s] influence seems to be spent.”
45
It is worth noting that, despite the large following the UP has generally retained, at the time of the
coup, “public confidence in [Mahdi’s] government had dissolved” (Kramer, Lobban, and Fluehr-Lobban 2013:
437).
46
The SPLA was the “main southern military resistance force” and fought the Sudanese state even before
the coup (Tachau 1994: 497). “Since the start of the second civil war in 1983, the SPLA has been the major
armed opposition to successive Khartoum governments” (Economist Intelligence Unit 1996c: 6). In addition
to involving the SPLA, “the NDA controlled its own military wing of northerners” (Kramer, Lobban, and
Fluehr-Lobban 2013: 312).
194
Fluehr-Lobban 2013:
437), the organization and its constituent groups cannot be
considered nonviolent in this period.
However, in March 2000, the UP left the NDA (Pineau 2000). The UP has engaged in
peaceful activism since then, and has refused to accept any cabinet posts (Economist
Intelligence Unit 2002b: 8; Economist Intelligence Unit 2009a: 6-7; BBC Monitoring:
Middle East 2010b; All Africa 2011b; BBC Monitoring: Africa 2013; Middle East North
Africa Financial Network 2015b.
“Sudanese Pres.
announces new cabinet lineup.” ;
Economist Intelligence Unit 2018f: 22; Woodside 2019).47 Throughout this period, sources
indicate that the UP was the most popular opposition group in Sudan. Reuters (2005)
writes that the UP is “widely considered to be Sudan’s largest political party.” All Africa
(2011b.
Country’s Largest Opposition Party Formally Rejects Participation in New
Government) describes it as the “largest opposition party” in Sudan, as does the Sudan
Tribune (2018), while the U.S. Department of State (2011a. “Background Notes: Sudan.” )
describes the Umma Party as the “single largest political grouping” and Agence France
Presse (2018) refers to it as “the main opposition.”
During 1992-1999 there was no significant (nonviolent) opposition group in Sudan.
TUNISIA
Hizb an-Nahda (Party of the Renaissance or Nahda for short), an Islamist movement that
47
In December 2014, the UP issued a joint communiqué with some armed groups, calling for democracy
in Sudan. However, the communiqué did not endorse armed resistance. The UP and the armed groups did
not set up a joint organization and the UP did not become involved in the armed groups’ violent activities
(Eljak 2014; Sudan Tribune 2014; Woodside 2019).
195
sought but was never granted legal recognition as a political party by the Ben Ali government,
was the main opposition group in Tunisia in the early 1990s. The movement was effectively
crushed in 1992, and no other opposition party with a broad base of support emerged before
the regime was overthrown in 2011.
At the time of Tunisia’s parliamentary elections in 1989, the Mouvement des Démocrates
Socialistes (Movement of Socialist Democrats—MDS) was considered “the main opposition
party” (Perkins 2016b: 185). Hizb al-Nahda, though not legally recognized, ran candidates
as nominal independents in most districts (ibid ). Nahda won approximately 13% of the vote
(Murphy 1999: 180), compared to about 4% for the MDS (Montabes Pereira 2003), thus
“replacing the MDS as the main opposition force” (Partrick 2019b).48
In late 1990, however, the Ben Ali government began cracking down on the Nahda movement,
and over the course of 1991, thousands of their members were arrested (Erdle 2010: 109).
In July of 1992, nearly 200 top Nahda activists were put on trial, effectively ending Nahda’s
ability to operate for the remainder of Ben Ali’s time in power (Partrick 2019b). “These
mass trials were seen as the culmination of the Tunisian Government’s long campaign against
al-Nahdah, whose organizational structures within the country were largely destroyed and
its leaders imprisoned or forced into exile” (ibid ). Similarly, Erdle (2010: 109) writes that
after these trials, “the Islamist opposition had effectively ceased to exist as an organized
force on the ground.”
49
Thus, I code Nahda as Tunisia’s main opposition group in 1992,
48
Likewise, Murphy (1999: 180) argues that, after the 1989 elections, “it was clear that Nahda represented
the only genuinely challenging opposition to the RCD,” the ruling party.
49
Perkins (2016b: 17) observes that, by 1992, the Tunisian government “had effectively crushed the [Nahda]
movement.” Willis (2012” 171-172) writes that “By the middle of 1992, authorities in both Tunisia and
Algeria had succeeded in banning and repressing their countries’ main Islamist movements, which a matter
of months earlier had appeared to present serious challenges to the rule and legitimacy of the existing
regimes.” Contemporary sources also gave such an account. For instance, the Middle East Economic Digest
(1992c. “An uneasy balance of political forces.” ) reported in July of that year: “Police action has crushed
Nahda, whose leadership is either exiled, in prison or has dropped out of active politics. ’I honestly see no
point in carrying on,’ says one former Nahda official. ’We entered into direct confrontation wanting to take
over and we lost.’”
196
and a significant opposition group in 1993-1999.
YEMEN
At-Tajammu’a al-Yamani lil-Islah (the Yemeni Congregation for Reform, commonly
known as Islah), is a coalition of Islamists and tribal leaders, and is generally regarded as
the main opposition group during the period from the unification of North and South Yemen
in 1990 to the end of Saleh’s rule (2012). For instance, Suwaed (2016: 552) observes that,
of all the legal political organizations in Yemen since unification, “only three are of national
significance — the GPC [General People’s Congress the ruling party under Saleh], Islaah, and
YSP [the Yemeni Socialist Party]” and claims that “the second largest political organization
in Yemen [after the GPC] is the Islaah Party.” Al Jazeera (2011a) describes Islah as “the
main opposition party in Yemen.” Schmitz and Burrowes (2018: 256) argue that Islah was
“the most formidable challenger to the Saleh regime.”
However, some sources describe Islah’s oppositional status as ambiguous, noting its gradual
evolution from an ally of the Saleh regime in the early 1990s to a foe in the 2000s. For
instance, Schmitz and Burrowes (2018: 256-257) denote Islah as:
the party with the most ambiguous relationship to that regime and its
General People’s Congress (GPC). Regarding the ambiguity, Islah when not
in government has often acted and been treated as if it were; when in the
government, it has often behaved as if in the opposition. In short, it often
has been difficult to describe Islah’s behavior as oppositional, even when it
197
was formally in opposition. . . . Between its creation in 1990 and the 2006
election, Islah had evolved from being something of an uncertain, ambiguous
appendage of President Saleh to being part of an increasingly credible and
formidable opposition.
Specifically, Schmitz and Burrowes (2018: 364) note that: “In the 2000s, Islah led the formal
opposition and was the most important party in the opposition coalition, Joint Meeting
Parties.”
Hamzawy (2009: 1) presents a similar narrative: “Islah did not enter the political scene
as part of the opposition. Rather, it began its participation in 1990 as an ally of the
ruling General People’s Congress (GPC), before turning against it and becoming the leading
opposition party by the end of the decade.”50 Islah was in fact created by members of
the General People’s Congress in 1990 (Durac 2019) and from 1990 to 1997, some of its
representatives held cabinet positions (Hamzawy 2009: 5-7; Schmitz and Burrowes 2018:
257).
However, most sources indicate that Islah entered the opposition after the 1997 parliamentary
elections, when it left the cabinet. For instance, the Economist Intelligence Unit (2002c: 6)
notes that, after 1997, “Islah became the only significant parliamentary opposition.” Schmitz
and Burrowes (2018: 257) write: “In the 1997 elections, having made a slightly poorer
showing than in 1993, Islah went into the opposition.” Hamzawy (2009: 5) observes that “in
50
These accounts are also echoed by Bonnefoy (2007): “Calling al-Islah an opposition party has also
long been inexact, as its relations with the government of President ‘Ali ‘Abd Allah Salih were ambivalent.
Indeed, in the context of the unification of North and South Yemen in the early 1990s, al-Islah became an
alternative ally of the General People’s Congress (GPC), the ruling party, which experienced tensions with
the Socialists. Between 1993 and 1997, al-Islah participated in government, controlling important ministries.
Its leader, ‘Abd Allah al-Ahmar, was consistently elected speaker of Parliament between 1993 and 2007 with
the support of the ruling party, and in 1994, al-Islah militants assisted the national army in its war against
southern secessionists headed by former socialist rulers. Starting at the end of the 1990s and becoming more
overt in the mid-2000s, al-Islah experienced a slow and hesitant transformation, accepting its role as an
opponent of ‘Ali ‘Abd Allah Salih’s rule.”
198
1997 Islah switched sides and joined the Yemeni Socialist Party and other parties in opposing
the GPC.”
51
In one indication of the new discord between Islah and the ruling party, the
GPC nationalized some of the religious schools run by Islah in the late 1990s (Hamzawy
2009: 7).52
However, the relationship between Islah and the GPC remained ambiguous for a time after
1997. As Hamzawy (2009: 6-7) notes, Islah’s transition to the opposition “has been far from
complete because of Islah’s unwillingness to break with the GPC at all levels and because
influential leaders within Islah have remained critical of its alliance with the opposition.”
Notably, Islah supported Saleh’s reelection for president in 1999 (Economist Intelligence Unit
2008b: 5; Durac 2019).
The Economist Intelligence Unit (2002c: 10) suggests that 2001 was a turning point, in which
Islah became more clearly an opposition party: “In early 2001, Islah underwent something
of a renaissance: growing fiercely critical of the GPC-backed constitutional reforms, and
opposing the government on other key issues.” The government began arresting Islah MPs
and activists, and toward the end of 2001, Islah’s leader “described the regime’s dealings
with his party as hostile” (ibid ). Islah also joined the Co-ordination Council of Opposition
Parties in 2001 (Sagar 2009: 651), and the following year it co-founded of the opposition
Joint Meeting Parties coalition (Hamzawy 2009: 8; Durac 2011: 343-344).53
51
Likewise, Durac (2019) writes that after the 1997 elections “al-Islah began to move closer to other
opposition parties and increasingly to espouse commitment to democratic ideals.”
52
Hiro (2003: 584) also notes that this move “further soured relations” between Islah and the GPC.
53
“The JMP developed a collective electoral platform, making the GPC their common enemy and
demanding the introduction of democratic safeguards and significant political reforms. They also coordinated
on candidates” (Hamzawy 2009: 8). (As an indication of the JMP’s popular support, its 2006 presidential
candidate won 22% of the vote [Economist Intelligence Unit 2008b: 6].) After co-founding the JMP, Islah’s
agenda was clearly reformist, as the party called for “constitutional amendments aimed at a fairer distribution
of power between the government and the opposition, reforms in electoral laws and laws pertaining to political
rights, improving parliament’s oversight of the government’s socioeconomic policies, and a reduction in
corruption” (Hamzawy 2009: 14).
199
However, after Saleh transferred power to his vice president, Hadi, in November 2011, the
Joint Meeting Parties endorsed the confirmation of Hadi’s presidency in a February 2012
referendum and entered into a coalition government with the GPC (Durac 2019), with Islah
taking seats in the cabinet (Reuters 2011c). Thus, Islah’s time in the opposition ended in
late 2011.
Sources indicate that the Yemeni Socialist Party has always had less support than Islah
since unification. As the former ruling party of South Yemen, its base was limited to the
less populous southern region, and its ability to organize was severely hampered after a
crackdown by the government in the wake of the 1994 civil war.54
Although many sources argue that Islah entered the opposition when it left the government
in 1997, its support for Saleh’s reelection in 1999 precludes coding it as an opposition party
during the late 1990s. Since Islah became clearly critical of the regime in 2001, and continued
to oppose the regime until the end of Saleh’s rule, it is coded as the main opposition party
for 2001-2011, its opposition role ending after it entered the Hadi government in late 2011
(Durac 2019).
No group is coded as a significant opposition group in Yemen during 1990 and 1991. The
following groups are coded as the significant opposition parties during 1992-2000.
At-Tajammu’a Al-Wahdawi (Yemeni Unionist Gathering) was founded in January 1990
54
For instance, the Economist Intelligence Unit (2002c: 10), reports: “The YSP ruled the one-party
PDRY [People’s Democratic Republic of Yemen, also known as South Yemen] before unification and won
almost all of the seats in the south in the 1993 election. However, following the 1994 civil war, during
which the YSP fought against the government, its leadership went into exile and the party lost much of
its support.” The Economist Intelligence Unit (2008b: 5) likewise notes that the YSP’s “poor showing in
the 2003 parliamentary election and subsequent local elections reflects the party’s waning political clout.”
The Economist Intelligence Unit (2001) also describes the YSP as “the main opposition party of the south”
(12-13, emphasis added) and concludes that the YSP “does not pose a sufficient threat to the ruling GPC’s
position” to concern the government (15).
200
to advocate for human rights and democracy (Middle East Economic Digest 1990a.:
Formation of the Yemeni Unionist Party) and was critical of the Saleh regime thereafter
(Schmitz and Burrowes 2018: 266). The party led an anti-government demonstration in
November 1991 (Middle East Economic Digest 1991b. Riots hit Sanaa as tensions mount
— shooting dead of traffic policeman by army colonel acts as catalyst), so it is coded as a
significant opposition group during 1992-1998.
The former ruling party of the People’s Democratic Republic of Yemen — which ceased to
exist in 1990 when the South and North were united as the Republic of Yemen (Schmitz and
Burrowes 2018: 486) — Hizb al-Ishtiraki al-Yemeni (Yemeni Socialist Party—YSP) won
over 10% of the vote in the April 1993 parliamentary elections Tachau 1994: 625). The party
also organized an anti-government demonstration in March 2000 (BBC Monitoring: Middle
East 2000b. Opposition rally urges release of Socialist Party leaders). Upon unification in
May 1990, the YSP formed a unity coalition with the GPC that lasted until the outbreak
of civil war between the North and the South in February 1994 (Schmitz and Burrowes
2018: xxxiii- xxxvii). Thus the YSP is not considered an opposition party during 1990-1993.
The civil war lasted until July 1994 (Schmitz and Burrowes 2018: xxxvii), so the YSP is
considered an armed group in 1994. The YSP was not included in subsequent governments
under Saleh and became an opposition party (Schmitz and Burrowes 2018: xxxvii, 527).
Thus, the YSP is coded as a significant opposition group during 1995-2000.
201
Appendix 4: Opposition Media
Narratives
ALGERIA
Islamic Salvation Front: 1992-1994
The FIS weekly newspapers El Mounqidh and El Fourqan were permanently banned in
February or March 1992 (sources vary on the timing). (Europa Publications 1989; Shehadi
1989; Reuters 1991b; Baggaley 1992a; Human Rights Watch 1993, 1994, 1995, 1997a, 1997b;
Europa Publications 1999; Willis 1999, 226; Rugh 2004; Willis 2012: 172).
Movement for Democracy in Algeria: 1992-1996
Throughout the 1992-1996 period, the MDA published the fortnightly newspapers Al-Badil
202
and Tribune d’Octobre, apparently without interference. (Europa Publications 1989; Reuters
1990d. “Ben Bella Supporters Stage Hunger Strike in Algeria.” ; Reuters 1990e. “Ben Bella’s
Party Launches Newspaper in Algeria.” ; Reuters 1990b. “Algerian Court Ends Seizure of
Ben Bella Party Magazine.” ; Human Rights Watch 1993, 1994, 1995, 1996, 1997a. World
Report 1997; Committee to Protect Journalists 1996, 1997; U.S. Department of State 1997;
U.S. Department of State 1998; Europa Publications 1999; Rugh 2004; Moussa 2015: 38
[FNs: 22, 23, 24, and 27])
Socialist Forces Front: 1992-2007
The FFS appears not to have had any media during 1992-2002. Beginning in September
2003, the party had an active website (www.ffs-dz.com)55 , which does not appear to have
been blocked during this period. (Europa Publications 1989; Human Rights Watch 1993,
1994, 1995, 1996, 1997a. World Report 1997, 1998, 1999, 2000, 2001, 2002, 2003; Committee
to Protect Journalists 1996, 1997, 1998, 1999, 2005, 2019; U.S. Department of State 19962008; Europa Publications 1999; Rugh 2004; Sagar 2009; Noman 2010; Europa Publications
2013; Europa World 2019l. Society and Media (Algeria); Front des Forces Socialistes 2019;
Reporters Sans Frontières 2019; Whois 2019j. “ffs.dz” )
55
The site is not currently active, but records from the Internet Archive show that the site had at least 24
posts on its home page during 2003, 2006, and 2007. Because of infrequent snapshots in some years, and in
some cases undated posts, there was no definitive evidence on frequency during 2004 or 2005, but they are
assumed to have been sufficiently frequent, since posting was frequent in 2003 and 2006.
203
Berber Cultural Movement: 1992-2007
The Berber Cultural Movement (MCB) does not appear to have had any media outlets
during this time frame. (Europa Publications 1989; Human Rights Watch 1993, 1994, 1995,
1996, 1997a. World Report 1997, 1998, 1999, 2000, 2001, 2002, 2003; Committee to Protect
Journalists 1996, 1997, 1998, 1999, 2005, 2019; U.S. Department of State 1996-2008; Europa
Publications 1999; Rugh 2004; Europa Publications 2013; BBC News 2017a; Europa World
2019l. Society and Media (Algeria); Kabyle Universel 2019; Reporters Sans Frontières 2019;
Whois 2019k. “kabyleuniversel.com” )
National Liberation Front: 1993-1996
The FLN had two newspapers, El Moudjahid and Ech Cha’ab, when it was the ruling party,
but under the new military regime, these papers were taken over by the state in May 1992
and do not appear to have been returned to FLN control until after it ceased to be an
oppositional group in 1997. The FLN founded the daily newspaper El Hiwar in April 1994.
The government suspended it in November 1994 and does not appear to have reauthorized it
while the FLN remained in the opposition. (Baggaley 1992a; Reuters 1992; Human Rights
Watch 1994, 1995, 1996, 1997s; Reuters 1994a; Reuters 1994c. Deux Autres Journaux
Suspendus en Algérie; Committee to Protect Journalists 1996, 1997; Reuters 1996a; U.S.
Department of State 1997; U.S. Department of State 1998; Immigration and Refugee Board
of Canada 2000)
204
Coordination of Aarchs, Dairas, and Communes: 2001-2010
The CADC appears to have had no media outlets during this period. (Human Rights Watch
2002, 2003, 2009, 2010, 2011; U.S. Department of State 2002, 2003, 2004, 2005, 2006, 2007,
2008, 2009, 2010; 2011b; Rugh 2004; Committee to Protect Journalists 2005, 2019; Europa
Publications 2013; BBC News 2017; Europa World 2019l. Society and Media (Algeria);
Reporters Sans Frontières 2019)
National Coordination for Change and Democracy: 2011-2017
The CNCD appears to have had no media during this period. (Freedom House 2012a.
Freedom of the Press, 2013a. Freedom of the Press, 2014a. Freedom of the Press, 2015a.
Freedom of the Press, 2016a. Freedom of the Press; Human Rights Watch 2012, 2013, 2014,
2015, 2016, 2017, 2018; U.S. Department of State 2012, 2013, 2014, 2015, 2016, 2017, 2018;
Committee to Protect Journalists 2019; Europa World 2019l. Society and Media (Algeria);
Reporters Sans Frontières 2019)
Rally for Culture and Democracy: 2011-2017
During this period, the RCD maintained a website: www.rcd-algerie.org.56 There is no
56
By the end of 2010, the site was redirecting visitors to a new site, http://rcd-algerie.net, but by the end
of 2011, the latter site was directing visitors back to the earlier site. Because www.rcd-algerie.org redirected
to http://rcd-algerie.net when the former was temporarily inactive, I treat the two sites as effectively a
single site (Rassemblement pour la Culture et la Démocratie 2019a; Rassemblement pour la Culture et la
205
evidence that the site was blocked in Algeria during this period. The site was active in each
year except 2015 and 2016. (Freedom House 2012a. Freedom of the Press, 2013a. Freedom
of the Press, 2014a. Freedom of the Press, 2015a. Freedom of the Press, 2016a. Freedom of
the Press; Human Rights Watch 2012, 2013, 2014, 2015, 2016, 2017, 2018; U.S. Department
of State 2012, 2013, 2014, 2015, 2016, 2017, 2018; Europa Publications 2013; Committee to
Protect Journalists 2019; Europa World 2019l. Society and Media (Algeria); Rassemblement
pour la Culture et la Démocratie 2019a; Rassemblement pour la Culture et la Démocratie
2019b; Reporters Sans Frontières 2019)
National Coordination for Liberties and Democratic Transition: 2015-2017
The CNLTD appears to have had no media during this period. (Freedom House 2016a.
Freedom of the Press; Human Rights Watch 2016, 2017, 2018; U.S. Department of State
2016, 2017, 2018; Committee to Protect Journalists 2019; Europa World 2019l. Society and
Media (Algeria); Reporters Sans Frontières 2019)
Démocratie 2019b). 24 or more posts were found through the Internet Archive in each year during 2011-2014,
as well as 2017. Specific evidence was found that the site posted infrequently in 2016. Due to a paucity of
Internet Archive captures during 2015, and the fact that the site did not post frequently in 2016, the site
was assumed not to have continued to post frequently in 2015.
206
BAHRAIN
The Accord National Islamic Society: 2002-2017
The Accord National Islamic Society (Al-Wefaq) maintained the websites AlWefaq.org and
AlWefaq.net since 2002. In 2013, the former began redirecting to the latter. Prior to that
the sites seemed to have hosted the same content or the .net site redirected to the .org
version. Thus, I treat them as different versions of a single site. At least one version
of the site was active throughout 2002-2016.57 They were blocked from September 2010
until January 2012. They were blocked again in 2016 and remained blocked at the end of
2017. The group’s weekly newspaper, Al-Wefaq, began publishing during or before 2008.58
It was banned in September 2010 and does not appear to have begun publishing again
subsequently.59 (Freedom House 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009a. “Freedom
of the Press 2009” , 2010, 2011a. “Freedom of the Press 2011” , 2011b. “Freedom on the Net
2011” , 2012a. “Freedom of the Press 2012” , 2012b. “Freedom on the Net 2012” , 2013a.
“Freedom of the Press 2013” , 2013b. “Freedom on the Net 2013” , 2014a. “Freedom of the
Press 2014” , 2014b. “Freedom on the Net 2014” , 2015a. “Freedom of the Press 2015” ,
2015b. “Freedom on the Net 2015” , 2016a. “Freedom of the Press 2016” , 2016b. “Freedom
on the Net 2016” , 2017a. “Freedom of the Press 2017” , 2017b. “Freedom on the Net 2017”
, 2018b; Committee to Protect Journalists 2003, 2004, 2005, 2007, 2008, 2009, 2010, 2011b.
“Attacks on the Press in 2010.” , 2012, 2013b. “Attacks on the Press in 2012.” , 2014a.
57
At least 24 posts were added to AlWefaq.org in every year from 2002 through 2004 and 2006 through
2012. There was insufficient evidence to determine the number of posts in 2005, but the site was assumed to
be active, since it was known to be active in 2004 and 2006. AlWefaq.net posted at least 24 times in every
year from 2013 through 2016 (before the site was blocked in the last case). After being blocked in 2016, but
AlWefaq.org and AlWefaq.net were inactive.
58
I did not find any reports about the paper prior to 2008.
59
The latest report on the paper that I found (Al Ayam 2012), was published in October 2012 and stated
that the paper was still banned.
207
“Attacks on the Press in 2013.” , 2019; U.S. Department of State 2003, 2004, 2005, 2006,
2007, 2008, 2009, 2010, 2011b. “Country Reports on Human Rights Practices for 2010.” ,
2012, 2013, 2014, 2015, 2016, 2017, 2018; OpenNet Initiative 2005, 2009a. “Bahrain” , 2019a.
“Internet Filtering in Bahrain in 2006-2007.” ; International Research & Exchanges Board
2006, 2008, 2009, 2010; Human Rights Watch 2007, 2009, 2010b. “World Report 2010.”
, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018; Reporters Sans Frontières 2008; Kenyon
2010; Toorani and Singh Grewal 2010; Energy Compass 2011; Al Ayam 2012; Reporters Sans
Frontières 2012; Europa Publications 2013; Al Ayam 2016; Gulf Daily News 2016; Europa
World 2019m. “Society and Media (Bahrain).” ; Accord National Islamic Society 2019a,
2019b; Reporters Sans Frontières 2019)
EGYPT
Muslim Brotherhood: 1992-2012, 2014-2017
Between 1992 and 2002, the Muslim Brotherhood had no frequently published media of its
own.60
The Brotherhood launched a website, IkhwanOnline.com, which was actively
posting in every year from 2003-2012 and 2014-2017.61 For parts of 2004 and 2005, the
government blocked the site in Egypt and also blocked the other versions of the site:
ikhwaonline.net and ikhwaonline.org.62
In December 2010, the government blocked
60
The Muslim Brotherhood did have a monthly newspaper during the 1990s, Liwa’ al Islam, the successor
to the monthly ad Dawah, which had been banned in 1981. Brotherhood members also often published their
views in Ash Shaab, the newspaper of the Labor Party until it was suspended in 2001 (Zahid 2012, 64).
61
Recall that Egypt 2013 is not in the dataset, because the country is coded as a democracy for the
majority of the year, although it reverted to authoritarianism in July of 2013.
62
It was not possible to determine whether the Muslim Brotherhood’s English-language website
208
IkhwanOnline.com,63 but did not block IkhwanOnline.net, which was actively posting that
year, particularly while IkhwanOnline.com was blocked.64 After the ousting of Mubarak in
February 2011, the Muslim Brotherhood launched many new media outlets under the
SCAF government, many of these outlets officially led by its Freedom and Justice Party.
These outlets were the website klmty.net, the newspaper Al Huriya wal Adala, and the
satellite TV channel Misr 25. The new military regime permanently shut down Al Huriya
wal Adala and Misr 25 in 2013, but in 2014-2016 IkhwanOnline.com and klmty.net were
both active and not blocked.
maintained www.fj-p.
com.
In 2012-2017, the Freedom and Justice Party actively
In addition, the Brotherhood also ran a website called
horriapost.net, which was active and not blocked in 2015 and 2016. However, all four of
these websites were blocked in 2017, although alternative versions of some sites remained
accessible
inside
Egypt
and
actively
posted,
including
IkhwanOnline.info
and
horiapost.de.65 (Europa Publications 1989; Human Rights Watch 1992, 1993, 1994, 1995,
1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2005, 2006, 2007, 2009, 2010, 2011, 2012,
2013, 2014, 2015, 2016, 2017, 2018; Abed-Kotob 1995; Abed-Kotob 1995; Committee to
Protect Journalists 1996, 1997, 1998, 1999, 2000, 2002, 2003, 2004, 2005, 2006, 2007, 2008,
2009, 2010, 2011b. “Attacks on the Press” , 2012, 2013b. “Attacks on the Press in 2012.” ,
2014a. Attacks on the Press; U.S. Department of State 1997, 1998, 1999, 2000, 2001, 2002,
2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011b. Country Reports on Human Rights
Practices for 2010, 2012, 2013, 2014, 2015, 2016, 2017, 2018; Europa Publications 1999;
Freedom House 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009a, 2009b. Freedom on the
Net 2009, 2010, 2011a, 2011b, 2012a, 2012b. Freedom on the Net 2012, 2013a, 2013b.
ikhwanweb.com was either actively posting or blocked while the IkhwanOnline sites were blocked. Since
all Muslim Brotherhood sites known to be active were blocked, each of these events is coded as a silencing
of the group.
63
The government also blocked ikhwanweb.com at this time.
64
According to the Internet Archive, for most of 2010, IkhwanOnline.net redirected to IkhwanOnline.com.
However in December it ceased redirecting while IkhwanOnline.com was blocked, and posted over 23 times
in this period alone.
65
The Freedom and Justice Party also had an English-language site fjponline.com, but it stopped posting
after 2012.
209
Freedom on the Net 2013, 2014a, 2014b. Freedom on the Net 2014, 2015a, 2015b. Freedom
on the Net 2015, 2016a, 2016b. Freedom on the Net 2016, 2017b. Freedom on the Net
2017, 2018b; International Research & Exchanges Board 2006, 2008, 2009, 2010; Rugh
2004; OpenNet Initiative 2009b. “Egypt” ; Sagar 2009; Reporters Sans Frontières 2011b.
“Egypt” ; Shenker 2011; Ahram Online 2013; Europa Publications 2013; Sakr 2013; Mada
Masr 2015; Ranko 2015; Aishima 2016; All Africa 2017; Mellor 2017; El-Taher et al 2018;
Kamal 2018; Europa World 2019n. “Society and Media (Egypt).” ; Freedom and Justice
Party 2019a; Freedom and Justice Party 2019 “Freedom and Justice Party” ; Klmty.net
2019; Muslim Brotherhood 2019a, 2019b, 2019c, 2019d; OpenNet Initiative 2019b.
“Internet Filtering in Egypt in 2006-2007.” ; Oxford Islamic Studies Online 2019)
April 6 Youth Movement: 2014-2017
The April 6 Movement had a website, www.6april.org, which was actively posting in 2014.66
In 2015 and 2016, the page was posting insufficiently frequently to qualify the group as
having media. In 2017, the page was updated frequently, and the government blocked access
to it within Egypt. The group also had an affiliated news website, elmasdr.com, which was
active 2013, 2014, and 2017, and was not blocked during this time. (Committee to Protect
Journalists 2014a Attacks on the Press; Committee to Protect Journalists 2014b. Egypt
jails two journalists, renews detention of another; Freedom House 2014a, 2014b. Freedom on
the Net 2014, 2015a, 2015b. Freedom on the Net 2015, 2016a, 2016b. Freedom on the Net
2016, 2017b. Freedom on the Net 2017, 2018b; Human Rights Watch 2014, 2015, 2016, 2017,
2018; U.S Department of State 2014, 2015, 2016, 2017, 2018; April 6 Youth Movement 2019;
66
It is important to note that April 6 began its organizing through a Facebook group, so this outlet has
been an important component of the group’s communication strategy. However, as discussed in the coding
rules, international social media platforms are not counted in determining whether governments allowed
opposition groups to have their own media.
210
Committee to Protect Journalists 2019; El Masdar 2019; Europa World 2019n. “Society and
Media (Egypt).” ; Reporters Sans Frontières 2019)
No to Military Trials for Civilians: 2014-2017
The group “No to Military Trials for Civilians” had a website during this period
(www.nomiltrials.org), but during 2014-2017, the group did not actively post to it.
(Committee to Protect Journalists 2014a. Attacks on the Press; U.S Department of State
2014, 2015, 2016, 2017, 2018; Freedom House 2014a, 2014b. Freedom on the Net 2014,
2015a, 2015b. Freedom on the Net 2015, 2016a, 2016b. Freedom on the Net 2016, 2017b.
Freedom on the Net 2017, 2018b; Human Rights Watch 2014, 2015, 2016, 2017, 2018;
Committee to Protect Journalists 2019; Europa World 2019n.
“Society and Media
(Egypt).” ; No to Military Trials for Civilians 2019; Reporters Sans Frontières 2019)
Salafist Front: 2014-2017
The Salafist Front ran the website www.GabhaSalafia.com, to which it actively posted in
2014-2017. The site was not blocked during this period. (Committee to Protect Journalists
2014a; U.S Department of State 2014, 2015, 2016, 2017, 2018; Freedom House 2014a, 2014b.
Freedom on the Net 2014, 2015a, 2015b. Freedom on the Net 2015, 2016a, 2016b. Freedom
on the Net 2016, 2017b. Freedom on the Net 2017, 2018b; Human Rights Watch 2014, 2015,
2016, 2017, 2018; Committee to Protect Journalists 2019; Europa World 2019n. “Society
and Media (Egypt).” ; Reporters Sans Frontières 2019; Salafist Front 2019a, 2019b)
211
JORDAN
Muslim Brotherhood: 1992-2017
In 1991, the Muslim Brotherhood launched a weekly newspaper, Ar-Ribat, an issue of
which was seized in the first year, but which was not interefered with subsequently. In
1993, the Brotherhood replaced Ar-Ribat with a new weekly, As-Sabeel, which became a
daily in 2009. As-Sabeel was published through 2017 without interference. (An associated
website, asSabeel.net, was never blocked in this period.) The Islamic Action Front, the
Brotherhood’s political party, maintained a website, www.Jabha.net, since as early as 2004,
which was blocked in January 2007 and hacked by regime supporters in March 2011.67 The
site was actively positing in both 2007 and 2011.
The Muslim Brotherhood website
www.IkhwanJo.com (which began in 2002) was also hacked in August 2011 and it was
actively posting in that year.68 The Muslim Brotherhood launched a satellite TV channel,
Al Yarmouk in 2011, but it was initially not permitted in Jordan, forcing the channel to
broadcast from Bahrain. In 2013, the Jordanian government allowed the channel to begin
broadcasting from Jordan, but later that year suspended the channel for five days. In 2015,
the government shut down the channel’s studios. Al Yarmouk began broadcasting with the
help of other Jordanian companies, but in 2016, the government pressured these companies
to stop working with Al Yarmouk, forcing the channel to resume broadcasting from outside
of Jordan. (A site associated with the channel, Yarmouk.tv, was not blocked during this
period.) (Europa Publications 1989; Department of State 1991, 1992, 1993, 1994, 1995,
67
I assume that the 2011 hacking was done with the support of the Jordanian regime, although it was not
possible to verify this assumption. The hackers posted messages on the site critical of the Islamic Action
Front and in support of King Abdullah II.
68
Although a version of the site was also blocked in 2006, there was insufficient evidence to determine if
the site was actively posting in that year.
212
1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010,
2011b. Country Reports on Human Rights Practices for 2010, 2012, 2013, 2014, 2015,
2016, 2017, 2018; Sabbagh 1991; Murphy 1992; Committee to Protect Journalists 1996,
1997, 1998, 1999, 2001, 2002, 2003, 2004, 2005, 2007, 2008, 2009, 2010, 2012, 2013, 2014,
2019; Al Abed 1997; Awadat 1997; Campagna 1998b; Europa Publications 1999; Freedom
House 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009a. Freedom of the Press 2009, 2010b.
Freedom of the Press 2010, 2011a. Freedom of the Press 2011, 2011b. Freedom on the Net
2011, 2012a. Freedom of the Press 2012, 2012b. Freedom on the Net 2012, 2013a. Freedom
of the Press 2013, 2013b. Freedom on the Net 2013, 2014a. Freedom of the Press 2014,
2014b. Freedom on the Net 2014, 2015a. Freedom of the Press 2015, 2015b. Freedom on
the Net 2015, 2016a. Freedom of the Press 2016, 2016b. Freedom on the Net 2016, 2017b.
Freedom on the Net 2017, 2018b. Freedom on the Net 2018; Index on Censorship 2002;
Ryan 2002; BBC Monitoring: Media 2006; Human Rights Watch 2006, 2007, 2009, 2010b.
World Report 2010; 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018; International Research
& Exchanges Board 2006, 2008, 2009, 2010; Schwedler 2006; BBC Monitoring: Media.
2007; BBC Monitoring: Middle East. 2007; Noman 2007; BBC Monitoring: Media 2009d;
OpenNet Initiative 2009c; Agence France Presse 2011b; Ammon News 2011; Jordan Times
2011, 2014, 2015a, 2015b, 2016, 2017; BBC Monitoring: Middle East 2013a; El-Shamayleh
2013; Europa Publications 2013; BBC Monitoring: Newsfile 2015; Middle East Monitor
2016; Europa 2019s; Europa 2019f; Islamic Action Front 2019; Muslim Brotherhood 2019e;
OpenNet Initiative 2019c; Reporters Sans Frontières 2019)
213
KUWAIT
Islamic Constitutional Movement: 1997-2003
The Social Reform Society, the precursor to the Islamic Constitutional Movement (Hadas)
founded the weekly Al-Mujtama in 1970, and published it without interruption throughout
1997-2003.69 (Ghabra 1991; Reuters 1994e. “Kuwaiti magazine urges Moslems to boycott
Russia.” ; Ersan 1996; Ghabra 1997; U.S. Department of State 1997, 1998, 1999, 2000, 2001,
2002, 2003, 2004; Fouad 1999; Fouad 2001; Human Rights Watch 2001; Freedom House 2002,
2003, 2004; Selvik 2011; Brown 2012; Brown 2019; Committee to Protect Journalists 2019;
Reporters Sans Frontières 2019; Social Reform Society 2019; Whois 2019i. “eslah.com” )
Enough: 2011-2017
The “Enough” (Kafi) movement had a blog (http://kafiq8.blogspot.com), but never posted
to it at a rate of 24 or more posts per year. The group had no other media in this period.
(Al Jazeera 2011b; Committee to Protect Journalists 2012; Freedom House 2012a. Freedom
of the Press, 2013a. Freedom of the Press, 2014a. Freedom of the Press, 2015a. Freedom of
the Press, 2016a. Freedom of the Press; Human Rights Watch 2012, 2013, 2014, 2015, 2016,
2017, 2018; U.S. Department of State 2012, 2013, 2014, 2015, 2016, 2017, 2018; Committee
to Protect Journalists 2013b. “Attacks on the Press in 2012.” ; Nordenson 2017; Committee
to Protect Journalists 2019; Enough 2019; Reporters Sans Frontières 2019)
69
The Social Reform Society also founded a website, www.eslah.com, in 2002, but there was insufficient
evidence to determine its posting frequency in 2002 and 2003.
214
Path: 2011-2017
The “Path” (Nahj) movement did not have any media during this period. (Committee to
Protect Journalists 2012; Freedom House 2012a. Freedom of the Press, 2013a. Freedom of
the Press, 2014a. Freedom of the Press, 2015a. Freedom of the Press, 2016a. Freedom of
the Press; Human Rights Watch 2012, 2013, 2014, 2015, 2016, 2017, 2018; U.S. Department
of State 2012, 2013, 2014, 2015, 2016, 2017, 2018; Committee to Protect Journalists 2013b.
“Attacks on the Press in 2012.” ; Committee to Protect Journalists 2019; Reporters Sans
Frontières 2019)
A Nation’s Dignity: 2013-2017
“A Nation’s Dignity” (Karamat Watan) had a blog (https://karametwatan.wordpress.com)
but did not post to it frequently. The group had no other media. (Freedom House 2014a.
Freedom of the Press, 2015a. Freedom of the Press, 2016a. Freedom of the Press Human
Rights Watch 2014, 2015, 2016, 2017, 2018; U.S. Department of State 2014, 2015, 2016,
2017, 2018; Nordenson 2017; Committee to Protect Journalists 2019; Nation’s Dignity 2019;
Reporters Sans Frontières 2019)
Popular Action Bloc: 2013-2017
The Popular Action Bloc had no media in this period. (Freedom House 2014a. Freedom
215
of the Press, 2015a. Freedom of the Press, 2016a. Freedom of the Press Human Rights
Watch 2014, 2015, 2016, 2017, 2018; U.S. Department of State 2014, 2015, 2016, 2017, 2018;
Committee to Protect Journalists 2019; Europa 2019g. “Political Organizations (Kuwait)”;
Popular Action Bloc 2019; Reporters Sans Frontières 2019)
MAURITANIA
Union of Democratic Forces/Rally of Democratic Forces: 1992-2004, 2009-2017
Beginning in 2003, the UFD/RFD had a website (www.rfd-mauritanie.org), but did not
post to it frequently during any of the years it was coded as a significant opposition
group.70 The group does not appear to have had any other media outlets during this
period. (Human Rights Watch 1993; Europa Publications 1995; U.S. Department of State
1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2009, 2010b. “World Report
2010” , 2011, 2012, 2013, 2014, 2015, 2016, 2017; Committee to Protect Journalists 1997,
1999, 2000, 2001, 2002, 2003, 2019; Freedom House 2002, 2003, 2004, 2005, 2006, 2007,
2008, 2009a, 2010, 2011a, 2012a. Freedom of the Press, 2013a. Freedom of the Press,
2014a. Freedom of the Press, 2015a. Freedom of the Press, 2016a. Freedom of the Press;
Europa Publications 2005; International Research & Exchanges Board 2010, 2011b.
Development of Sustainable Independent Media in Africa; Europa World 2019o. Society
70
Due to a paucity of successful captures by the Internet Archive for 2003 and 2004, these years did not
qualify the RCD as having media. There was also a lack of evidence for 2009. Although the site had at
least 24 posts in 2008 (when Mauritania was a democracy), in 2010, there were only five posts over the 249
days for which there was evidence from the Internet Archive, a rate of approximately one every seven weeks.
Thus 2009 was not counted. In 2011 and 2012 there were no posts. There was a lack of sufficient captures
in 2013-2017; therefore, the latter years were not counted.
216
and Media (Mauritania); Rassemblement des Forces Démocratiques 2019; Reporters Sans
Frontières 2019)
Opposition Parties Front: 1998-2005
The FPO does not appear to have had any media outlets during this period. (Committee to
Protect Journalists 1999, 2000, 2001, 2002, 2003; U.S. Department of State 1999, 2000, 2001,
2002, 2003, 2004, 2005; Freedom House 2002, 2003, 2004, 2005, 2006; Europa Publications
2005; Reporters Sans Frontières 2019)
National Front for the Defense of Democracy: 2009-2016
The FNDD does not appear to have had any media outlets during this period. (Freedom
House 2010, 2011a, 2012a, 2013a, 2014a, 2015a, 2016a; International Research & Exchanges
Board 2010, 2011b.
Development of Sustainable Independent Media in Africa; U.S.
Department of State 2010, 2011b. Country Reports on Human Rights Practices for 2010,
2012, 2013, 2014, 2015, 2016, 2017; Committee to Protect Journalists 2019; Europa World
2019o. Society and Media (Mauritania); Reporters Sans Frontières 2019)
217
Coordination of Democratic Opposition: 2010-2017
The COD does not appear to have had any media outlets during this period. (Freedom House
2011a, 2012a, 2013a, 2014a, 2015a, 2016a; International Research & Exchanges Board 2011b.
Development of Sustainable Independent Media in Africa; U.S. Department of State 2011b.
Country Reports on Human Rights Practices for 2010, 2012, 2013, 2014, 2015, 2016, 2017,
2018; Committee to Protect Journalists 2019; Europa World 2019o. Society and Media
(Mauritania); Reporters Sans Frontières 2019)
National Rally for Reform and Development: 2013-2017
The RNRD (Tewassoul)-affiliated satellite TV channel Al Mourabiton was launched in 2012,
initially broadcasting illegally from outside the country. The government gave Al Mourabiton
a license to broadcast from within Mauritania in May 2013. Although some programs were
suspended, the channel was allowed to broadcast until the government shut down its offices
and prevented broadcasting in October 2017.71 By 2013, the party also had a website
(http://tewassoul.org), to which it posted actively in 2013 and 2014, but abandoned in
2015.72 That year, the party launched a new website (http://tewassoul.mr), on which it
posted frequently during 2015-2017.73 Neither site appears to have been blocked while it
71
The channel had an associated website (http://elmourabiton.tv) since 2012, but evidence from the
website and the Internet Archive was insufficient to determine how often the site was updated in any year
in 2012-2017 (El Mourabiton 2019a; El Mourabiton 2019b; Whois. 2019h).
72
The site had more than 23 posts in 2013 and 2014. In 2015, evidence from the Internet Archive is only
available from January 1 through April 29, but during that time, there were only 2 new posts. At the only
other snapshot that year (July 18), the site displayed a “Service Temporarily Unavailable” error. In 2016
and 2017, the site had been taken over by another organization, publishing in a foreign language.
73
Evidence from the Internet Archive indicates that there were more than 23 posts in each year in this
period.
218
was active. (L’Expression 2012; Gueye 2013; All Africa 2014; Freedom House 2014a, 2015a,
2016a; BBC Monitoring: Middle East 2015; Agence Ecofin 2016; APANews 2017; Morgan
2017; BBC Monitoring: Middle East 2018; Committee to Protect Journalists 2019; Europa
World 2019o. Society and Media (Mauritania); Rassemblement National pour la Réforme et
le Developpement 2019a, 2019b; Reporters Sans Frontières 2019; U.S. Central Intelligence
Agency. 2019c. “Mauritania.” )
National Forum for Democracy and Unity: 2014-2017
The FNDU does not appear to have had any media outlets during this period. (Freedom
House 2015a, 2016a; U.S. Department of State 2015, 2016, 2017, 2018; Committee to Protect
Journalists 2019; Europa World 2019o. Society and Media (Mauritania); Reporters Sans
Frontières 2019)
MOROCCO
Independence Party: 1992-1997
The Parti de l’Istiqlal founded a daily newspaper, Al-Alam, in the 1940s.
The party
founded another daily newspaper, l’Opinion, in the 1960s. Neither paper was suspended
during the period 1992-1997. (Damis 1972; Europa Publications 1989; Drost 1991; U.S.
219
Department of State 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998; Human Rights Watch
1992; Reuters 1995b. “Press Digest — Morocco.” ; Human Rights Watch 1996; Reuters
1997c. “Press Digest — Morocco.” ; Europa Publications 1999; Rugh 2004; Boum and
Park 2016; Independence Party 2019)
Socialist Union of Popular Forces: 1992-1997
The Socialist Union of Popular Forces (USFP) founded a daily called Libération in 1964
and another daily called Al-Ittihad Al-Ichtiraki in 1983. Neither publication was suspended
during 1992-1997. (Europa Publications 1989; Drost 1991; U.S. Department of State 1991,
1992, 1993, 1994, 1995, 1996, 1997, 1998; Human Rights Watch 1992; Reuters 1995b “Press
Digest — Morocco.” ; Human Rights Watch 1996; Reuters 1997c. “Press Digest — Morocco.”
; Europa Publications 1999; Rugh 2004; Boum and Park 2016; Europa World 2019p. “Society
and Media (Morocco).” ; Socialist Union of Popular Forces 2019a, 2019b)
Justice and Charity: 1990-2017
Justice and Charity (al-Adl wal-Ihsan, commonly known as Ihsan for short) had no media
during 1990-1998. In 1983, it had attempted to launch a twice-monthly magazine, As-Subh,
but it was banned after its second issue. Thus, the group is coded as having a suppressed
media outlet in 1990. In 1999, the group received authorization to print the weekly Risalat
al-Fotowa (Message of the Youth), the official publication of Ihsan’s youth wing, but the
government banned the newspaper in 2000 and it remained banned through 2017. In 2000,
220
the group tried to print a weekly entitled al-Adl wal-Ihsan, but the government immediately
prevented its distribution, and the publication remained banned through 2017. Ihsan’s
leader, Sheikh Abdesslam Yassine, launched the website Yassine.net in 2000, although it
remained under construction until 2004. The site was active during 2004-2008 and 20112017.74 Ihsan launched a website with multiple versions in 2000: AlJamaa.org, AlJamaa.net,
and AlJamaa.com (Whois 2019b., 2019c., 2019d., 2019e.).75 Typically one site was active at
any given time, with the other versions redirecting to it, so I treat AlJamaa as a single site
with multiple versions. There was insufficient evidence for posting frequency on any version
of the site in 2000, but at least one version of the site posted actively in each year during
2001-2017.76 Ihsan launched a site for its female followers, Mouminate.net, in 2005 and
it posted actively throughout 2006-2016.77 Nadia Yassine, daughter of Abdesslam Yassine
and a spokesperson for the group, launched a website, nadiayassine.net in 2005. In most
years there was either insufficient evidence on posting frequency or the evidence indicated
that the site posted less than 24 times in a year, except 2011, when the site was active.78
Ihsan launched an online TV channel, Chahed.tv, in 2013, and the site was active throughout
2013-2017.79 The Moroccan government blocked access in the country to all of Ihsan’s active
74
In 2009, it was only possibly to determine the number of posts during a 123-day period, during which
time 8 posts were added to the site. The annualized rate is slightly less than 24, so the year is treated
as inactive. In 2010, there was insufficient evidence to determine frequency, so that year is also treated as
inactive, since it was inactive in 2009.
75
AlJamaa.info was added in 2004.
76
AlJamma.org posted at least 24 posts in 2001, AlJamaa.com posted sufficiently frequently in 2002-2005,
and AlJamaa.net posted frequently throughout 2006-2017.
77
There was insufficient evidence to determine posting frequency in 2005. In 2008, there were 20 posts
during the 120-day period for which there was evidence, an annualized rate of more than 23. There was
insufficient evidence to directly assess positing frequency in 2010, but the site was assumed to be active in
this year, since it was active in both 2009 and 2011. Similarly, despite sufficient direct evidence to determine
2014 positing frequency, this year was assumed to be active because the site was active in 2013 and 2015.
There was insufficient evidence to determine posting frequency in 2017, but that year was assumed inactive,
because starting in July 2017 and persisting through 2018, all captures of the site by the Internet Archive
resulted in 403 errors.
78
In 2005-2007, 2009-2010, and 2012-2017, there was insufficient evidence on posting frequency. In 2008,
there were 5 new posts in the 165-day period for which there was evidence, which is an annualized rate of
only 11 posts. During March 2016, all Internet Archive captures of the site encountered 404 errors, and
there were no further captures of the site. The site is currently inactive as of October 2019. It appears that
the site was no longer in operation from March 2016 on.
79
The group also launched the website fotowa.com in 2000, but it never posted sufficiently frequently to
qualify.
221
websites from April 2001 until 2005, and for parts of 2006 and 2009. (Europa Publications
1989; Drost 1991; U.S. Department of State 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011b. “Country
Reports on Human Rights Practices for 2010.” , 2012, 2013, 2014, 2015, 2016, 2017, 2018;
Human Rights Watch 1992, 1996, 2002, 2006, 2007, 2008, 2009, 2010b. “World Report 2010”
, 2011, 2013, 2014, 2015, 2016, 2017, 2018; Kokan 1994; Bouzerda 1995; BBC Monitoring:
Middle East. 1998a. “Islamist movement denounces ’repression’.” ; Committee to Protect
Journalists 1999, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2014a. Attacks on the
Press in 2013; Europa Publications 1999; Bouzerda, Ali. 2000a. “Moroccan Islamist says to
end 10-year-house arrest.” ; Bouzerda, Ali. 2000b. “Morocco’s Islamist chief free, group still
shackled.” ; Khalaf 2000; Reuters 2000; BBC Monitoring: Media 2001; BBC Monitoring:
Middle East 2001; Reuters. 2001c. “Morocco bans access to radical Islamists’ websites.”
; Freedom House 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009a. Freedom of the Press
2009; 2010b. Freedom of the Press 2010; 2011a. Freedom of the Press 2011; 2012a. Freedom
of the Press 2012; 2013a. Freedom of the Press; 2013b. Freedom on the Net 2013; 2014a.
Freedom of the Press 2014a.; 2014b. Freedom on the Net 2014; 2015a. Freedom of the Press
2015; 2015b. Freedom on the Net 2015; 2016a. Freedom of the Press 2016; 2016b. Freedom
on the Net 2016; 2017. Freedom of the Press 2017; 2017b. Freedom on the Net; 2018b.
Freedom on the Net 2018b; International Research & Exchanges Board 2006; Thorne 2006;
BBC Monitoring: Media 2009a “Morocco: Websites of Islamic movement inaccessible since
17 January.” ; Rubin 2010; BBC Monitoring: Middle East. 2013c. “Morocco’s Islamist
party reportedly launches TV channel.” ; Europa Publications 2013; Motaouakal 2014; AlChahed 2019; A. Yassine 2019; Europa World 2019h. “Political Organizations (Morocco).”
; Europa World 2019p. “Society and Media (Morocco).” ; Justice and Spirituality 2019a,
2019b, 2019c, 2019d; N. Yassine 2019; Partrick, Neill 2019a. “History (Morocco).” ; Sisters
of the Hereafter 2019; Whois 2019b “aljamma.com” ; Whois 2019c “aljamma.info” ; Whois
2019e “aljamma.org” ; Whois 2019d “aljamma.net” ; Whois 2019o. “yassine.net” ; Whois
222
2019m. “nadiayassine.net” ; Whois 2019l. “mouminate.net” ; Whois 2019g. “chahed.tv” )
Democratic Bloc: 1993-1997
The Democratic Bloc did not have any media of its own. It is nonetheless coded as having
permitted media during this period its leading members, Istiqlal and the USFP, were
significant opposition parties that had legally permitted media. (U.S. Department of State
1993, 1994, 1995, 1996, 1997, 1998; Human Rights Watch 1996; Europa Publications 1999;
Rugh 2004; Boum and Park 2016)
SUDAN
Umma National Party: 2000-2017
The Umma National Party launched a website, umma.org, in 1998. However, it was only
intermittently active, posting 24 or more times per year only in 2010 and 2017.
The
website was never blocked.80 The party started legally publishing a weekly newspaper,
Sawt Al Umma, in 2007 and it published without interference until closed due to financial
difficulties in 2012.81 (Committee to Protect Journalists 2001, 2002, 2003, 2004, 2005,
80
Previously, the Umma National Party (then the Umma Party) had a daily newspaper, Al Umma, but it
was not published during 2000-2017.
81
Sawt al-Umma had previously been banned in 1989, but this is outside the seven-year window, so the
group is not coded as having suppressed media in any year.
223
2008, 2009, 2010, 2011b. Attacks on the Press in 2010, 2012, 2013b. Attacks on the Press
in 2012, 2014a. Attacks on the Press in 2013; Human Rights Watch 2001, 2002, 2003, 2006,
2007, 2008, 2009, 2010b. “World Report 2010” , 2011, 2012, 2013, 2014, 2015, 2016, 2017,
2018; U.S. Department of State 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010,
2011b. “Country Reports on Human Rights Practices for 2010.” ; 2012, 2013, 2014, 2015,
2016, 2017, 2018; Freedom House 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009a. Freedom
of the Press 2009, 2010b. Freedom of the Press 2010, 2011a. Freedom of the Press 2011,
2012a Freedom of the Press, 2013a. Freedom of the Press, 2013b. Freedom on the Net
2013, 2014 Freedom of the Press 2014a., 2014b. Freedom on the Net 2014, 2015a. Freedom
of the Press 2015, 2015b. Freedom on the Net 2015, 2016a. Freedom of the Press 2016,
2016b. Freedom on the Net 2016, 2017b. Freedom on the Net 2017, 2018b. Freedom on the
Net 2018; Europa Publications 2005; OpenNet Initiative 2009d. “Sudan.” ; Deckert 2012;
International Research & Exchanges Board 2013; Kramer, Lobban, and Fluehr-Lobban
2013; Committee to Protect Journalists 2019; Cover SD 2019; Elawad 2019a; Elawad
2019b; Europa World 2019 “Media and Society (Sudan).” ; Europa World 2019i. “Political
Organizations (Sudan).” ; Gallab 2019; OpenNet Initiative 2019d. “Internet Filtering in
Sudan in 2006-2007.” ; Reporters Sans Frontières 2019; Shaib 2019; Umma National Party
2019; Whois 2019n. “umma.org” )
224
TUNISIA
Renaissance Party: 1992-1999
The Renaissance Party (Nahda) did not publish any media during 1992-1999.82 (Donnadieu
1992; U.S. Department of State 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000; Kr´’amer
1994)
YEMEN
Yemeni Unionist Gathering: 1992-1998
Shortly after it was founded in 1990, the Yemeni Unionist Gathering began publishing a
weekly newspaper, At-Tajammu. The newspaper was temporarily suspended in 1994 and
1996, but otherwise published throughout 1992-1998. (Reuters 1991a. “500 Somali Refugees
Try Their Luck in Aden” ; Index on Censorship 1992; Human Rights Watch 1993; Youssef
1993; BBC Monitoring: Middle East 1994a. “Coeducation in South Yemen Abolished; Press
Censorship Reportedly Introduced.” ; “Tachau 1994; BBC Monitoring: Middle East 1996c.
“Newspapers Report Increased Opposition Activity in the South.” ; Committee to Protect
Journalists 1996; Index on Censorship 1996; Committee to Protect Journalists 1997; Human
82
Nahda began publishing a weekly, El-Fajr, in 1990. In 1991, the government permanently closed El-Fajr.
225
Rights 1997; U.S. Department of State 1997; Committee to Protect Journalists 1999; Europa
Publications 1999; Rugh 2004)
Yemeni Socialist Party: 1995-2000
The Yemeni Socialist Party launched Ath-Thawri as its official weekly journal when it was
the ruling party of South Yemen, prior to the 1990 unification. Although the paper was
suspended temporarily in 1994 after the outbreak of the civil war that pitted the YSP against
the ruling GPC, it was operational again by 1995. Though its journalists were subjected
to regular harassment, Ath-Thawri was not suspended during the 1995-2000 period. (Cigar
1990; BBC Monitoring: Middle East 1995; Reuters 1995c. “Yemen opposition party says
members being arrested.” ; Committee to Protect Journalists 1996; Reuters 1996b. “Yemeni
tribesmen release release kidnapped oilmen — paper.” ; Committee to Protect Journalists
1997; Human Rights Watch 1997; U.S. Department of State 1997, 1998, 1999, 2000, 2001;
Committee to Protect Journalists 1999; Europa Publications 1999; Human Rights Watch
2000; Rugh 2004; Schmitz and Burrowes 2018)
Yemeni Congregation for Reform: 2001-2011
The Yemeni Congregation for Reform (Islah) founded a weekly newspaper, As-Sahwa, in
1985. It published continuously during 2001-2010, so is coded as permitted in those years.
The publication was actively printing and allowed during the first few months of 2011, but
on May 25, 2011 armed supporters of Saleh attacked the newspaper’s headquarters and
226
the publication of the print edition was halted, so this event is coded as a banning. The
party launched a website affiliated with the paper, www.alsahwa-yemen.net, in 2001, and the
site published actively throughout 2001-2010, without interference. On May 25, 2011, armed
supporters of Saleh attacked the office of the website, delaying posts to the site for a few days.
Later that year, the government blocked access to the site within Yemen.83 Islah launched the
website www.al-islah.net in 2003 and it posted actively in 2004-2011 without interference.84
Islah launched a satellite television channel, Suhail, which began broadcasting regularly in
January 2010. I code the channel as not permitted by the Saleh government in 2010, as it
was broadcast from outside Yemen, but by 2011, the channel had an office in Sana’a. Armed
supporters of Saleh attacked the Sana’a office on May 25, 2011 and its transmission was halted
until the next day. A website affiliated with Suhail TV, www.suhail.net, was launched in 2009
and actively posted in 2010.85 (Europa Publications 1999; Human Rights Watch 2002, 2009,
2010b. “World Report 2010.” , 2011, 2012, 2016, 2017, 2018; Freedom House 2002, 2003,
2004, 2005, 2006, 2007, 2008, 2009a. “Freedom of the Press” , 2010, 2011a. “Freedom of the
Press” , 2012a. “Freedom of the Press” , 2016a. “Freedom of the Press” ; U.S. Department
of State 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011b. “Country Reports
on Human Rights Practices for 2010.” , 2012, 2016, 2017, 2018; BBC Monitoring: Middle
East 2003; Seddon 2004; Committee to Protect Journalists 2005, 2008, 2009, 2010 “Attacks
on the Press in 2009.” , 2011b. “Attacks on the Press in 2010.” , 2012; Browers 2007;
BBC Monitoring: Media 2009b “Yemen: Suhayl satellite television channel launches news
website.” ; BBC Monitoring: Middle East 2009b. “Paper reports on trial broadcasting of new
Yemeni satellite channel.” ; BBC Monitoring: Middle East 2009a. “Kuwait-based Yemeni
TV shut down on official orders.” ; Salmoni, Loidolt, and Wells 2010; BBC Monitoring:
Middle East 2011; Committee to Protect Journalists 2011a. “Yemen shells TV station, news
83
Prior to the May 25, 2011 attack, the website was publishing at an annualized rate of at least 24 posts
per year, so it is regarded as active at the time of the effective banning of the site.
84
There was insufficient evidence to determine posting frequency in 2003. During most of 2007, the site
redirected to www.islah-forums.net.
85
There was insufficient evidence to determine the frequency of posting to the site in 2009 or 2011.
227
agency, online newspaper” ; Reporters Sans Frontières 2011a; Reuters 2011a ; Reporters Sans
Frontières 2011c; Yemen Post 2011; BBC Monitoring: Media 2012; CDAC Network 2012;
Dubai Press Club 2012; Europa Publications 2013; ForeignAffairs.co.nz 2014; Yadav 2014;
Committee to Protect Journalists 2015; Bakhsh 2017; Fanack 2017; Schmitz and Burrowes
2018; Al-Islah 2019; Al-Sahwah Net 2019; Anonymous Yemeni journalist 2019; OpenNet
Initiative. 2019e. “Yemen.” Suhail Net 2019; Whois 2019f. “alsahwa-yemen.net” ; Whois
2019a. “al-islah.net” )
228
Appendix 5: Regime Leaders
The following is a list of all the de facto regime leaders who ruled during the regime-years
in the dataset. This information was used to construct the variables leader year office and
regime first leader. Goemans, Gleditsch, and Chiozza (2009)86 and Geddes, Wright, and
Frantz (2014) also code de facto regime leaders. Where my coding of leaders differs from
either of theirs, I explain my reasoning in the footnotes and provide sources.
86
The
Archigos
dataset
of
government
leaders
https://www.rochester.edu/college/faculty/hgoemans/data.htm
229
can
be
found
at:
ALGERIA
Nezzar
Start: January 11, 199287
End: January 31, 1994
Years in dataset: 1992-1993
First leader of regime: Yes
Zeroual
Start: January 31, 1994
87
My coding for this period departs from Goemans, Gleditsch, and Chiozza (2009), who code Mohamed
Boudiaf and Ali Kafi as Algeria’s leaders from January 14, 1992 to June 29, 1992 and July 2, 1992 to January
31, 1994, respectively. My coding is consistent with the Geddes, Wright, and Frantz (2014). Naylor (2015,
407) observes: “Nezzar played the most prominent role in plotting and executing the deposal of President
Benjedid by the coup of January 1992. He subsequently became recognized as the most powerful figure
within the Haut Conseil d’état (HCE), the collegial executive that took over the government.” This view
of Nezzar as first among equals in the junta was observed in numerous contemporaneous accounts. An
article in the Middle East Economic Digest (1992f. “Political Crisis Reaches Climax with Resignation of
President Chadli” ) observed two weeks after the coup that “The dominant figures in the military leadership
are Defence Minister General Khaled Nezzar and chief-of-staff Abdelmalek Guenaizia” and a Reuters (1993)
article argued in February of the following year that “Nezzar is seen as the strongman in the army-backed
presidency.” This view did not alter after Nezzar resigned as defense minister in July 1993. Ghanmi (1993)
reported the following month that Nezzar was still “widely regarded as the strongest figure” on the ruling
council and the Middle East Economic Digest (1993b. “Nezzar to Withdraw from Political Life When
Mandate of Higher Council of State Expires” ) reported in October of 1993 that “Nezzar has been regarded
as the directing force within the HCS [High Council of State].” See also: Baggaley (1992f. “Five-Man Body
Takes Charge of Algeria” ) and Middle East Economic Digest (1993a. “Military Shake-Up Strengthens Hard
Line.” ).
230
End: April 27, 1999
Years in dataset: 1994-1998
First leader of regime: No
Bouteflika
Start: April 27, 1999
End: Remained in power December 31, 201788
Years in dataset: 1999-2017
First leader of regime: No
BAHRAIN
Hamad Isa Ibn Al-Khalifah
Start: March 6, 1999
88
Economist Intelligence Unit 2018a.
231
End: Remained in power December 31, 201789
Years in dataset: 2002-2017
First leader of regime: No
EGYPT
Mubarak
Start: October 14, 1981
End: February 11, 2011
Years in dataset: 1992-2010
First leader of regime: No
Tantawi
Start: February 11, 2011
89
Economist Intelligence Unit 2018b.
232
End: August 12, 201290
Years in dataset: 2011-2012
First leader of regime: Yes
As-Sisi
Start: July 3, 201391
End: Remained in power December 31, 201792
Years in dataset: 2014-2017
First leader of regime: Yes
90
President Mohammed Morsi forced Tantawi and several other top military leaders in the Supreme Council
of the Armed Forces (SCAF) to resign and revoked the SCAF’s June decree expanding its powers. (After
dissolving the parliament in June 2012, the SCAF had placed restrictions on the power of the president
and assumed authority over law-making, the military budget, and the drafting of a new constitution.) The
dismissals were not challenged by the SCAF and are seen as the moment when the balance of power shifted
from the SCAF to Morsi’s government. (Freedom House 2013 “Freedom in the World” ; Lansford 2017;
Europa World 2019a. “Contemporary Political History (Egypt).” )
91
The military ousted Morsi in a coup and Egypt was thereafter ruled by General Abd al Fatah As-Sisi, the
Commander in Chief of the armed forces and Minister of Defense and Military Production. As-initially was
a de facto ruler, having appointed a civilian president. After violently suppressing the Muslim Brotherhood,
As-Sisi was elected president in 2014 (Lansford 2017; Europa World 2019a. “Contemporary Political History
(Egypt).” ).
92
Europa World 2019a. “Contemporary Political History (Egypt)” .
233
JORDAN
Hussein Ibn Talal Al-Hashemi
Start: August 11, 1952
End: February 7, 1999
Years in dataset: 1992-1998
First leader of regime: No
Abdullah Ibn Hussein Al-Hashemi
Start: February 7, 1999
End: Remained in power December 31, 201793
Years in dataset: 1999-2017
First leader of regime: No
93
Economist Intelligence Unit 2018c.
234
KUWAIT
Jabir As-Sabah
Start: January 1, 197894
End: September 21, 200195
Years in dataset: 1997-2003
First leader of regime: No
94
From August 8, 1990 to April 20, 1991, Goemans, Gleditsch, and Chiozza (2009) reasonably code Saddam
Hussein as the de facto leader of Kuwait, due to the Iraqi occupation of the country. Nonetheless, I code
the leader year in office variable for Jabir As-Sabah with reference to his ascension to the throne in 1978,
because this variable is meant to capture how long a ruler has had to consolidate his support base. I assume
that the king did not have to completely start over in this respect after the Iraqi withdrawal in 1991.
95
The end date of Jabir As-Sabah’s rule/the start date of Sabah Al-Ahmad’s rule is different from Goemans,
Gleditsch, and Chiozza (2009) and Geddes, Wright, and Frantz (2014), who code the former as ruling until
his death in 2006 and the latter as beginning his rule soon after when he was formally crowned as emir.
Numerous sources report that Sabah Al-Ahmad became Kuwait’s “de facto” ruler when his predecessor had
a stroke on September 21, 2001 (Agence France Presse 2001b; Economist Intelligence Unit 2003; Agence
France Presse 2006; Fattah 2006; Global Security 2019). Goemans, Gleditsch, and Chiozza (2009) also code
Saad Abdullah As-Salem As-Sabah as ruling from January 15, 2006, when he was formally crowned, until
January 24, 2006, when the ruling family forced Saad to abdicate, transferring power to Sabah Al-Ahmad
(Economist Intelligence Unit 2003; Fattah 2006). Saad was in extremely poor health during the brief period
he held the title of emir, and there is no indication that he exercised any meaningful power during this
time. (During the 2001-2006 period, Saad had been the crown prince, but by the time of the emir’s stroke in
2001, “due to [Saad’s] health problems,” Saad had “delegated extensive authority to Sheikh Sabah al-Ahmad
al-Sabah” [Agence France Presse 2001b].) In any case, if this period were treated as a genuine interregnum,
it would not affect the coding of Sabah Al-Ahmad as the leader of Kuwait for most of 2006. Nor would it
be sensible to code the leader year in office variable as resetting to 1 when Sabah Al-Ahmad was crowned,
for reasons similar to the coding of his predecessor after 1990. (See previous footnote.) Underscoring that
Sabah Al-Ahmad had built up power since 2001 that he continue to wield on formally becoming emir, the
Economist Intelligence Unit (2003: 3) said of Sabah Al-Ahmad a couple months after the succession that
“power is firmly in his hands, as confirmed by his choice of two members of his branch of the family as crown
prince and prime minister, and by his appointments to the new government.”
235
Sabah Al-Ahmad
Start: September 21, 2001
End: Remained in power December 31, 201796
Years in dataset: 2011-2017
First leader of regime: No
MAURITANIA
Ould Taya
Start: December 12, 1984
End: August 3, 2005
Years in dataset: 1992-2005
First leader of regime: No
96
Economist Intelligence Unit 2018d.
236
Ould Abdel Aziz
Start: August 6, 2008
End: Remained in power December 31, 201797
Years in dataset: 2009-2017
First leader of regime: Yes
97
Goemans, Gleditsch, and Chiozza (2009) code Ba Mamadou M’Baré as Mauritania’s leader from April
15 to August 5, 2009. Abdel Aziz resigned from the presidency and the military on April 15, 2009, in order
to be legally eligible to run in the upcoming presidential election (Agence France Presse 2009b “Mauritanian
junta leader to run for president” ). Senate President M’Baré was appointed the interim president (BBC
Monitoring: Middle East 2009c. “Senegal’s foreign minister meets ousted Mauritanian leader behind closed
doors.” ). As Boucek (2009) writes, “It is unclear, however, whether Abdel-Aziz ever truly resigned form the
military junta” and the junta continued to govern in the form of the High Council of State, which was led by
an ally of Abdel Aziz (Agyeman-Togobo, Kissy 2009b “Election 2009: Mauritanian Junta Alters Timetable
for Presidential Candidates, Opposition Cries Foul.” ; Reuters 2009; Voice of America 2009a. Mauritania
Military and Politicians Agree to Postpone Saturday’s Election). In particular, it controlled state media,
which promoted Abdel Aziz as “the poor people’s president” (Ould Sadi 2009). Abdel Aziz also became the
head of the new ruling party within weeks of resigning (Agence France Presse 2009a “Former Mauritanian
junta chief elected party head.” ). In June, a few weeks before the election, a transitional government was
formed that included members of the opposition. However, Abdel Aziz appointed half of the ministers in the
cabinet, including the prime minister (Voice of America 2009b. “Power-Sharing Deal Signed in Mauritania.”
). Moreover, Abdel Aziz retained clear control of “the machinery of state” as he campaigned for president,
for instance, “lowering prices on electricity, water, sugar, and gas” (Pitman 2009). Abdel Aziz won the
presidential election on July 18, but the head of Mauritania’s election commission resigned soon after over
concerns over “the reliability of the election” (Agyeman-Togobo 2009a. Election 2009: Court Confirms
Former Junta Leader Mauritanian President) and Freedom House (2010a. Freedom in the World 2010)
notes: “Serious doubts have been raised about the legitimacy of the 2009 presidential election.” Thus, I code
Abdel Aziz as remaining in power throughout 2009.
237
MOROCCO
Hassan II
Start: February 26, 1961
End: July 23, 1999
Years in dataset: 1992-1999
First leader of regime: No
Mohammed VI
Start: July 23, 1999
End: Remained in power December 31, 201798
Years in dataset: 2000-2017
First leader of regime: No
98
Europa World 2019b. “Contemporary Political History (Morocco)”
238
SUDAN
Al-Bashir
Start: December 12, 199999
End: Remained in power December 31, 2017100
Years in dataset: 2000-2017
First leader of regime: No
TUNISIA
Ben Ali
Start: November 7, 1987
99
Although Goemans, Gleditsch, and Chiozza (2009) code Al-Bashir as the ruler of Sudan since the
1989 coup, numerous sources indicate that Hassan At-Turabi, leader of the National Islamic Front, was the
“power behind the throne” in the first decade of the regime (Joffe 2016; Human Rights Watch 2019). (See also
Woodward 2009.) In 1999, the National Assembly, of which At-Turabi was the speaker, introduced legislation
to restrict the power of the presidency, which Al-Bashir held (Europa World 2019d. “Contemporary Political
History (Sudan)” ; Human Rights Watch 2019). Al-Bashir dissolved the assembly on December 12, 1999,
before the vote could be held and imposed emergency rule (Europa World 2019d. “Contemporary Political
History (Sudan)” ). Thus, I follow Geddes, Wright, and Frantz (2014) in coding this episode as the leadership
transition from At-Turabi to Al-Bashir.
100
Economist Intelligence Unit 2018f.
239
End: January 14, 2011
Years in dataset: 1992-1999
First leader of regime: No
YEMEN
Saleh
Start: July 17, 1978
End: February 27, 2012
Years in dataset: 1992-2011
First leader of regime: Yes
240
Appendix 6: Histograms of
continuous explanatory variables
Figure 5.1: Distribution of unemployment rate (percentage)
241
Figure 5.2: Distribution of annual GDP per capita growth (percentage)
Figure 5.3: Distribution of fuel revenue per capita (2018 USD)
242
Figure 5.4: Distribution of oil revenue per capita (2018 USD)
Figure 5.5: Distribution of natural gas revenue per capita (2018 USD)
243
Figure 5.6: Distribution of years regime leader has been in power
Figure 5.7: Distribution of fuel revenue per capita (2018 USD), given greater than 0
244
Figure 5.8: Distribution of oil revenue per capita (2018 USD), given greater than 0
Figure 5.9: Distribution of natural gas revenue per capita (2018 USD), given greater than 0
245
Appendix 7: Dropping one regime at
a time: tolerated
This appendix shows the results of re-estimating the linear probability model of tolerated,
with one regime dropped at a time.
246
Table 5.1: Tolerated Opposition Media (Regime FE), dropped Algeria 1992-NA
Dependent variable:
tolerated
log(L unemployment)
−1.202∗∗∗
(0.168)
sqrt L GDP pc growth
0.050∗∗
(0.021)
log(L fuel + 1)
0.463∗∗∗
(0.078)
log(leader year office)
0.140∗∗∗
(0.047)
yrs00 05
0.038
(0.083)
yrs06 11
−0.205∗∗
(0.094)
yrs12 17
−0.397∗∗∗
(0.094)
Observations
176
R2
0.447
Adjusted R2
0.359
Note:
∗
p<0.1;
247
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.2: Tolerated Opposition Media (Regime FE), continued, dropped Algeria 1992-NA
Dependent variable:
tolerated
pres elec
0.170∗
(0.097)
leg elec
−0.090
(0.063)
leader fail
0.101
(0.169)
L civil war
0.081
(0.164)
is main
0.234∗
(0.120)
groups N
0.237∗∗∗
(0.060)
Observations
176
R2
0.447
Adjusted R2
0.359
Note:
∗
p<0.1;
∗∗
248
p<0.05;
∗∗∗
p<0.01
Table 5.3: Tolerated Opposition Media (Regime FE), dropped Bahrain 1971-NA
Dependent variable:
tolerated
log(L unemployment)
−0.913∗∗∗
(0.154)
sqrt L GDP pc growth
0.055∗∗
(0.023)
log(L fuel + 1)
0.308∗∗∗
(0.090)
log(leader year office)
0.102∗∗
(0.049)
regime first leader
−0.074
(0.319)
yrs00 05
−0.040
(0.088)
yrs06 11
−0.229∗∗
(0.106)
yrs12 17
−0.467∗∗∗
(0.107)
Observations
R
186
2
0.324
Adjusted R2
Note:
0.219
∗
p<0.1;
249
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.4: Tolerated Opposition Media (Regime FE), continued, dropped Bahrain 1971-NA
Dependent variable:
tolerated
pres elec
0.158∗
(0.094)
leg elec
−0.081
(0.069)
leader fail
0.117
(0.153)
L civil war
0.188
(0.148)
is main
0.335∗∗
(0.136)
groups N
0.295∗∗∗
(0.056)
Observations
186
R2
0.324
Adjusted R2
0.219
Note:
∗
p<0.1;
∗∗
250
p<0.05;
∗∗∗
p<0.01
Table 5.5: Tolerated Opposition Media (Regime FE), dropped Egypt 1952-2011
Dependent variable:
tolerated
log(L unemployment)
−1.061∗∗∗
(0.141)
sqrt L GDP pc growth
0.052∗∗
(0.021)
log(L fuel + 1)
0.298∗∗∗
(0.079)
log(leader year office)
0.074
(0.047)
regime first leader
−0.190
(0.303)
yrs00 05
−0.075
(0.092)
yrs06 11
−0.386∗∗∗
(0.102)
yrs12 17
−0.582∗∗∗
(0.097)
Observations
R
183
2
0.422
Adjusted R2
Note:
0.329
∗
p<0.1;
251
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.6: Tolerated Opposition Media (Regime FE), continued, dropped Egypt 1952-2011
Dependent variable:
tolerated
pres elec
0.239∗∗
(0.095)
leg elec
−0.058
(0.065)
leader fail
0.072
(0.144)
L civil war
0.136
(0.141)
is main
0.414∗∗∗
(0.119)
groups N
0.323∗∗∗
(0.051)
Observations
183
R2
0.422
Adjusted R2
0.329
Note:
∗
p<0.1;
∗∗
252
p<0.05;
∗∗∗
p<0.01
Table 5.7: Tolerated Opposition Media (Regime FE), dropped Egypt 2011-2012
Dependent variable:
tolerated
log(L unemployment)
−0.979∗∗∗
(0.143)
sqrt L GDP pc growth
0.049∗∗
(0.022)
log(L fuel + 1)
0.358∗∗∗
(0.079)
log(leader year office)
0.087∗
(0.047)
regime first leader
−0.101
(0.311)
yrs00 05
−0.024
(0.085)
yrs06 11
−0.252∗∗
(0.100)
yrs12 17
−0.502∗∗∗
(0.097)
Observations
R
200
2
0.375
Adjusted R2
Note:
0.286
∗
p<0.1;
253
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.8: Tolerated Opposition Media (Regime FE), continued, dropped Egypt 2011-2012
Dependent variable:
tolerated
pres elec
0.145
(0.093)
leg elec
−0.064
(0.065)
leader fail
0.151
(0.155)
L civil war
0.215
(0.142)
is main
0.338∗∗∗
(0.123)
groups N
0.306∗∗∗
(0.053)
Observations
200
R2
0.375
Adjusted R2
0.286
Note:
∗
p<0.1;
∗∗
254
p<0.05;
∗∗∗
p<0.01
Table 5.9: Tolerated Opposition Media (Regime FE), dropped Egypt 2013-NA
Dependent variable:
tolerated
log(L unemployment)
−0.971∗∗∗
(0.144)
sqrt L GDP pc growth
0.046∗∗
(0.022)
log(L fuel + 1)
0.373∗∗∗
(0.081)
log(leader year office)
0.085∗
(0.048)
regime first leader
−0.080
(0.312)
yrs00 05
−0.025
(0.085)
yrs06 11
−0.264∗∗∗
(0.100)
yrs12 17
−0.498∗∗∗
(0.098)
Observations
R
198
2
0.374
Adjusted R2
Note:
0.283
∗
p<0.1;
255
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.10: Tolerated Opposition Media (Regime FE), continued, dropped Egypt 2013-NA
Dependent variable:
tolerated
pres elec
0.171∗
(0.094)
leg elec
−0.073
(0.065)
leader fail
0.115
(0.150)
L civil war
0.220
(0.143)
is main
0.339∗∗∗
(0.123)
groups N
0.303∗∗∗
(0.053)
Observations
198
R2
0.374
Adjusted R2
0.283
Note:
∗
p<0.1;
∗∗
256
p<0.05;
∗∗∗
p<0.01
Table 5.11: Tolerated Opposition Media (Regime FE), dropped Jordan 1946-NA
Dependent variable:
tolerated
log(L unemployment)
−1.033∗∗∗
(0.154)
sqrt L GDP pc growth
0.021
(0.025)
log(L fuel + 1)
0.357∗∗∗
(0.081)
log(leader year office)
0.063
(0.061)
regime first leader
−0.096
(0.317)
yrs00 05
0.016
(0.090)
yrs06 11
−0.161
(0.115)
yrs12 17
−0.337∗∗∗
(0.118)
Observations
R
176
2
0.377
Adjusted R2
Note:
0.273
∗
p<0.1;
257
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.12: Tolerated Opposition Media (Regime FE), continued, dropped Jordan 1946-NA
Dependent variable:
tolerated
pres elec
0.149
(0.091)
leg elec
−0.045
(0.070)
leader fail
0.143
(0.159)
L civil war
0.268∗
(0.143)
is main
0.306∗∗
(0.124)
groups N
0.309∗∗∗
(0.055)
Observations
176
R2
0.377
Adjusted R2
0.273
Note:
∗
p<0.1;
∗∗
258
p<0.05;
∗∗∗
p<0.01
Table 5.13: Tolerated Opposition Media (Regime FE), dropped Kuwait 1961-NA
Dependent variable:
tolerated
log(L unemployment)
−0.508∗∗∗
(0.177)
sqrt L GDP pc growth
0.048∗∗
(0.022)
log(L fuel + 1)
0.503∗∗∗
(0.084)
log(leader year office)
0.066
(0.051)
regime first leader
−0.171
(0.302)
yrs00 05
−0.031
(0.086)
yrs06 11
−0.158
(0.102)
yrs12 17
−0.250∗∗
(0.113)
Observations
R
188
2
0.401
Adjusted R2
Note:
0.308
∗
p<0.1;
259
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.14: Tolerated Opposition Media (Regime FE), continued, dropped Kuwait 1961-NA
Dependent variable:
tolerated
pres elec
0.177∗∗
(0.088)
leg elec
−0.065
(0.065)
leader fail
0.146
(0.159)
L civil war
0.234∗
(0.138)
is main
0.335∗∗∗
(0.121)
groups N
0.356∗∗∗
(0.053)
Observations
188
R2
0.401
Adjusted R2
0.308
Note:
∗
p<0.1;
∗∗
260
p<0.05;
∗∗∗
p<0.01
Table 5.15: Opposition Media Bans (Regime FE), dropped Mauritania 2008-NA
Dependent variable:
banned
log(L unemployment)
0.237
(0.266)
sqrt L GDP pc growth
−0.083∗∗∗
(0.028)
log(L fuel + 1)
−0.454∗∗∗
(0.149)
log(leader year office)
−0.078
(0.056)
regime first leader
−0.019
(0.328)
yrs00 05
0.061
(0.142)
yrs06 10
0.264∗
(0.155)
yr11
0.456∗∗
(0.202)
yrs12 17
0.179
(0.150)
Observations
R
124
2
Adjusted R
Note:
0.256
2
261
∗
p<0.1;
0.076
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.16: Opposition Media Bans (Regime FE), continued, dropped Mauritania 2008-NA
Dependent variable:
banned
pres elec
−0.056
(0.140)
leg elec
0.069
(0.078)
leader fail
0.178
(0.183)
L civil war
−0.269
(0.199)
is main
−0.292
(0.227)
groups N
−0.133
(0.097)
Observations
124
R2
0.256
Adjusted R2
0.076
Note:
∗
p<0.1;
∗∗
262
p<0.05;
∗∗∗
p<0.01
Table 5.17: Opposition Media Bans (Regime FE), dropped Morocco 1956-NA
Dependent variable:
banned
log(L unemployment)
0.233
(0.262)
sqrt L GDP pc growth
−0.092∗∗∗
(0.030)
log(L fuel + 1)
−0.455∗∗∗
(0.154)
log(leader year office)
−0.038
(0.065)
regime first leader
0.084
(0.324)
yrs00 05
0.043
(0.148)
yrs06 10
0.246
(0.161)
yr11
0.560∗∗
(0.218)
yrs12 17
0.231
(0.157)
Observations
R
108
2
Adjusted R
Note:
0.320
2
263
∗
p<0.1;
0.123
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.18: Opposition Media Bans (Regime FE), continued, dropped Morocco 1956-NA
Dependent variable:
banned
pres elec
−0.042
(0.128)
leg elec
0.103
(0.083)
leader fail
0.274
(0.205)
L civil war
−0.232
(0.198)
is main
−0.343
(0.232)
groups N
−0.176
(0.106)
Observations
108
R2
0.320
Adjusted R2
0.123
Note:
∗
p<0.1;
∗∗
264
p<0.05;
∗∗∗
p<0.01
Table 5.19: Opposition Media Bans (Regime FE), dropped Sudan 1989-NA
Dependent variable:
banned
log(L unemployment)
0.354
(0.258)
sqrt L GDP pc growth
−0.094∗∗∗
(0.027)
log(L fuel + 1)
−0.498∗∗∗
(0.156)
log(leader year office)
−0.078
(0.058)
regime first leader
−0.005
(0.338)
yrs00 05
0.048
(0.147)
yrs06 10
0.286∗
(0.158)
yr11
0.473∗∗
(0.218)
yrs12 17
0.144
(0.156)
Observations
R
124
2
Adjusted R
Note:
0.290
2
265
∗
p<0.1;
0.117
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.20: Opposition Media Bans (Regime FE), continued, dropped Sudan 1989-NA
Dependent variable:
banned
pres elec
−0.027
(0.140)
leg elec
0.069
(0.079)
leader fail
0.174
(0.187)
L civil war
−0.393
(0.263)
is main
−0.425∗
(0.215)
groups N
−0.217∗∗
(0.083)
Observations
124
R2
0.290
Adjusted R2
0.117
Note:
∗
p<0.1;
∗∗
266
p<0.05;
∗∗∗
p<0.01
Table 5.21: Opposition Media Bans (Regime FE), dropped Yemen 1978-2012
Dependent variable:
banned
log(L unemployment)
0.393
(0.271)
sqrt L GDP pc growth
−0.097∗∗∗
(0.026)
log(L fuel + 1)
−0.526∗∗∗
(0.148)
log(leader year office)
−0.060
(0.063)
regime first leader
0.184
(0.323)
yrs00 05
0.156
(0.161)
yrs06 10
0.365∗∗
(0.169)
yr11
0.342
(0.228)
yrs12 17
0.200
(0.157)
Observations
R
110
2
Adjusted R
Note:
0.335
2
267
∗
p<0.1;
0.147
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.22: Opposition Media Bans (Regime FE), continued, dropped Yemen 1978-2012
Dependent variable:
banned
pres elec
−0.016
(0.144)
leg elec
0.097
(0.079)
leader fail
0.266
(0.180)
L civil war
−0.222
(0.227)
is main
−0.763∗∗
(0.352)
groups N
−0.321∗∗∗
(0.109)
Observations
110
R2
0.335
Adjusted R2
0.147
Note:
∗
p<0.1;
∗∗
268
p<0.05;
∗∗∗
p<0.01
Appendix 8: Dropping one regime at
a time: banned
This appendix shows the results of re-estimating the linear probability model of banned, with
one regime dropped at a time.
269
Table 5.23: Opposition Media Bans (Regime FE), dropped Algeria 1992-NA
Dependent variable:
banned
log(L unemployment)
0.424
(0.317)
sqrt L GDP pc growth
−0.084∗∗∗
(0.029)
log(L fuel + 1)
−0.530∗∗∗
(0.177)
log(leader year office)
−0.068
(0.068)
yrs00 05
0.045
(0.165)
yrs06 10
0.273
(0.165)
yr11
0.505∗∗
(0.226)
yrs12 17
0.131
(0.162)
Observations
R
115
2
0.253
Adjusted R2
Note:
0.064
∗
p<0.1;
270
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.24: Opposition Media Bans (Regime FE), continued, dropped Algeria 1992-NA
Dependent variable:
banned
pres elec
−0.045
(0.160)
leg elec
0.062
(0.083)
leader fail
−0.029
(0.217)
L civil war
−0.080
(0.249)
is main
−0.428∗
(0.238)
groups N
−0.235∗∗
(0.094)
Observations
115
R2
0.253
Adjusted R2
0.064
Note:
∗
p<0.1;
∗∗
271
p<0.05;
∗∗∗
p<0.01
Table 5.25: Opposition Media Bans (Regime FE), dropped Bahrain 1971-NA
Dependent variable:
banned
log(L unemployment)
0.439
(0.274)
sqrt L GDP pc growth
−0.083∗∗∗
(0.030)
log(L fuel + 1)
−0.445∗∗
(0.189)
log(leader year office)
−0.085
(0.058)
regime first leader
0.032
(0.336)
yrs00 05
0.038
(0.147)
yrs06 10
0.258
(0.164)
yr11
0.482∗∗
(0.213)
yrs12 17
0.228
(0.171)
Observations
R
116
2
Adjusted R
Note:
0.264
2
272
∗
p<0.1;
0.070
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.26: Opposition Media Bans (Regime FE), continued, dropped Bahrain 1971-NA
Dependent variable:
banned
pres elec
−0.079
(0.133)
leg elec
0.059
(0.086)
leader fail
0.179
(0.188)
L civil war
−0.288
(0.211)
is main
−0.430∗
(0.223)
groups N
−0.215∗∗
(0.085)
Observations
116
R2
0.264
Adjusted R2
0.070
Note:
∗
p<0.1;
∗∗
273
p<0.05;
∗∗∗
p<0.01
Table 5.27: Opposition Media Bans (Regime FE), dropped Egypt 1952-2011
Dependent variable:
banned
log(L unemployment)
0.429∗
(0.250)
sqrt L GDP pc growth
−0.088∗∗∗
(0.026)
log(L fuel + 1)
−0.414∗∗
(0.158)
log(leader year office)
−0.089
(0.054)
regime first leader
0.001
(0.313)
yrs00 05
0.011
(0.140)
yrs06 10
0.305∗∗
(0.148)
yr11
0.467∗∗
(0.194)
yrs12 17
0.200
(0.145)
Observations
R
122
2
Adjusted R
Note:
0.308
2
274
∗
p<0.1;
0.137
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.28: Opposition Media Bans (Regime FE), continued, dropped Egypt 1952-2011
Dependent variable:
banned
pres elec
−0.186
(0.133)
leg elec
−0.005
(0.076)
leader fail
0.156
(0.175)
L civil war
−0.283
(0.191)
is main
−0.443∗∗
(0.203)
groups N
−0.216∗∗∗
(0.079)
Observations
122
R2
0.308
Adjusted R2
0.137
Note:
∗
p<0.1;
∗∗
275
p<0.05;
∗∗∗
p<0.01
Table 5.29: Opposition Media Bans (Regime FE), dropped Egypt 2011-2012
Dependent variable:
banned
log(L unemployment)
0.325
(0.254)
sqrt L GDP pc growth
−0.087∗∗∗
(0.027)
log(L fuel + 1)
−0.484∗∗∗
(0.150)
log(leader year office)
−0.072
(0.056)
regime first leader
0.025
(0.326)
yrs00 05
0.071
(0.142)
yrs06 10
0.292∗
(0.154)
yr11
0.510∗∗
(0.208)
yrs12 17
0.170
(0.150)
Observations
R
128
2
Adjusted R
Note:
0.284
2
276
∗
p<0.1;
0.118
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.30: Opposition Media Bans (Regime FE), continued, dropped Egypt 2011-2012
Dependent variable:
banned
pres elec
−0.093
(0.132)
leg elec
0.069
(0.078)
leader fail
0.242
(0.192)
L civil war
−0.257
(0.199)
is main
−0.428∗∗
(0.211)
groups N
−0.211∗∗
(0.082)
Observations
128
R2
0.284
Adjusted R2
0.118
Note:
∗
p<0.1;
∗∗
277
p<0.05;
∗∗∗
p<0.01
Table 5.31: Opposition Media Bans (Regime FE), dropped Egypt 2013-NA
Dependent variable:
banned
log(L unemployment)
0.417
(0.254)
sqrt L GDP pc growth
−0.091∗∗∗
(0.026)
log(L fuel + 1)
−0.384∗∗
(0.161)
log(leader year office)
−0.085
(0.056)
regime first leader
0.060
(0.324)
yrs00 05
0.044
(0.142)
yrs06 10
0.268∗
(0.153)
yr11
0.467∗∗
(0.201)
yrs12 17
0.192
(0.149)
Observations
R
126
2
Adjusted R
Note:
0.277
2
278
∗
p<0.1;
0.106
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.32: Opposition Media Bans (Regime FE), continued, dropped Egypt 2013-NA
Dependent variable:
banned
pres elec
−0.055
(0.135)
leg elec
0.073
(0.078)
leader fail
0.187
(0.181)
L civil war
−0.248
(0.198)
is main
−0.417∗∗
(0.210)
groups N
−0.214∗∗∗
(0.081)
Observations
126
R2
0.277
Adjusted R2
0.106
Note:
∗
p<0.1;
∗∗
279
p<0.05;
∗∗∗
p<0.01
Table 5.33: Opposition Media Bans (Regime FE), dropped Jordan 1946-NA
Dependent variable:
banned
log(L unemployment)
0.282
(0.284)
sqrt L GDP pc growth
−0.070∗∗
(0.032)
log(L fuel + 1)
−0.476∗∗∗
(0.155)
log(leader year office)
−0.101
(0.070)
regime first leader
−0.098
(0.335)
yrs00 05
−0.043
(0.178)
yrs06 10
0.147
(0.208)
yr11
0.237
(0.256)
yrs12 17
−0.032
(0.217)
Observations
R
104
2
Adjusted R
Note:
0.305
2
280
∗
p<0.1;
0.094
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.34: Opposition Media Bans (Regime FE), continued, dropped Jordan 1946-NA
Dependent variable:
banned
pres elec
−0.079
(0.130)
leg elec
−0.034
(0.088)
leader fail
0.220
(0.197)
L civil war
−0.322
(0.200)
is main
−0.341
(0.230)
groups N
−0.224∗∗
(0.088)
Observations
104
R2
0.305
Adjusted R2
0.094
Note:
∗
p<0.1;
∗∗
281
p<0.05;
∗∗∗
p<0.01
Table 5.35: Opposition Media Bans (Regime FE), dropped Kuwait 1961-NA
Dependent variable:
banned
log(L unemployment)
0.394
(0.275)
sqrt L GDP pc growth
−0.091∗∗∗
(0.028)
log(L fuel + 1)
−0.483∗∗∗
(0.158)
log(leader year office)
−0.084
(0.068)
regime first leader
0.042
(0.340)
yrs00 05
0.095
(0.180)
yrs06 10
0.320∗
(0.182)
yr11
0.502∗∗
(0.220)
yrs12 17
0.213
(0.175)
Observations
R
123
2
Adjusted R
Note:
0.288
2
282
∗
p<0.1;
0.114
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.36: Opposition Media Bans (Regime FE), continued, dropped Kuwait 1961-NA
Dependent variable:
banned
pres elec
−0.076
(0.134)
leg elec
0.073
(0.082)
leader fail
0.172
(0.240)
L civil war
−0.265
(0.212)
is main
−0.465∗∗
(0.234)
groups N
−0.219∗∗
(0.084)
Observations
123
R2
0.288
Adjusted R2
0.114
Note:
∗
p<0.1;
∗∗
283
p<0.05;
∗∗∗
p<0.01
Table 5.37: Opposition Media Bans (Regime FE), dropped Mauritania 2008-NA
Dependent variable:
banned
log(L unemployment)
0.237
(0.266)
sqrt L GDP pc growth
−0.083∗∗∗
(0.028)
log(L fuel + 1)
−0.454∗∗∗
(0.149)
log(leader year office)
−0.078
(0.056)
regime first leader
−0.019
(0.328)
yrs00 05
0.061
(0.142)
yrs06 10
0.264∗
(0.155)
yr11
0.456∗∗
(0.202)
yrs12 17
0.179
(0.150)
Observations
R
124
2
Adjusted R
Note:
0.256
2
284
∗
p<0.1;
0.076
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.38: Opposition Media Bans (Regime FE), continued, dropped Mauritania 2008-NA
Dependent variable:
banned
pres elec
−0.056
(0.140)
leg elec
0.069
(0.078)
leader fail
0.178
(0.183)
L civil war
−0.269
(0.199)
is main
−0.292
(0.227)
groups N
−0.133
(0.097)
Observations
124
R2
0.256
Adjusted R2
0.076
Note:
∗
p<0.1;
∗∗
285
p<0.05;
∗∗∗
p<0.01
Table 5.39: Opposition Media Bans (Regime FE), dropped Morocco 1956-NA
Dependent variable:
banned
log(L unemployment)
0.233
(0.262)
sqrt L GDP pc growth
−0.092∗∗∗
(0.030)
log(L fuel + 1)
−0.455∗∗∗
(0.154)
log(leader year office)
−0.038
(0.065)
regime first leader
0.084
(0.324)
yrs00 05
0.043
(0.148)
yrs06 10
0.246
(0.161)
yr11
0.560∗∗
(0.218)
yrs12 17
0.231
(0.157)
Observations
R
108
2
Adjusted R
Note:
0.320
2
286
∗
p<0.1;
0.123
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.40: Opposition Media Bans (Regime FE), continued, dropped Morocco 1956-NA
Dependent variable:
banned
pres elec
−0.042
(0.128)
leg elec
0.103
(0.083)
leader fail
0.274
(0.205)
L civil war
−0.232
(0.198)
is main
−0.343
(0.232)
groups N
−0.176
(0.106)
Observations
108
R2
0.320
Adjusted R2
0.123
Note:
∗
p<0.1;
∗∗
287
p<0.05;
∗∗∗
p<0.01
Table 5.41: Opposition Media Bans (Regime FE), dropped Sudan 1989-NA
Dependent variable:
banned
log(L unemployment)
0.354
(0.258)
sqrt L GDP pc growth
−0.094∗∗∗
(0.027)
log(L fuel + 1)
−0.498∗∗∗
(0.156)
log(leader year office)
−0.078
(0.058)
regime first leader
−0.005
(0.338)
yrs00 05
0.048
(0.147)
yrs06 10
0.286∗
(0.158)
yr11
0.473∗∗
(0.218)
yrs12 17
0.144
(0.156)
Observations
R
124
2
Adjusted R
Note:
0.290
2
288
∗
p<0.1;
0.117
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.42: Opposition Media Bans (Regime FE), continued, dropped Sudan 1989-NA
Dependent variable:
banned
pres elec
−0.027
(0.140)
leg elec
0.069
(0.079)
leader fail
0.174
(0.187)
L civil war
−0.393
(0.263)
is main
−0.425∗
(0.215)
groups N
−0.217∗∗
(0.083)
Observations
124
R2
0.290
Adjusted R2
0.117
Note:
∗
p<0.1;
∗∗
289
p<0.05;
∗∗∗
p<0.01
Table 5.43: Opposition Media Bans (Regime FE), dropped Yemen 1978-2012
Dependent variable:
banned
log(L unemployment)
0.393
(0.271)
sqrt L GDP pc growth
−0.097∗∗∗
(0.026)
log(L fuel + 1)
−0.526∗∗∗
(0.148)
log(leader year office)
−0.060
(0.063)
regime first leader
0.184
(0.323)
yrs00 05
0.156
(0.161)
yrs06 10
0.365∗∗
(0.169)
yr11
0.342
(0.228)
yrs12 17
0.200
(0.157)
Observations
R
110
2
Adjusted R
Note:
0.335
2
290
∗
p<0.1;
0.147
∗∗
p<0.05;
∗∗∗
p<0.01
Table 5.44: Opposition Media Bans (Regime FE), continued, dropped Yemen 1978-2012
Dependent variable:
banned
pres elec
−0.016
(0.144)
leg elec
0.097
(0.079)
leader fail
0.266
(0.180)
L civil war
−0.222
(0.227)
is main
−0.763∗∗
(0.352)
groups N
−0.321∗∗∗
(0.109)
Observations
110
R2
0.335
Adjusted R2
0.147
Note:
∗
p<0.1;
∗∗
291
p<0.05;
∗∗∗
p<0.01
Appendix 9: Alternative coding of
main opposition
The following four main opposition groups are affected by the alternative coding rule in
which main opposition groups that are driven underground by repression continue to be
coded as the main opposition group for seven years. In each case, I list the additional years
after a crackdown in which the group continues to be coded as the main group under the
alternative rule, but not under the original rule. I also list other major opposition groups,
which appear in the dataset during the affected years under the original rule but not under
the alternative rule.
Algeria: 1992-1994
Regime: 1992-NA
Main Group: Islamic Salvation Front (FIS)101
101
Unlike the other main groups affected by the alternative coding, the FIS does not appear as a main
opposition group in the original dataset, because the first year it appears is also the year it was driven
underground. However, as it was clearly the main opposition group during 1990 and 1991, it would count
as a repressed main opposition group starting in 1992 under the alternative coding rule. The FIS is coded
as an armed group after 1994.
292
Other Major Groups: Berber Cultural Movement, Movement for Democracy in Algeria,
National Liberation Front, Socialist Forces Front
Bahrain: 2017
Regime: 1971-NA
Main Group: The Accord National Islamic Society (Al-Wefaq)
Other Major Groups: None
Egypt: 2014-2017
Regime: 2013-NA
Main Group: Muslim Brotherhood
Other Major Groups: April 6 Youth Movement, No to Military Trials, Salafist Front
Tunisia 1993-1999
Regime: 1956-2011
Main Group: Party of the Renaissance (Ennahda)
Other Major Groups: None
293
References
Abdelaziz, Salma.
2014.
“2 dead in clashes between Egyptian security, Muslim
Brotherhood protestors.” January 1.
CNN. Accessed March 28, 2019:
www.dowjones.com/products/factiva
Abed-Kotob, Sana. 1995. “The Accommodationists Speak: Goals and Strategies of the
Muslim Brotherhood of Egypt.” International Journal of Middle East Studies 27(3):
321- 339
Abou El-Magd, Nadia. 2005. “5,000 reform activists, mostly Islamists, protest and support
judges’ insistence on fair elections.” July 20. Associated Press. Accessed February 16,
2020: www.dowjones.com/products/factiva
Abouyoub, Younes.
2009.
“Al-Wefaq National Islamic Society.” In The Oxford
Encyclopedia of Islam and Politics. John L. Esposito, editor. Accessed February 8,
2019: www.oxfordislamicstudies.com
Accord National Islamic Society, The. 2019a. “Al-Wefaq National Islamic Society.” Internet
Archive. Accessed November 7, 2019: http://web.archive.org/web/∗ /alwefaq.net
Accord National Islamic Society, The. 2019b. “Al-Wefaq National Islamic Society.” Internet
Archive. Accessed November 7, 2019: http://web.archive.org/web/∗ /alwefaq.org
Acemoglu, Daron and James A. Robinson. 2006. Economic Origins of Dictatorship and
Democracy. New York: Cambridge University Press.
294
Addi, Lahouari. 2014. “Algeria.” In The Middle East. Thirteenth Edition. Ellen Lust,
editor. Washington, DC: Sage and CQ Press. Pages 429-447.
Aflou, Lyes. 2014. “Little Change in New Bouteflika Cabinet.” May 9. All Africa. Accessed
April 22, 2019: www.dowjones.com/products/factiva
Africa Confidential. 1988. “Tunisia: Ben Ali’s Boys.” 29(17): 7.
Africa Confidential. 1990. “Morocco: Watching for trouble.” April 20. 31 (8): 5-7.
Africa Confidential. 1991a. “Morocco: Hassan’s minor troubles.” April 5. 32 (7): 2-5.
Africa Confidential. 1991b. “Tunisia: Old Habits Return.” 32(4).
Africa Confidential. 1992. “Morocco: The Sahara consensus breaks down.” July 17. 33 (14):
4-6.
Africa Confidential. 1993. “Morocco: Pomp and changing circumstance.” September 10. 34
(18): 5-6.
Africa Confidential. 1994. “Morocco: All the King’s men.” January 7. 35 (1): 6-7.
Africa Confidential. 2019. “Mohamed Ould Abdelaziz.” Accessed November 15, 2019:
www.africa-confidential.com
Africa Report. 1988. “Ben Ali Tackles Reforms in Post-Bourguiba Tunisia.” 1: 8.
Agence de Presse Africaine. 2011. “Une marche pacifique de jeunes à Nouakchott dispersée
par la police.” March 9. Accessed May 22, 2019: www.dowjones.com/products/factiva
Agence Ecofin. 2016. “Mauritanie: suspension de l’émission ‘Vissamim’ sur la chaine
Mourabitoun
TV.”
February
22.
Accessed
August
16,
2019:
www.dowjones.com/products/factiva
Agence France Presse. 2001a. “Spiral of violence in Algeria’s Kabylie region.” April 29.
Accessed July 24, 2019: www.dowjones.com/products/factiva
295
Agence France Presse. 2001b. “Transition of power in Kuwait promises to be tricky.”
September 21. Accessed November 19, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2002. “Bouteflika names new Algerian government.” June 18.
Accessed May 13, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2003. “Mauritanians protest against war on Iraq.” February 6.
Accessed May 21, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2004. “Algeria’s re-elected Bouteflika has way clear to consolidate
peace.” April 11. Accessed May 11, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2005. “Syrians stage rare protest at trial of rights activists.” April
24. Accessed June 21, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2006. “Kuwait parliament to meet to remove ailing emir.” January
23. Accessed November 15, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2007. “New Kuwait cabinet sworn in.” March 26. Accessed May 27,
2019: www.dowjones.com/products/factiva
Agence France Presse. 2008a. “Algerian president brings back Ouyahia for third stint as
PM.” June 23. Accessed April 22, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2008b. “Mauritanie: l’Assemblée vote la confiance au gouvernement
issu
du
putsch.”
September
19.
Accessed
May
23,
2019:
www.dowjones.com/products/factiva
Agence France Presse. 2008d. “Mauritanie: plusieurs milliers de participants à une marche
anti-putsch.” August 20. Accessed July 26, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2008e. “Mauritanie/gouvernement: ‘rejet absolu de cette formation
illégale.’” September 1. Accessed July 26, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2009a. “Former Mauritanian junta chief elected party head.” May 6.
Accessed November 10, 2019: www.dowjones.com/products/factiva
296
Agence France Presse. 2009b. “Mauritanian junta leader to run for president.” April 15.
Accessed November 10, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2009c. “Mauritanian speaker and opposition leader to run for
president.” June 15. Accessed July 26, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2009d. “Mauritanie: des milliers de manifestants contre la
présidentielle du 6 juin.” May 18.
Accessed July 31,
2019:
www.dowjones.com/products/factiva
Agence France Presse. 2010. “Kuwait MPs oppose deal for French Rafale warplanes.” March
21. Accessed May 31, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2011a. “Algerian Islamists’ former number two indicted.” January
19. Accessed May 4, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2011b. “Jordan Islamist opposition says website hacked.” March 27.
Accessed October 19, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2012. “Kuwaitis rally against amending electoral districts.” August
28. Accessed May 31, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2014. “Mauritania since Abdel Aziz seized power.” June 21. Accessed
May 2, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2015. “Yemen president resigns: advisers.” January 22. Accessed
November 11, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2016a. “Bahrain court orders Shiite opposition bloc dissolved.” July
17. Accessed April 9, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2016b. “Opposition vows to rid Morocco of Islamists in Oct vote.”
August 29. Accessed April 8, 2019: www.dowjones.com/products/factiva
Agence France Presse. 2018. “AFP and Reuters journalists detained over covering Sudan
protests.” January 18. Accessed October 17, 2019: www.dowjones.com/products/factiva
297
Agenzia Nazionale Stampa Associata. 2004. “Algeria: Berbers Boycott Presidential
Election.” February 13. Accessed May 18, 2019: www.dowjones.com/products/factiva
Aghrout, Ahmed and Yahia Zoubir. 2019. “History (Algeria).” In Europa World Online.
Routledge. Accessed May 4, 2019: http://europaworld.com
Agyeman-Togobo, Kissy. 2009a. “Election 2009: Court Confirms Former Junta Leader
Mauritanian President.” July 24. IHS Global Insight Daily Analysis. Accessed November
10, 2019: www.dowjones.com/products/factiva
Agyeman-Togobo, Kissy. 2009b. “Election 2009: Mauritanian Junta Alters Timetable for
Presidential Candidates, Opposition Cries Foul.” June 25. IHS Global Insight Daily
Analysis. Accessed November 10, 2019: www.dowjones.com/products/factiva
Agyeman, Kissy. 2008. “U.S. Imposes Sanctions on Mauritania as Pro- and Anti-Coup
Demonstrations Hit Capital.” August 8. IHS Markit. Accessed May 21, 2019:
www.dowjones.com/products/factiva
Ahdaf. Accessed June 7, 2019: www.wahdawi.org
Ahram Online. 2011. “Ghad Al-Thawra Party.” December 3. Accessed June 25, 2019:
http://english.ahram.org.eg
Ahram Online. 2013. “Brotherhood’s Freedom and Justice newspaper suspended.”
December 26. Accessed September 24, 2019: http://english.ahram.org.eg
Aishima, Hatsuki. 2016. Public Culture and Islam in Modern Egypt: Media, Intellectuals
and Society. London: Bloomsbury Publishing.
Al A‘ali, Mohammed. 2011. “Protest Rally.” March 7.
www.dowjones.com/products/factiva
Accessed June 4, 2019:
Al Abed, Raed. 1997. “Weeklies Almost Declare Victory Over Press Law.” July 31. The
Star. Accessed August 10, 2019: www.dowjones.com/products/factiva
Al Arabiya. 2013. “Military trials back to haunt Egyptian civilians.” December 2. Accessed
June 24, 2019: www.dowjones.com/products/factiva
298
Al
Qå JkQK Im ZAªËAK. A¯ñË@ øñ«X ¯QK éK P@XB @” [The
Ayam.
2012.
“ AîE@
.
Administration Rejects Al-Wefaq’s Lawsuit to Revoke the Withdrawal of its
Publications License] October 1. Accessed November 16, 2019: www.alayam.com
Al Ayam. 2016. “ A¯ñË@
Ï @ Imk” [Al-Wefaq Website Blocked] June 14.
éJ ªÒm.Ì úGð QºËB @ ©¯ñÖ
..
Accessed November 6, 2019: www.alayam.com
Al Jazeera 2019. “Yemen conflict: Who controls what.” March 24. Accessed December 29,
2019: www.aljazeera.com
Al Jazeera. 2005. “Profile: Maaouya Ould Sid’Ahmed Taya.” August 3. Accessed November
15, 2019: www.aljazeera.com
Al Jazeera. 2010. “Q and A: Iraq’s Election Row.” February 13. Accessed September 10,
2019: https://www.aljazeera.com
Al Jazeera. 2011a. “Who’s who in Yemen’s opposition? As protests against Ali Abdullah
Saleh continue, opposition remains fractured, riven with differences.” March 10. Accessed
March 21, 2019: www.aljazeera.com
é»QmÌ '@” [Kuwaiti Youth Movement] March 8. Accessed
Al Jazeera. 2011b. “ éJK ñºË@ éJK. AJ.Ë@
October 4, 2019: www.aljazeera.com
Al Jazeera. 2014. “Yemen’s Houthi leader warns of civil unrest.” September 1. Accessed
November 11, 2019: www.dowjones.com/products/factiva
Al Jazeera. 2016. “Yemeni parties welcome cabinet reshuffle.” April 4. Accessed June 23,
2019: www.dowjones.com/products/factiva
Al Jazeera. 2017b. “How did Yemen’s Houthi-Saleh alliance collapse?” December 4.
Accessed June 23, 2019: www.dowjones.com/products/factiva
Al Jazeera. 2017c. “Morocco’s king names new coalition government.” April 5. Accessed
April 8, 2019: www.dowjones.com/products/factiva
Al-Anani, Khalil. 2016. Inside the Muslim Brotherhood: Religion, Identity, and Politics.
New York: Oxford University Press.
299
Al-Anani, Khalil. 2017. “What Happened to Egypt’s Muslim Brotherhood?” February 14.
Al Jazeera. Accessed March 28, 2019: www.aljazeera.com
Al-Araby. 2017. “Strained relations: Houthis shut down pro-Saleh newspaper.” May 16.
The New Arab blog. Accessed October 1, 2019: www.alaraby.co.uk
Al-Chahed. 2019. “ YëAË@
èAJ¯” [Al-Chahed (Witness) Channel] Internet Archive.
Accessed
October 10, 2019: http://web.archive.org/web/∗ /chahed.tv
Al-Falahi, Ashraf. 2015. “Islah’s Houthi Gamble.” January 30. Carnegie Endowment for
International Peace. Sada blog. Accessed June 23, 2019: https://carnegieendowment.org
Al-Haj, Ahmed and Brian Rohan. 2015. “Leader of Yemen’s rebels urges ‘peaceful transfer
of power.’” January 27.
Associated Press.
Accessed November 11, 2019:
www.dowjones.com/products/factiva
Al-Haj, Ahmed. 2012. “Yemeni president to remain in country in apparent bid to maintain
unravelling
control.”
January
1.
Accessed
January
5,
2020:
www.dowjones.com/products/factiva
Al-Islah. 2019. “Yemeni Islah Party.” Internet Archive.
http://web.archive.org/web/∗ /al-islah.net
Accessed October 3, 2019:
Al-Masmari, Hakim and Margaret Coker. 2011. “Thousands protest regime in Yemen.”
January 28.
Wall Street Journal.
Accessed February 15, 2020:
www.dowjones.com/products/factiva
Al-Masry Al-Youm. 2012. “" l. ' AJK áʪK " AJʪË@” [‘Supreme’ Announces Results] January 21.
Accessed June 25, 2019: www.almasryalyoum.com
Al-Qarawee, Harith Hasan. 2014. “Iraq’s Sectarian Crisis: A Legacy of Exclusion.” April
23.
Carnegie Endowment of International Peace.
Accessed April 4, 2019:
https://carnegieendowment.org
Al-Rasheed, Madawi. 2007. Contesting the Saudi State: Islamic Voices from a New
Generation. New York: Cambridge University Press.
300
K
Al-Sahwa Net. 2019. “ I
èñjË@”
[Al-Sahwa [The Awakening] Net] Internet Archive.
Accessed October 1, 2019: http://web.archive.org/web/∗ /alsahwa-yemen.net
Al-Sakkaf, Nasser and Andrew England. 2018. “Yemen: Trapped in crossfire of regional
struggle.” January 11.
Financial Times.
Accessed November 11, 2019:
www.dowjones.com/products/factiva
Albloshi, Hamad H. and Michael Herb. 2018. “Karamet Watan: An Unsuccessful Nonviolent
Movement.” The Middle East Journal 72(3): 408-430.
Albrecht, Holger. 2005. “How can opposition support authoritarianism? Lessons from
Egypt.” Democratization 12(3): 378-397.
Alexander, Christopher. 1997. “Back from the Democratic Brink: Authoritarianism and
Civil Society in Tunisia.” Middle East Report 205: 34-38.
Alexander, Christopher. 2010. Tunisia: Stability and Reform in the Modern Maghreb. New
York: Routledge.
All Africa. 2011d. “Le parti Tawassoul decide de rejoindre la l’apposition [sic] démocratique.”
June 2. Accessed March 15, 2020: www.dowjones.com/products/factiva
All Africa. 1999. “Current Cabinet to Remain Until At Least OAU Summit.” April 22.
Accessed April 22, 2019: www.dowjones.com/products/factiva
All Africa. 2004. “Démonstration de force des arouch à Tizi Ouzou.” April 21. Accessed
July 25, 2019: www.dowjones.com/products/factiva
All Africa. 2009a. “Crise politique — Quand Ahmed Ould Daddah souffle le chaud et le
froid.” February 5. Accessed May 23, 2019: www.dowjones.com/products/factiva
All Africa. 2009b. “Election sénatoriale partielle — L’UPR et Tewassoul present une list
commune.”
October
8.
Accessed
Accessed
May
23,
2019:
www.dowjones.com/products/factiva
All Africa. 2011a. “Le parti Tawassoul decide de rejoindre la l’apposition [sic] démocratique.”
June 2. Accessed December 7, 2019: www.dowjones.com/products/factiva
301
All Africa. 2011b. “Country’s Largest Opposition Party Formally Rejects Participation in
New
Government.”
September
25.
Accessed
April
1,
2019:
www.dowjones.com/products/factiva
All Africa. 2011c. “President Names New Government, Political Police Dissolved.” March
7. Accessed January 5, 2020: www.dowjones.com/products/factiva
All Africa. 2012. “Des sympathisants de Biram Ould Dah manifestent pour exiger sa
liberation.” May 5. Accessed July 26, 2019: www.dowjones.com/products/factiva
All Africa. 2013. “Aleg — Le parti El-Wiam organise un sit-in devant le tribunal de la
Wilaya.” December 30. Accessed May 21, 2019: www.dowjones.com/products/factiva
All Africa. 2014. “Election - Des jeunes journalistes invitent les candidats à un débat
présidentiel.” May 22. Accessed August 16, 2019: www.dowjones.com/products/factiva
All Africa. 2017. “Updated List of Blocked Websites in Egypt Reveal False Allegations By
Egyptian Authorities.” May 30.
Accessed September 23,
2019:
www.dowjones.com/products/factiva
Allen, Nathaniel. 2019. “Interrogating Ethnic Stacking: The Uses and Abuses of Security
Force Ethnicity in Sudan.” Civil Wars: 1-23.
Allison, Paul D. 2009. Fixed Effects Regression Models. Thousand Oaks, CA: SAGE
Publications.
Althawrah.
2019a.
“Althawrah.” Internet Archive.
http://web.archive.org/web/∗ /althawrah.ye
Accessed October 3, 2019:
Althawrah.
2019b.
“Althawrah.” Internet Archive.
http://web.archive.org/web/∗ /althawranews.net
Accessed October 3, 2019:
Álvarez-Ossorio, Ignacio and Naomı́ Ramı́rez Dı́az. 2017. “Understanding the Muslim
Brotherhood in Syria.” In Political Islam in a Time of Revolt. Ferran Izquierdo Brichs,
John Etherington, and Laura Feliu, editors. New York: Palgrave Macmillan. Pages
81-102.
302
Ammon News. 2011. “ áÒÊÖÏ @
www.ammonnews.net
à@ñkB@ éJQ¯”
August 13. Accessed October 21, 2019:
Amnesty International. 1991. “Tunisia: Imprisonment of Journalists.” February.
Amnesty International. 1994. “Tunisia: Rhetoric vs Reality: The Failure of a Human Rights
Bureaucracy.” January 12.
Amnesty International. 2011. “Syrian woman activist wins human rights award.” October
7. Accessed June 21, 2019: www.amnesty.org
Amnesty International. 2015. “Urgent Action: Opposition Politician Jailed for Comment.”
February 19. Accessed February 14, 2020: www.amnesty.org
Anderson, Liam. 2019. “History (Iraq).” In Europa World Online. Routledge. Accessed
April 4, 2019: http://europaworld.com
Anonymous Yemeni journalist. 2019. Personal communication. October 6.
Anzalone, Christopher. 2009. “Islamic Action Front.” In The Oxford Encyclopedia of the
Islamic World.
John L. Esposito, editor.
Accessed February 6, 2019:
www.oxfordislamicstudies.com
APANews. 2017. “Arrôt de diffusion des télévisions privées en Mauritanie.” October 17.
Accessed August 16, 2019: www.dowjones.com/products/factiva
April 6 Youth Movement.
2019.
“ ÉK QK. @
”
6 H. AJ. é»Qk
Internet Archive.
Accessed
September 24, 2019: http://web.archive.org/web/∗ /6april.org
Asharq Al-Awsat. 2017. “Algeria Appoints New Prime Minister.” May 25. Accessed April
22, 2019: www.dowjones.com/products/factiva
Associated Press. 2003. “Russian Newspaper Announces Closure Over Interference.”
February 28. Accessed February 3, 2020: www.dowjones.com/products/factiva
Associated Press. 2019. “Yemen rebel leader meets UN envoy in effort to revive peace talks.”
303
October 1. Accessed November 11, 2019: https://apnews.com
Awadat, Ibtisam. 1997. “PPD, 8 weeklies await High Court verdict on suspension order
next week.” November 20.
The Star.
Accessed August 10, 2019:
www.dowjones.com/products/factiva
Ayish, Muhammad. 2011. “Television Broadcasting in the Arab World: Political
Democratization and Cultural Revivalism.” In Arab Media. Noha Mellor, Muhammad
Ayish, Nabil Dajani, and Khalil Rinnawi, editors. Malden, MA: Polity.
Ayubi, Nazih N., Joseph A. Kéchichian, Denis J. Sullivan, Fred H. Lawson, and Marion
Boulby. 2009. “Muslim Brotherhood.” In The Oxford Encyclopedia of the Islamic World.
John L. Esposito, editor. Accessed March 28, 2019: www.oxfordislamicstudies.com
Bacon, Tricia. 2018. “Somalia’s Current Security and Stability Status.” Senate Foreign
Relations Subcommittee on Africa and Global Health Policy Hearing.
Baggaley, John. 1992a. “Algeria’s newspapers struggle for survival.” September 28. Reuters.
Accessed August 5, 2019: www.dowjones.com/products/factiva
Baggaley, John. 1992b. “Algeria’s FIS Has Record For Survival.” February 10. Reuters.
Accessed August 5, 2019: www.dowjones.com/products/factiva
Baggaley, John. 1992c. “Algeria’s FLN Appears Headed for History.” January 21. Accessed
July 31, 2019: www.dowjones.com/products/factiva
Baggaley, John. 1992d. “Five-Man Body Takes Charge of Algeria.” January 14. Accessed
November 10, 2019: www.dowjones.com/products/factiva
Baghdadi, Rima. 2012. “Kuwait Youth Emerge as Force in Protests Against the State.”
October 23. Al-Monitor. Accessed May 28, 2019: www.al-monitor.com
Bahrain Mirror. 2015. “Al-Wefaq’s Jameel Kadhem Released after Spending Six Months
Behind Bars.” July 28. Accessed February 14, 2020: http://bahrainmirror.com
Bakhsh, Sumaya. 2017. “Yemen TV presenter announces brother’s death on air.” December
22. Accessed October 1, 2019: www.bbc.com
304
Barany, Zoltan. 2011. “Comparing the Arab revolts: The role of the military.” Journal of
Democracy 22(4): 24-35.
Baron, Adam. 2016. “Yemen’s Forgotten War: How Europe Can Lay the Foundations for
Peace.”
December
20.
Accessed
December
29,
2019:
www.dowjones.com/products/factiva
Barzegar, Karine G. 2014. “Protests, controversy mark elections in Algeria.” April 17.
Gannett News Service. Accessed May 7, 2019: www.dowjones.com/products/factiva
Batty, David and Jack Shenker. 2011. “Syria suspended from Arab League.” November 12.
The Guardian. Accessed February 10, 2020: www.theguardian.com
Baum, Matthew A. and Tim Groeling. 2009. “Shot by the Messenger: Partisan Cues and
Public Opinion Regarding National Security and War.” Politicl Behavior 31:157—186.
Baykal, Erol A. F. 2019. The Ottoman Press (1908-1923). Boston: BRILL.
BBC Monitoring: Africa. 1997. “Five main opposition parties form common front.” March
3. Accessed May 23, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Africa. 2013. “Sudanese opposition parties reject offer to join
government.”
November
9.
Accessed
April
1,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Media. 2001. “Islamist weekly seized again.” April 12. Accessed August
17, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Media. 2003. “Saudi opposition Sawt al-Islah radio observed active.”
January 22. Accessed June 4, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Media. 2004. “Saudi rulers ease their grip on the media.” May 28.
Accessed June 4, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Media. 2006. “Jordan: Muslim Brotherhood’s website blocked.” August
17. Accessed August 10, 2019: www.dowjones.com/products/factiva
305
BBC Monitoring: Media. 2007. “Jordan: Ban lifted on Al-Majd publication; chief editor
cited.” May 7. Accessed August 10, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Media. 2009a. “Morocco: Websites of Islamic movement inaccessible
since
17
January.”
January
21.
Accessed
October
16,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Media. 2009b. “Yemen: Suhayl satellite television channel launches
news
website.”
October
12.
Accessed
September
29,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Media. 2009c. “Three Tunisian newspapers call one-week strike, protest
government restrictions.” November 10.
Accessed August 22,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Media. 2009d. “Jordanian independent weekly Al-Sabil begins daily
publication.”
February
10.
Accessed
August
9,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Media. 2012. “Yemeni paper profiles growing number of private TV
channels.”
March
13.
Accessed
September
29,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1990. “Demonstration by Islamic movement members in
Rabat.” May 10. Accessed June 14, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1990a. “New Sudanese rebel radio heard following
announcement by Radio SPLA.” October 2.
Accessed April 1, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1991. “Over 100,000 protestors attempt to march to
presidential
office.”
October
26.
Accessed
June
29,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1992. “FLN Political Bureau affirms need to appeal
against seizure decisions.” May 7.
Accessed August 26,
2019:
www.dowjones.com/products/factiva
306
BBC Monitoring: Middle East. 1993. “King Husayn Gives News Conference on Election
Results and Peace Process.” November 11.
Accessed August 9, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1994a. “Coeducation in South Yemen Abolished; Press
Censorship Reportedly Introduced.” October 3.
Accessed June 18, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1994b. “Shops Close in Tizi Ouzou in Response to Berber
Cultural Movement’s Strike Call.” January 27.
Accessed July 13, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1995. “Saudi Troops ‘Hoisted the Saudi Flag on Yemeni
Territory.’”
January
14.
Accessed
September
30,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1996a. “Ahmed Ould Daddah Re-elected as Head of Main
Opposition
Party.”
June
27.
Accessed
May
23,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1996b. “HAMAS and the Algerian Renewal Party Given
Portfolios in New Government.” January 8.
Accessed May 6, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1996c. “Newspapers Report Increased Opposition Activity
in the South.” June 11. Accessed June 18, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1998a. “Islamist movement denounces ‘repression’.” January
21. Accessed August 17, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1998b. “Official comments on crisis within UFD opposition
party.” March 20. Accessed May 24, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1998c. “Thousands march in Algiers in protest at Berber
singer’s death.” July 11. Accessed May 6, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 1999. “Demonstrators in Nouakchott demand release of
opposition figures.” Accessed May 20, 2019: www.dowjones.com/products/factivs
307
BBC Monitoring: Middle East. 2000a. “Opposition front lists conditions for ‘genuine’
political
détente.”
December
8.
Accessed
May
23,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2000b. “Opposition rally urges release of Socialist Party
leaders.” March 17. Accessed July 27, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2001. “Authorities said to block access to Islamic opposition
websites.” April 17. Accessed August 17, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2003. “BBC Monitoring guide to Yemeni parliamentary
elections 27 April 2003.” March 22.
Accessed August 17,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2004. “Libya: Brotherhood detainees reportedly stage
sit-in to protest trial delay.” April 8.
Accessed June 26, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2007.
‘sabotaging’ websites.” January
www.dowjones.com/products/factiva
“Opposition accuses Jordanian government of
18.
Accessed August 11,
2019:
BBC Monitoring: Middle East. 2009a. “Kuwait-based Yemeni TV shut down on official
orders.” August 26. Accessed September 29, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2009b. “Paper reports on trial broadcasting of new Yemeni
satellite
channel.”
July
15.
Accessed
September
28,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2009c. “Senegal’s foreign minister meets ousted
Mauritanian leader behind closed doors.” May 9. Accessed November 10, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2010a. “Al-Jazeera website reports on Internet monitoring
in Tunisia.” April 25. Accessed February 14, 2020: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2010b. “Sudanese opposition party criticizes new
government, to establish shadow cabinet.” June 18.
Accessed April 1, 2019:
308
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2011. “Yemeni minister raps foreign media over protest
coverage.” April 25. Accessed September 29, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2012a. “Algeria’s Islamist party votes against joining
cabinet after ‘fraudulent’ poll.” May 21.
Accessed April 22, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2013. “Mauritanian leader to continue ‘peaceful struggle’
to
oust
president.”
January
4.
Accessed
May
22,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2013a. “Jordanian Islamist TV channel resumes
transmission after five-day suspension.” September 3. Accessed August 10, 2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2013b. “Mauritanian Islamists stage protest against army
‘rule’.” February 28. Accessed July 26, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2013c. “Morocco’s Islamist party reportedly launches TV
channel.” July 28. Accessed August 17, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2014. “Algerian opposition leader says ‘the regime has no
future’,
elections ‘farce’.” March 14.
Accessed July 13,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2015a. “Mauritania suspends three TV channels from
transmitting.” July 7. Accessed August 16, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2015b. “Retired Algerian general says president’s brother
wields true power.” September 7.
Accessed December 29,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2016. “Reform debate heats up in Morocco ahead of
elections.” October 6. Accessed April 8, 2019: www.dowjones.com/products/factiva
309
BBC Monitoring: Middle East. 2017. “‘Largest’ protest since 2011 in Moroccan capital.”
June 12. Accessed October 16, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Middle East. 2018. “Victory ‘for the Mauritanian people’ — president
casting
vote.”
September
1.
Accessed
August
16,
2019:
www.dowjones.com/products/factiva
BBC Monitoring: Newsfile. 2003. “Berber towns, villages paralysed by protest on uprising
anniversary.” April 20. Accessed July 24, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Newsfile. 2011. “Algerian anti-government protest reportedly ends in
confusion.” February 12. Accessed July 25, 2019: www.dowjones.com/products/factiva
BBC Monitoring: Newsfile. 2015. “Jordan shuts down pro-Islamist Al-Yarmuk TV — Al
Jazeera.” August 31. Accessed August 10, 2019: www.dowjones.com/products/factiva
BBC Monitoring. 2014. “Huthists continue to storm, loot houses, public offices in Yemeni
capital.”
September
30.
Accessed
September
10,
2019:
www.dowjones.com/products/factiva
BBC News. 2004. “Algeria’s presidential challengers.” April 9. Accessed July 13, 2019:
http://news.bbc.co.uk
BBC News. 2006. “Bahrain pro-democracy cleric dies.” December 18. Accessed February
7, 2019: http://news.bbc.co.uk
BBC News. 2012. “Profile: Mohammed Hussein Tantawi.” August 12. Accessed November
15, 2019: www.bbc.com
BBC News. 2017a. “Algeria profile.” May 2. Accessed August 22, 2019: www.bbc.com
BBC News. 2017b.
www.bbc.com
“Somaliland profile.” December 14.
Accessed January 23, 2019:
BBC News. 2017c. “Twitter user numbers overtaken by China’s Sina Weibo.” May 17.
Accessed February 4, 2020: www.bbc.com
310
BBC News. 2017d. “Profile: Arab League.” August 24. Accessed February 18, 2020:
www.bbc.com
Beaumont, Peter. 2011. “The truth about Twitter, Facebook and the uprisings in the Arab
world.” February 25. The Guardian. Accessed February 4, 2020: www.theguardian.com
Beck, Nathaniel. 2020. “Estimating Grouped Data Models with a Binary-Dependent
Variable and Fixed Effects via a Logit versus a Linear Probability Model: The Impact
of Dropped Units.” Political Analysis 28:139—145.
Bekaye, Sheikh. 2003. “Mauritania Bans Anti-U.S. Protests.” January 29. Associated Press.
Accessed May 21, 2019: www.dowjones.com/products/factiva
Bellin, Eva. 1995. “Civil Society in Formation: Tunisia.” In Civil Society in the Middle
East. Richard Augustus Norton, editor. Leiden: Brill. 120-147.
Bellin, Eva. 2012. “Reconsidering the Robustness of Authoritarianism in the Middle East:
Lessons from the Arab Spring.” Comparative Politics 44(2): 127-149.
Benhaida, Sarah. 2014. “‘Fearless’ Egypt Islamists rally in defiance of ban.” January 3.
Agence France Presse. Accessed March 28, 2019: www.dowjones.com/products/factiva
Benmehdi, Hassan. 2013. “Istiqlal Quits Morocco Government.” July 11. All Africa.
Accessed April 7, 2019: www.dowjones.com/products/factiva
Berenger, Ralph. 2006. “Media in the Middle East and North Africa.” In Global
Communication: Theories, Stakeholders, and Trends. Second edition. Thomas L.
McPhail, editor. Malden, MA:Â Blackwell. 192-225.
Berry, LaVerle Bennette. 2015. “Sudan, a Country Study.” Library of Congress. Accessed
April 1, 2019: www.loc.gov
Besley, Timothy and Andrea Prat. 2006. “Handcuffs for the Grabbing Hand? Media Capture
and Government Accountability.” American Economic Review 96(3): 720-736.
Bonnefoy, Laurent. 2007. “Yemeni Congregation for Reform.” In Oxford Islamic Studies
Online. John L. Esposito, M. A. Abdel Haleem, A.J. Arberry, and Hanna E. Kassis,
311
editors. Accessed March 22, 2019: www.oxfordislamicstudies.com
Bouasria, Abdelilah. 2015. Sufism and Politics in Morocco: Activism and Dissent. New
York: Routledge.
Boucek, Christopher. 2009. “Security over democracy: how the West views Mauritania.”
May 29. Daily Star. Accessed November 10, 2019: www.dowjones.com/products/factiva
Boulby, Marion. 1988. “The Islamic challenge: Tunisia since independence.” Third World
Quarterly 10(2): 590-614.
Boum, Aomar and Thomas K. Park. 2016. Historical Dictionary of Morocco. Third Edition.
New York: Rowman & Littlefield.
Bouzerda, Ali. 1995. “Morocco to free Islamist leader from house arrest.” October 2.
Reuters. Accessed April 4, 2019: www.dowjones.com/products/factiva
Bouzerda, Ali. 2000a. “Moroccan Islamist says to end 10-year-house arrest.” May 15.
Reuters. Accessed October 6, 2019: www.dowjones.com/products/factiva
Bouzerda, Ali. 2000X. “Morocco’s Islamist chief free, group still shackled.” May 20. Reuters.
Accessed October 6, 2019: www.dowjones.com/products/factiva
Bove, Vincent, Jean-Philippe Platteau, and Petros G. Sekeris. 2017. “Political repression in
autocratic regimes.” Journal of Comparative Economics 45: 410-428.
BP. 2019. BP Statistical Review of World Energy 2019. 68th edition. Accessed November
28, 2019: www.bp.com
Brancati, Dawn. 2014. “Democratic Authoritarianism: Origins and Effects.” Annual Review
of Political Science 17: 313-326.
Bratton, Michael, and Nicolas van de Walle. 1997. Democratic Experiments in Africa:
Regime Transitions in Comparative Perspective. New York: Cambridge University Press.
Brennan, David. 2018. “Democracy in Egypt: Five Years After the Military Coup,
Opposition Has Been Silenced.” July 7. Newsweek. Accessed March 28, 2019:
312
www.newsweek.com
Brinkley, Joel. 1991. “Jordanian King Names Palestinian Prime Minister.” June 19. New
York Times. Accessed March 22, 2019: www.nytimes.com
Browers, Michaelle. 2007. “Origins and Architects of Yemen’s Joint Meeting Parties.”
International Journal of Middle East Studies 39(4): 565-586.
Brown, Nathan J. 2012. When Victory is Not an Option. Ithaca, NY: Cornell University
Press.
Brown, Nathan J. 2019. Professor of Political Science and International Affairs, Georgetown
University. Personal communication. October 7.
Brown, Nathan J., Amr Hamzawy, and Marina Ottaway. 2006. “Islamist Movements and
the Democratic Process in the Arab World: Exploring the Gray Zones.” 2006. Carnegie
Endowment for International Peace and Herbert-Quandt-Stiftung. Accessed February
16, 2020: https://carnegieendowment.org
Brownlee, Jason, Tarek Masoud, and Andrew Reynolds. 2015. The Arab Spring: Pathways
of Repression and Reform. New York: Oxford University Press.
Brüderl, Josef and Volker Ludwig. 2014. “Fixed-effects panel regression.” In The SAGE
Handbook of Regression Analysis and Causal Inference. Henning Best and Christof
Wolf, editors. Thousand Oaks, CA: SAGE Publications.
Bueno de Mesquita, Bruce and Alastair Smith. 2010. “Leader Survival, Revolutions, and the
Nature of Government Finance.” American Journal of Political Science 54(4): 936—950.
Bueno de Mesquita, Ethan. 2010. “Regime Change and Revolutionary Entrepreneurs.”
American Political Science Review 104(3): 446-466.
Burgat, François and William Dowell. 1997. The Islamic Movement in North Africa.
Second edition. Austin, TX: Center for Middle Eastern Studies, University of Texas at
Austin.
Cameron, A. Colin and Douglas L. Miller. 2015. “A Practitioner’s Guide to Cluster-Robust
313
Inference.” Journal of Human Resources 50(2): 317-372.
Cameron, Peter D. and Michael C. Stanley. 2017. Oil, Gas, and Mining: A Sourcebook
for Understanding the Extractive Industries.” Washington, DC: International Bank for
Reconstruction and Development / The World Bank.
Campagna, Joel. 1998a. “Overview of the Middle East and North Africa.” Attacks on
the Press in 1997. Committee to Protect Journalists. Accessed September 21, 2019:
https://cpj.org
Campagna, Joel. 1998b. “Jordan Reins in the Press.” Attacks on the Press in 1997.
Committee to Protect Journalists. Accessed October 19, 2019: https://cpj.org
Campagna, Joel. 1999. “Siege Mentality: Press Freedom and the Algerian Conflict.” March.
Committee to Protect Journalists. Accessed August 5, 2019: https://cpj.org
Campagna, Joel. 2009. “Pre-empting the Satellite TV Revolution.” February 10. Committee
to Protect Journalists. Accessed February 14, 2020: https://cpj.org
Caren, Neal, Sarah Gaby, and Catherine Herrold.
Collective Action.” Social Problems 64: 133-155.
2017.
“Economic Breakdown and
Carnegie Endowment for International Peace. 2012. “Constitution Party.” October 31.
Accessed June 25, 2019: https://carnegieendowment.org
CDAC Network. 2012. “Yemen Media and Telecoms Landscape Guide.” February. Accessed
October 2, 2019: www.cdacnetwork.org
Charlton, Angela. 2007. “Moroccans Go With Status Quo in Election.” September 9.
Associated Press. Accessed April 7, 2019: www.dowjones.com/products/factiva
Chen, Jidong and Yiqing Xu. 2017. “Why Do Authoritarian Regimes Allow Citizens to
Voice Opinions Publicly?” Journal of Politics 79(3): 792-803.
Cho, Joan E., Jae Seung Lee, and B. K. Song. 2017. “Media Exposure and Regime Support
under Competitive Authoritarianism: Evidence from South Korea.” Journal of East
Asian
314
Chouikha, Larbi, Kamel Labidi, and Hassen Jouini. 1992. “Etat de la Liberté de la Presse
en Tunisie de Janvier 1990 à Mai 1991.” In L’Information au Maghreb. Pierre Albert et
al, editors. Tunis: Cérès. 94-119.
Chtatou, Mohamed. 2019. “The State of Amazigh Culture in Algeria and Morocco.” January
31. International Policy Digest. Accessed July 13, 2019: https://intpolicydigest.org
Chwe, Michael Suk-Young. 2001. Rational Ritual: Culture, Coordination, and Common
Knowledge. Princeton, NJ: Princeton University Press.
Cigar, Norman. 1990. “Islam and the State in South Yemen: The Uneasy Coexistence.”
Middle Eastern Studies 26(2): 185-203.
Clark, Justin, Robert Faris, Ryan Morrison-Westphal, Helmi Noman, Casey Tilton, and
Jonathan Zittrain. 2017. “The Shifting Landscape of Global Internet Censorship.”
Berkman Klein Center for Internet & Society Research Publication. Accessed February
4, 2020: http://nrs.harvard.edu
Clayton, Jonathan. 1987. “Tunisian Prime Minister Promises Multi-Party State.” November
10. Reuters News. Accessed March 11, 2020: www.dowjones.com/products/factiva
Coates Ulrichsen, Kristian. 2019. “History (The United Arab Emirates)” in Europa World.
London: Routledge. Accessed January 21, 2019: http://europaworld.com
Cochrane, Richard. 2017. “Algerian president’s demonstrable infirmity likely to accelerate
contest for a successor ahead of 2019 election.” March 30. Accessed December 29, 2019:
www.dowjones.com/products/factiva
Coll, Steve. 2011. “The Casbah Coalition.” April 4. The New Yorker. Accessed January 5,
2020: www.dowjones.com/products/factiva
Commins, David and David W. Lesch. 2014. Historical Dictionary of Syria. Third Edition.
Lanham, MD: Scarecrow.
Committee to Protect Journalists. 1996. “Attacks on the Press in 1995.” Accessed September
21, 2019: https://cpj.org
315
Committee to Protect Journalists. 1997. “Attacks on the Press in 1996.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 1998. “Attacks on the Press in 1997.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 1999. “Attacks on the Press in 1998.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 2000. “Attacks on the Press in 1999.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 2002. “Attacks on the Press in 2001.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 2003. “Attacks on the Press in 2002.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 2005. “Attacks on the Press in 2004.” Accessed September
21, 2019: https://cpj.org
Committee to Protect Journalists. 2006. “Attacks on the Press in 2005.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2007. “Attacks on the Press in 2006.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2008. “Attacks on the Press in 2007.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2009. “Attacks on the Press in 2008.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2010. “Attacks on the Press in 2009.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2011a. “Yemen shells TV station, news agency, online
316
newspaper.” May 26. Accessed September 30, 2019: https://cpj.org
Committee to Protect Journalists. 2011b.
September 27, 2019: https://cpj.org
“Attacks on the Press in 2010.” Accessed
Committee to Protect Journalists. 2011c. “Egypt: new accreditation rules; military
obstructs media.” February 7. Accessed September 23, 2019: https://cpj.org
Committee to Protect Journalists. 2012. “Attacks on the Press in 2011.” Accessed September
27, 2019: https://cpj.org
Committee to Protect Journalists. 2013a. “At least three journalists abducted in Yemen.”
May 22. Accessed September 30, 2019: https://cpj.org
Committee to Protect Journalists. 2013b.
September 27, 2019: https://cpj.org
“Attacks on the Press in 2012.” Accessed
Committee to Protect Journalists. 2014a.
September 27, 2019: https://cpj.org
“Attacks on the Press in 2013.” Accessed
Committee to Protect Journalists. 2014b. “Egypt jails two journalists, renews detention of
another.” April 30. Accessed September 27, 2019: https://cpj.org
Committee to Protect Journalists. 2014c. “Yemeni journalist shot dead in series of attacks
on state-run media.” August 19. Accessed October 3, 2019: https://cpj.org
Committee to Protect Journalists. 2015. “Media outlets raided and banned as conflict spirals
in Yemen.” March 27. Accessed September 30, 2019: https://cpj.org
Committee to Protect Journalists. 2019. “Committee to Protect Journalists.” Accessed
September 20, 2019: https://cpj.org
Confidentiel Afrique.
2018.
“Mauritanie: L’égérie Naha Mint Mouknass,nouvelle
anti-soufre douleur de la politique d’Ould Abdel Aziz” . June 13. Accessed May 21,
2019:
http://confidentielafrique.com/gouvernement/mauritanie-legerie-naha-mintmouknassnouvelle-anti-soufre-douleur-de-politique-dould-abdel-aziz/
317
Coonan, Clifford. 2013. “Chinese journalists go on strike over government censorship of
Southern Weekend newspaper.” January 7. The Independent. Accessed February 4,
2020: www.independent.co.uk
Cotet, Anca M. and Kevin K.Tsui. 2013. “Oil and Conflict: What Does the Cross Country
Evidence Really Show?” American Economic Journal: Macroeconomics 5(1): 49-80.
Cousin, Eduardo. 2014. “Anti-Coup Alliance condemns military trial verdict for five
civilians.” September 10.
Daily News Egypt.
Accessed June 25, 2019:
www.dowjones.com/products/factiva
. JºÖÏ @
Cover SD. 2019. “ HAJ
Hñ
”
ú¯ éÓB@
[The Voice of the Nation in Libraries] July 5.
Accessed November 16, 2019: www.cover-sd.com
Cowell, Alan. 1990. “Tunis at Impasse on Political Change.” June 17. New York Times.
Accessed March 10, 2020: www.dowjones.com/products/factiva
Cowen, Tyler and Daniel Sutter. 1998. “Why only Nixon could go to China.” Public Choice
97: 605-615.
CQ Press. 2006. Political Handbook of the Middle East 2006. Washington, DC.
Crystal, Jill. 2014. “Eastern Arabian States: Kuwait, Bahrain, Qatar, United Arab
Emirates, and Oman.” In The Government and Politics of the Middle East and North
Africa. Seventh edition. Mark Gasiorowski, editor. San Francisco: Westview. Pages
157-195.
Crystal, Jill. 2019. “History (Kuwait)” in Europa World. London: Routledge. Accessed
January 21, 2019: http://www.europaworld.com
Cudsi, Alexander S. and John O. Voll.
2009.
“Ummah-Ansar.” In The Oxford
Encyclopedia of the Islamic World. John L. Esposito, editor. Accessed March 31, 2019:
www.oxfordislamicstudies.com
Cukierman, Alex and Mariano Tommasi. 1998. “When Does It Take a Nixon to Go to
China?” American Economic Review 88: 180-197.
318
Daadaoui, Mohamed. 2010. “Rituals of Power and Political Parties in Morocco: Limited
Elections as Positional Strategies.” Middle Eastern Studies 46(2): 195-219.
Daily News Egypt. 2011. “Downtown march calls for end to military trials for civilians.”
July 21. Accessed July 27, 2019: www.dowjones.com/products/factiva
Daily Star. 2011. “Libyan rebels enter Tripoli, crowds celebrate.” August 22. Accessed
September 10, 2019: www.dowjones.com/products/factiva
Daily Tribune. 2013. “Morsi Backers Stage Rallies as Crackdown Looms.” August 10.
Accessed July 27, 2019: www.dowjones.com/products/factiva
Damis, John. 1972. “The Moroccan Political Scene.” Middle East Journal 26(1): 25-36.
Danish Immigration Service. 2017. “South and Central Somalia: Security Situation, alShabaab Presence, and Target Groups.”
Dassin, Joan R. 1979. “Press censorship-how and why.” Index on Censorship 4: 13-19.
Dazi-Heni, Fatiha. 2015. “The Arab Spring Impact on Kuwait ‘Exceptionalism’.” Arabian
Humanities (4). Accessed July 12, 2019: https://journals.openedition.org
de Bendern, Paul. 2004. “Algerian leader seen picking reform government.” April 13.
Reuters. Accessed April 22, 2019: www.dowjones.com/products/factiva
Deckert, Roman. 2012. “The history of the Sudanese press: Background to conflict.” In “The
Sudanese press after separation — Contested identities of journalism.” Tammi Coles, Jess
Smee, and Dirk Spilker, editors. Berlin: Media in Cooperation and Transition.
DeNardo, James. 1985. Power in Numbers: The Political Strategy of Protest and Rebellion.
Princeton, NJ: Princeton University Press.
Derpanopoulos, George, Erica Frantz, Barbara Geddes, and Joseph Wright. 2017. “Are
coups good for democracy? A response to Miller (2016)” Research and Politics 4(2):
1—4.
Deutsche Welle. 2011. “New interim president takes oath in Tunisia.” January 15. Accessed
319
November 11, 2019: www.dowjones.com/products/factiva
Devitt, Polina and Denis Pinchuk. 2018. “Russian opposition leader Navalny’s website
blocked before election.” February 15.
Reuters.
Accessed February 4, 2020:
www.reuters.com
Dick, Charles. 1988. “Tunisian Ruling Party Sets New Course after Bourguiba Era.” July
31. Accessed March 11, 2020: www.dowjones.com/products/factiva
Diwan. 2012. “Syria in Crisis: National Coordination Body for Democratic Change.”
January 15.
Carnegie Middle East Center.
Accessed June 21, 2019:
https://carnegie-mec.org/diwan
Dodge, Toby. 2013. “State and society in Iraq ten years after regime change: the rise of a
new authoritarianism.” International Affairs, 89(2): 241—257.
Donnadieu, Jean-Louis. 1992. “Tunisia: freedom to discriminate.” Index on Censorship
21(1): 39.
Dougherty, Beth K. and Edmund A. Ghareeb. 2013. Historical Dictionary of Iraq. Second
Edition. Lanham, MD: Scarecrow.
Dow Jones International News. 2003. “Saudi Opposition Group: Government Jammed
Radio
Stations.”
October
23.
Accessed
July
30,
2019:
www.dowjones.com/products/factiva
Drost, Harry. 1991. The World’s News Media: A Comprehensive Reference Guide. Essex,
United Kingdom: Longman Group.
Drummond, James. 2001. “Berber protests open second front for Algeria’s military
regime.” June 11.
Financial Times.
Accessed February 16, 2020:
www.dowjones.com/products/factiva
Dubai Press Club. 2012. “Arab Media Outlook: 2011-2015.”
Dunne, Michelle and Katie Bentivoglio. 2014. “Egypt’s Student Protests: The Beginning
or the End of Youth Dissent?” October 22. Diwan blog. Carnegie Endowment for
320
International Peace. Accessed June 25, 2019: https://carnegie-mec.org/diwan
Durac, Vincent. 2011. “The Joint Meeting Parties and the Politics of Opposition in Yemen.”
British Journal of Middle Eastern Studies 38(3): 343-365.
Durac, Vincent. 2019. “History (Yemen).” In Europa World Online. Routledge. Accessed
March 16, 2019: http://europaworld.com
Economist Intelligence Unit. 1996a. “Country Profile: Egypt.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 1996b. “Country Profile: Morocco.” Accessed February 11,
2019: www.eiu.com
Economist Intelligence Unit. 1996c. “Country Profile: Sudan.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 1996d. “Country Profile: Algeria.” Accessed February 11,
2019: www.eiu.com
Economist Intelligence Unit. 1996e. “Country Profile: Jordan.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 1997. “Country Report: Senegal, The Gambia, Mauritania.”
1st Quarter. Accessed February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2000. “Country Report: Bahrain.” 1st Quarter. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2001. “Country Report: Yemen.” February. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2002a. “Country Profile: Bahrain.” Accessed February 11,
2019: www.eiu.com
Economist Intelligence Unit. 2002b. “Country Profile: Sudan.” Accessed February 11, 2019:
www.eiu.com
321
Economist Intelligence Unit. 2002c. “Country Profile: Yemen.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2002d. “Country Profile: Algeria.” Accessed February 11,
2019: www.eiu.com
Economist Intelligence Unit. 2002e. “Country Profile: Jordan.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2006. “Country Report: Bahrain.” March. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2008a. “Country Profile: Morocco.” Accessed February 11,
2019: www.eiu.com
Economist Intelligence Unit. 2008b. “Country Profile: Yemen.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2008c. “Country Profile: Egypt.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2008d. “Country Profile: Jordan.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2009a. “Country Profile: Sudan.” Accessed February 11, 2019:
www.eiu.com
Economist Intelligence Unit. 2009b. “Mauritania politics: In with the Ould.” July 21.
Accessed July 26, 2019: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2010a. “Country Report: Morocco.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2010b. “Country Report: Egypt.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit.
2011a.
“Country Report: Tunisia.” February.
322
Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2011b.
February 11, 2019: www.eiu.com
“Country Report: Jordan.” January.
Accessed
Economist Intelligence Unit. 2011c.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2011d. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2011e. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2012a. “Algeria politics: New cabinet will disappoint
reformers.”
September
12.
Accessed
May
11,
2019:
www.dowjones.com/products/factiva
Economist Intelligence Unit. 2012b.
February 11, 2019: www.eiu.com
“Country Report: Jordan.” January.
Accessed
Economist Intelligence Unit. 2012c.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2012d. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2012e. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2013a. “Country Report: Lebanon.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2013b. “Country Report: Lebanon.” July. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2013c. “Country Report: Jordan.” January. Accessed February
323
11, 2019: www.eiu.com
Economist Intelligence Unit. 2013d.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2013e. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2013f. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2014a. “Country Report: Egypt.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2014b. “Egypt Politics: Struggle looms over control of the
security
state.”
April
29.
Accessed
December
17,
2019:
www.dowjones.com/products/factiva
Economist Intelligence Unit. 2014c. “Country Report: Jordan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2014d.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2014e. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2014f. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2015a. “Algeria politics: The bumpy road to shale gas
production.” March 27. Accessed May 8, 2019: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2015b. “Algeria: Key figures.” June 24. Accessed December
16, 2019: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2015c. “Algeria politics: Clipping the wings of the DRS.”
324
October 8. Accessed January 7, 2020: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2015d.
February 11, 2019: www.eiu.com
“Country Report: Jordan.” January.
Accessed
Economist Intelligence Unit. 2015e.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2015f. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2015g. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2016a. “Country Report: Morocco.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2016b. “Algeria politics: Who is in charge?” February 6.
Accessed December 16, 2019: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2016c.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2016d. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2016e. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2016f. “Country Report: Jordan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2017a. “Algeria: Political forces at a glance.” September 26.
Accessed April 22, 2019: www.dowjones.com/products/factiva
Economist Intelligence Unit. 2017b. “Country Report: Morocco.” January. Accessed
February 11, 2019: www.eiu.com
325
Economist Intelligence Unit. 2017c. “Country Report: Jordan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2017d.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2017e. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2017f. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2018a.
February 11, 2019: www.eiu.com
“Country Report: Algeria.” January.
Accessed
Economist Intelligence Unit. 2018b.
February 11, 2019: www.eiu.com
“Country Report: Bahrain.” January.
Accessed
Economist Intelligence Unit. 2018c. “Country Report: Jordan.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2018d.
February 11, 2019: www.eiu.com
“Country Report: Kuwait.” January.
Accessed
Economist Intelligence Unit. 2018e. “Country Report: Mauritania.” January. Accessed
February 11, 2019: www.eiu.com
Economist Intelligence Unit. 2018f. “Country Report: Egypt.” January. Accessed February
11, 2019: www.eiu.com
Economist Intelligence Unit. 2018g. “Country Report: Sudan.” January. Accessed February
11, 2019: www.eiu.com
Economist, The. 2012. “Egypt’s president: New broom in action: The senior generals who
once ran the country are booted out of office.” August 18. Accessed January 27, 2019:
https://www.economist.com/middle-east-and-africa/2012/08/18/new-broom-in-action
326
Economist, The. 2013. “From president to prisoner: Egypt puts yet another former president
on trial.” November 4. Accessed March 28, 2019: www.economist.com
Economist, The. 2014. “Al-Sisi Ascendant: The general has a good first 100 days—at the
cost of political freedom.” September 20. Accessed March 28, 2019: www.economist.com
Editors of Encyclopedia Britannica. 2019a. “Abdel Fattah al-Sisi.” Accessed November 15,
2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019b. “Abdelaziz Bouteflika.” Accessed November 15,
2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019c. “Abdullah II.” Accessed November 15, 2019:
www.britannica.com
Editors of Encyclopedia Britannica. 2019d. “Ali Abdullah Saleh.” Accessed November 15,
2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019e. “Hassan II.” Accessed November 15, 2019:
www.britannica.com
Editors of Encyclopedia Britannica. 2019f. “Hosni Mubarak.” Accessed November 15, 2019:
www.britannica.com
Editors of Encyclopedia Britannica.
www.britannica.com
2019g.
“Hussein.” Accessed November 15, 2019:
Editors of Encyclopedia Britannica. 2019h. “Liamine Zeroual.” Accessed November 15,
2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019i. “Omar al-Bashir.” Accessed November 15, 2019:
www.britannica.com
Editors of Encyclopedia Britannica. 2019j. “Sheikh Hamad ibn ‘Isa Al Khalifah.” Accessed
November 15, 2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019k. “Sheikh Jabir al-Ahmad al-Jabir al-Sabah.”
327
Accessed November 15, 2019: www.britannica.com
Editors of Encyclopedia Britannica. 2019l. “Zine al-Abidine Ben Ali.” Accessed November
15, 2019: www.britannica.com
Egorov, Georgy, Sergei Guriev, and Konstantin Sonin. 2009. “Why Resource-poor
Dictators Allow Freer Media: A Theory and Evidence from Panel Data.” American
Political Science Review 103(4): 645-668.
El Gody, Ahmed. 2007. “New Media, New Audience, New Topics, and New Forms of
Censorship in the Middle East.” In New Media and the New Middle East. Philip Seib,
editor. New York: Palgrave Macmillan.
El Masdar. 2019. “ éK PAJ.kB @
PYÖÏ @” [News Source] Internet Archive.
Accessed September
27, 2019: http://web.archive.org/web/2018∗ /http://elmasdr.com
El Mourabiton. 2019a. “ àñ¢. @QÖÏ @” [El Mourabiton (The Almoravids)] Internet Archive.
Accessed September 21, 2019: http://web.archive.org/web/∗ /elmourabiton.tv
El Mourabiton. 2019b. “ àñ¢. @QÖÏ @” [El Mourabiton (The Almoravids)] Accessed September
21, 2019: http://elmourabiton.tv
El Sharnoubi, Osman. 2014. “Who will Egypt’s political groups vote for? A guide.” May
16. Al-Ahram Gate. Accessed June 25, 2019: www.dowjones.com/products/factiva
El Watan. 2014. “Une place baptisée par le MCB.” May 20. Accessed July 13, 2019:
www.dowjones.com/products/factiva
El-Battahani, Atta. 2016. “The Sudan Armed Forces and Prospects of Change.” April. Chr.
Michelsen Institute. Accessed January 7, 2020: www.cmi.no
El-Fekki, Amira. 2014. “Al-Sisi appoints new Director of Intelligence Agency.” December 21.
Daily News Egypt. Accessed December 17, 2019: www.dowjones.com/products/factiva
El-Naggar, Mona. 2015. “Urgin unity, rebels seek legitimacy in Yemen.” January 30. New
York Times. Accessed November 11, 2019: www.dowjones.com/products/factiva
328
El-Shamayleh, Nisreen. 2013. “Jordan’s Muslim Brotherhood goes live-to-air.” April 7.
Accessed August 10, 2019: www.aljazeera.com
El-Sherif, Ashraf. 2014. “The Muslim Brotherhood and the Future of Political Islam in
Egypt.” Carnegie Endowment for International Peace. Accessed April 3, 2019:
https://carnegieendowment.org
El-Taher, Mohammad, Hassan Al-Azhary, Sarah Mohsen, Leonid Evdokimov, and Maria
Xynou. 2018. “The State of Internet Censorship in Egypt.” July 2. Open Observatory of
Network Interference and Association for Freedom of Thought and Expression. Accessed
September 26, 2019: https://ooni.torproject.org
Elawad, Saifeldin Hassan. 2019a. Professor of Communication, Omdurman Islamic
University. Personal communication.
Elawad, Saifeldin Hassan. 2019b. “ à@XñË@
ú¯ éJ K. QmÌ '@ é¯AjË@” [The Partisan Press in Sudan]
ÈAB@ ÐñÊ« éÊm .× (Journal of Communication Sciences) 5(7).
Elbagir, Nima and Faith Karimi. 2012. “Hundreds teargassed amid clampdown on Sudan
protests.” June 26.
CNN Wire.
Accessed April 1,
2019:
www.dowjones.com/products/factiva
Eljak, Monim. 2014. “Sudan’s political opposition unites under new call for democracy.”
December
11.
The
Guardian.
Accessed
April
1,
2019:
www.dowjones.com/products/factiva
Energy Compass. 2011. “Bahrain: Crackdown Politics.” March 25. Accessed August 8,
2019: www.dowjones.com/products/factiva
Energy Compass. 2012. “UAE: Abu Dhabi Battens Down Hatches Against Islamist Threat.”
July 27. Accessed March 23, 2019: www.dowjones.com/products/factiva
Enough. 2019. “ ú¯A»” [Enough] Accessed October 4, 2019: http://kafiq8.blogspot.com
Entelis, John P. and Lisa Arone. 1995. “Democratic and Popular Republic of Algeria.” In
The Government and Politics of the Middle East and North Africa. David E. Long and
Bernard Reich, editors. Boulder, CO: Westview. 394-422.
329
Erdbrink, Thomas. 2011. “Libyan rebels come under attack near center of Tripoli.” August
23. Washington Post. Accessed June 19, 2019: www.dowjones.com/products/factiva
Erdle, Steffen. 2010. Ben Ali’s ‘New Tunisia’ (1987-2009): A Case Study of Authoritarian
Modernization in the Arab World. Berlin: Klaus Schwarz Verlag.
Errazzouki, Samia. 2017. “Led by Islamists, thousands of Moroccans rally in support of
northern
protests.”
Reuters.
Accessed
October
16,
2019:
www.dowjones.com/products/factiva
Ersan, Inal. 1996. “Kuwaitis slowly shed post-war gloom, seek fun.” January 22. Reuters
News. Accessed August 11, 2019: www.dowjones.com/products/factiva
Europa Publications. 1989. The Middle East and North Africa: 35th edition. New York:
Taylor & Francis.
Europa Publications. 1994. The Middle East and North Africa 1995: 41st edition. New
York: Taylor & Francis.
Europa Publications. 1995. Africa South of the Sahara: 24th edition. New York: Taylor &
Francis.
Europa Publications. 1999. The Middle East and North Africa: 45th edition. New York:
Taylor & Francis.
Europa Publications. 2005. Africa South of the Sahara: 34th edition. New York: Taylor &
Francis.
Europa Publications. 2013. The Middle East and North Africa: 59th edition. New York:
Taylor & Francis.
Europa World. 2019a. “Contemporary Political History (Egypt).” In Europa World Online.
Routledge. Accessed November 11, 2019: http://europaworld.com
Europa World. 2019b. “Contemporary Political History (Morocco).” In Europa World
Online. Routledge. Accessed November 10, 2019: http://europaworld.com
330
Europa World. 2019c. “Contemporary Political History (Qatar).” In Europa World Online.
Routledge. Accessed March 25, 2019: http://europaworld.com
Europa World. 2019d. “Contemporary Political History (Sudan).” In Europa World Online.
Routledge. Accessed November 11, 2019: http://europaworld.com
Europa World. 2019e. “Contemporary Political History (Tunisia).” In Europa World Online.
Routledge. Accessed November 11, 2019: http://europaworld.com
Europa World. 2019f. “Political Organizations (Jordan).” In Europa World Online.
Routledge. Accessed October 20, 2019: http://europaworld.com
Europa World. 2019g. “Political Organizations (Kuwait).” In Europa World Online.
Routledge. Accessed October 4, 2019: http://europaworld.com
Europa World. 2019h. “Political Organizations (Morocco).” In Europa World Online.
Routledge. Accessed October 11, 2019: http://europaworld.com
Europa World. 2019i. “Political Organizations (Sudan).” In Europa World Online.
Routledge. Accessed October 17, 2019: http://europaworld.com
Europa World. 2019j. “Political Organizations (Tunisia).” In Europa World Online.
Routledge. Accessed October 21, 2019: http://europaworld.com
Europa World. 2019k. “Political Organizations (Yemen).” In Europa World Online.
Routledge. Accessed September 28, 2019: http://europaworld.com
Europa World. 2019l. “Society and Media (Algeria).” In Europa World Online. Routledge.
Accessed September 10, 2019: http://europaworld.com
Europa World. 2019m. “Society and Media (Bahrain).” In Europa World Online. Routledge.
Accessed November 7, 2019: http://europaworld.com
Europa World. 2019n. “Society and Media (Egypt).” In Europa World Online. Routledge.
Accessed September 27, 2019: http://europaworld.com
Europa World.
2019o.
“Society and Media (Mauritania).” In Europa World Online.
331
Routledge. Accessed September 20, 2019: http://europaworld.com
Europa World. 2019p. “Society and Media (Morocco).” In Europa World Online. Routledge.
Accessed October 10, 2019: http://europaworld.com
Europa World. 2019q. “Society and Media (Tunisia).” In Europa World Online. Routledge.
Accessed October 21, 2019: http://europaworld.com
Europa World. 2019r. “Society and Media (Yemen).” In Europa World Online. Routledge.
Accessed September 28, 2019: http://europaworld.com
Europa World. 2019s. “Society and Media (Jordan).” In Europa World Online. Routledge.
Accessed October 20, 2019: http://europaworld.com
Europa World. 2019t. “Contemporary Political History (Algeria).” In Europa World Online.
Routledge. Accessed December 29, 2019: http://europaworld.com
Europa World. 2020. “Contemporary Political History (Yemen).” In Europa World Online.
Routledge. Accessed January 5, 2020: http://europaworld.com
European Council on Foreign Relations. 2019. “Mapping the Yemen Conflict.” Accessed
June 18, 2019: https://www.ecfr.eu
Evans, Martin and John Phillips. 2007. Algeria: Anger of the Dispossessed. New Haven,
CT: Yale University Press.
Fakihani, Abdelfettah and Dominique Pettit. 2003. “Casablanca bombings throw spotlight
on Morocco’s Islamic militants.” May 19. Agence France Presse. Accessed April 7, 2019:
www.dowjones.com/products/factiva
Fanack. 2017. “Yemen’s Media Landscape: An Overview.” March 2. Updated September
11, 2019. Accessed: October 1, 2019: https://fanack.com
Fargues, Phillipe. 2015. “Immigration vs. Population in the Gulf.” In “The Gulf
Monarchies Beyond the Arab Spring: Changes and Challenges.” European University
Institute. Accessed February 4, 2020: https://cadmus.eui.eu
332
Fattah, Hassan M. 2006. “Ailing Emir of Kuwait Steps Down, Ending a Succession Crisis.”
January 24.
New York Times.
Accessed November 15,
2019:
www.dowjones.com/products/factiva
Faucon, Benoit. 2011. “Algerians Rally in Capital, Defying Protest Ban.” February 12. Dow
Jones News Service. Accessed July 25, 2019: www.dowjones.com/products/factiva
Financial Times. 1993. “Cabinet shuffle in Algeria.” February 4. Accessed May 13, 2019:
www.dowjones.com/products/factiva
Financial Times. 1995. “Hassan Walks a Fine Line with Islamists.” June 2. Accessed March
29, 2019: www.dowjones.com/products/factiva
Financial Times. 1997a. “Moroccans Learn Algeria’s Lessons.” March 13. Accessed May 6,
2019: www.dowjones.com/products/factiva
Financial Times. 1997b. “Trappings of Democracy.” June 5. Accessed March 29, 2019:
www.dowjones.com/products/factiva
Finn, Tom. 2019. “Abdel-Malek al-Houthi: from shadow rebel leader to kingmaker.”
January 29. Middle East Eye. Accessed November 15, 2019: www.middleeasteye.net
ForeignAffairs.co.nz. 2014. “Yemen — Houthi rebels still harassing media and journalists
in
Sanaa.”
September
25.
Accessed
September
29,
2019:
www.dowjones.com/products/factiva
Foster, Noel. 2011. Mauritania: The Struggle for Democracy. Boulder, CO: First Forum.
Fouad, Ashraf. 1999. “Kuwait urged to form representative government.” July 5. Accessed
August 11, 2019: www.dowjones.com/products/factiva
Fouad, Ashraf. 2001. “U.S. to send 2,000 more troops to Kuwait.” November 17. Accessed
April 4, 2019: www.dowjones.com/products/factiva
France 24. 2011. “Ahmed Néjib Chebbi, l’homme qui cherche à rassembler l’opposition.”
January 14. Accessed June 11, 2019: www.france24.com
333
France 24. 2019. “Egypt: the Muslim Brotherhood in 10 dates.” June 17. Accessed February
18, 2020: www.france24.com
Freedom and Justice Party.
2019a.
“ éË@YªË@ð
éK QmÌ '@ éK . @ñK.”
Internet Archive.
Accessed
September 25, 2019: http://web.archive.org/web/∗ /http://www.fj-p.com
Freedom and Justice Party. 2019b. “Freedom and Justice Party.” Internet Archive. Accessed
September 27, 2019: http://web.archive.org/web/∗ /fjponline.com
Freedom House. 2002. Freedom of the Press 2002.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2003. Freedom of the Press 2003.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2004. Freedom of the Press 2004.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2005. Freedom of the Press 2005.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2006. Freedom of the Press 2006.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2007. Freedom of the Press 2007.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2008. Freedom of the Press 2008.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2009a. Freedom of the Press 2009.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2009b. Freedom on the Net 2009.
https://freedomhouse.org
Accessed September 26, 2019:
334
Freedom House. 2010a. Freedom in the World 2010.
https://freedomhouse.org
Accessed November 10, 2019:
Freedom House. 2010b. Freedom of the Press 2010.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2011a. Freedom of the Press 2011.
https://freedomhouse.org
Accessed September 20, 2019:
Freedom House. 2011b. Freedom on the Net 2011.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2012a. Freedom of the Press 2012.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House. 2012b. Freedom on the Net 2012.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2013c. Freedom in the World 2013.
https://freedomhouse.org
Accessed January 28, 2019.
Freedom House. 2013a. Freedom of the Press 2013.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House. 2013b. Freedom on the Net 2013.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2014a. Freedom of the Press 2014.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House. 2014b. Freedom on the Net 2014.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2015a. Freedom of the Press 2015.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House.
Accessed September 26, 2019:
2015b.
Freedom on the Net 2015.
335
https://freedomhouse.org
Freedom House. 2016a. Freedom of the Press 2016.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House. 2016b. Freedom on the Net 2016.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2017a. Freedom in the World 2017.
https://freedomhouse.org
Accessed September 10, 2019:
Freedom House. 2017b. Freedom on the Net 2017.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2017c. Freedom of the Press 2017.
https://freedomhouse.org
Accessed December 6, 2019:
Freedom House. 2018a. Freedom in the World 2018.
https://freedomhouse.org
Accessed January 29, 2019:
Freedom House. 2018b. Freedom on the Net 2018.
https://freedomhouse.org
Accessed September 26, 2019:
Freedom House. 2019. Freedom in the World 2019.
https://freedomhouse.org
Accessed February 17, 2020:
Freer, Courtney. 2017. “Renter Islamism in the Absence of Elections: The Political Role of
Muslim Brotherhood Affiliates in Qatar and the United Arab Emirates.” International
Journal of Middle East Studies 49(3): 479-500.
Freer, Courtney. 2019. “History (Jordan)” in Europa World. London: Routledge. Accessed
January 20, 2019: http://www.europaworld.com
Front des Forces Socialistes. 2019. “Site Officiel du Front des Forces Socialistes.” Internet
Archive. Accessed August 22, 2019: http://web.archive.org/web/∗ /http://www.ffsdz.com/
336
Gallab, Abdullahi. 2019. Professor of Religious Studies, Arizona State University, Tempe.
Personal communication.
Garon, Lise. 2003. Dangerous Alliances: Civil Society, the Media and Democratic Transition
in North Africa. London: Zed Books.
Geddes, Barbara and John Zaller. 1989. “Sources of Popular Support for Authoritarian
Regimes.” American Journal of Political Science 33(2): 319-347.
Geddes, Barbara, Joseph Wright and Erica Frantz. 2014. “Autocratic Breakdown and
Regime Transitions: A New Data Set.” Perspectives on Politics 12(2): 313-331.
Geddes, Barbara, Joseph Wright, and Erica Frantz. 2018. How Dictatorships Work. New
York: Cambridge University Press.
Geddes, Barbara.
Manuscript.
2006.
“Why Parties and Elections in Authoritarian Regimes?”
Gehlbach, Scott and Konstantin Sonin. 2014. “Government Control of the Media.” Journal
of Public Economics 118: 163-171.
Gerschewski, Johannes. 2013. “The three pillars of stability: legitimation, repression, and
co-optation in autocratic regimes.” Democratization 20(1): 13-38.
Ghabra, Shafeeq. 1991. “Voluntary Associations in Kuwait: The Foundation of a New
System?” Middle East Journal 45(2): 199-215.
Ghabra, Shafeeq. 1997. “Balancing State and Society: The Islamic Movement in Kuwait.”
Middle East Policy Council V(2). Accessed August 11, 2019: www.mepc.org
Ghabra, Shafeeq. 2014. “Kuwait: At the Crossroads of Change or Political Stagnation.”
Middle East Institute. Accessed May 31, 2019: www.mei.edu
Ghanmi, Lamine. 1993. “Algeria sacks prime minister amid economic gloom.” August 21.
Accessed November 10, 2019: www.dowjones.com/products/factiva
Ghanmi, Lamine. 2004. “Moroccan Islamists call Spain’s bombers ‘aliens’.” March 20.
337
Reuters. Accessed April 7, 2019: www.dowjones.com/products/factiva
Ghareeb, Edmund. 2000. “New Media and the Information Revolution in the Arab World:
An Assessment.” Middle East Journal 54(3): 395-418.
Ghiles, Francis. 1984. “Tunisia — Hard thinking follows jolt to elite — How the bread riots
tarnished the image of the Bouguiba government.” March 16. The Financial Times.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Ghiles, Francis. 1986. “The Applause Dies Out / Uncertain future for President Bourguiba
of Tunisia.” July 14.
Financial Times.
Accessed June 11, 2019:
www.dowjones.com/products/factiva
Ghiles, Francis. 1989. “Tunisian Ruling Party Finds its Polling Habits Hard to Break:
The Turn-Out was Low in Yesterday’s Elections.” April 3. Accessed March 11, 2020:
www.dowjones.com/products/factiva
Ginkel, John and Alastair Smith. 1999. “So You Say You Want a Revolution: A
Game-Theoretic Explanation of Revolution in Repressive Regimes.” Journal of Conflict
Resolution 43(3): 291-316.
Gläßel, Christian and Katrin Paula. 2019. “Sometimes Less Is More: Censorship, News
Falsification, and Disapproval in 1989 East Germany.” American Journal of Political
Science, Early View, December.
Gleditsch, Nils Petter, Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg, and
Håvard Strand. 2002. “Armed Conflict 1946-2001: A New Dataset.” Journal of Peace
Research 39(5): 615-637.
Global Security. 2011. “Bahrain — Political Parties.” September 7. Accessed June 7, 2019:
www.globalsecurity.org
Global Security. 2019. “Sheikh Sabah Al Ahmad Al Jaber Al Sabah.” Accessed November
15, 2019: www.globalsecurity.org
Global Security. 2020. “The Riots of 1984 and Their Aftermath.” Accessed February 13,
2020: www.globalsecurity.org
338
Globe and Mail, The. 1994. “Oxford Analytica Gulf States: Islamic Opposition groups
threaten
gulf
elites.”
October
18.
Accessed
May
27,
2019:
www.dowjones.com/products/factiva
Goemans, H. E., Kristian Skrede Gleditsch, and Giacomo Chiozza. 2009. “Introducing
Archigos: A Data Set of Political Leaders.” Journal of Peace Research 46(2): 269-183.
Goldschmidt, Arthur Jr. 2013. Historical Dictionary of Egypt. Fourth Edition. Lanham,
MD: Scarecrow.
Goldstein, Eric. 1999. “The Internet in the Mideast and North Africa.” Human Rights
Watch. Accessed February 4, 2020: www.hrw.org
Guardian, The. 1989. “Fundamentalists Mount Strong Election Push.” April 3. Accessed
March 11, 2020: www.dowjones.com/products/factiva
Guardian, The. 2013. “Egypt: Back with a vengeance.” December 26. Accessed June 25,
2019: www.dowjones.com/products/factiva
Gueye, Bakari. 2013. “Maghreb Journalists Mark World Press Freedom Day.” May 7.
Accessed August 16, 2019: www.dowjones.com/products/factiva
Gulf Daily News. 2016. “Court suspends Al Wefaq.” June 15. Accessed August 8, 2019:
www.dowjones.com/products/factiva
Gulf, The. 2011. “Old problems, new rules.” May 1.
www.dowjones.com/products/factiva
Accessed July 12, 2019:
Gunter, Barrie and Roger Dickinson. 2013. “The Changing Nature of News and the Arab
World.” In “News Media in the Arab World: A Study of 10 Arab and Muslim Countries.”
Barrie Gunter and Roger Dickinson, editors. New York: Bloomsbury. 1-20.
Guriev, Sergei and Daniel Treisman.
Manuscript.
2019.
“A Theory of Informational Autocracy.”
Gurr, Ted Robert. 1970. Why Men Rebel. Princeton, NJ: Princeton University Press.
339
Haddad, Yvonne. 1992. “Islamists and the ‘Problem of Israel’: The 1967 Awakening.”
Middle East Journal 46(2): 266-285.
Haddadin, Haitham. 2006. “Kuwait emir names new cabinet with few key changes.”
February 9. Reuters. Accessed May 27, 2019: www.dowjones.com/products/factiva
Hafez, Kai. 2010. “The Role of Media in the Arab World’s Transformation Process.” In
Bound to Cooperate: Europe and the Middle East II. Christian-Peter Hanelt
and Almut M´’oller, editors. G´’utersloh, Germany: Verlag Bertelsmann Stiftung.
Hamid, Shadi. 2013. “The Islamic Action Front in Jordan.” In The Oxford Handbook of
Islam and Politics. John L. Esposito and Emad El-Din Shahin, editors. Accessed March
22, 2019: www.oxfordhandbooks.com
Hamid, Shadi. 2017. “The end of Nasserism: How the 1967 War opened new space for
Islamism in the Arab world.” June 5.
Markaz blog.
Brookings Institution.
www.brookings.edu
Hamzawy, Amr. 2009. “Between Government and Opposition: The Case of the Yemeni
Congregation for Reform.” Washington, DC: Carnegie Endowment for International
Peace.
Harik, Iliya. 1994. “Rethinking Civil Society: Pluralism in the Arab World.” Journal of
Democracy 5(3): 43-56.
Harris, D. R. 1988. “Tunisia.” In The Middle East and North Africa 1989. London: Europa.
Harris, D. R. 1990. “Tunisia.” In The Middle East and North Africa 1991. London: Europa.
Hasan, Omar. 2009. “Kuwait appoints fifth oil minister in three years.” February 9. Agence
France Presse. Accessed May 28, 2019: www.dowjones.com/products/factiva
Hasan, Omar. 2011. “Youth take wealthy Kuwait into Arab Spring.” November 29. Agence
France Presse. Accessed May 31, 2019: www.dowjones.com/products/factiva
Hedges, Matthew and Giorgio Cafiero. 2017. “The GCC and the Muslim Brotherhood:
What Does the Future Hold?” Middle East Policy Council XXIV(1). Accessed March
340
26, 2019: www.mepc.org
Hendawi, Hamza. 2018. “Sudan journalists go on strike in support of protests.” December
27. Associated Press. Accessed February 4, 2020: https://apnews.com
Hendrix, Cullen, Stephan Haggard, and Beatriz Magaloni. 2009. “Grievance and
Opportunity: Food Prices, Political Regime, and Protest.” Manuscript.
Herb, Michael.
2019.
www.kuwaitpolitics.org
“Kuwait Politics Database.” Accessed May 27, 2019:
Hindley, Angus. 1992. “Fresh blood for Saad cabinet.” October 30. Middle East Economic
Digest. Accessed May 28, 2019: www.dowjones.com/products/factiva
Hinnebusch, Raymond. 2014. “Syria.” In The Middle East. Thirteenth Edition. Ellen Lust,
editor. Washington, DC: Sage and CQ Press. Pages 764-788.
Hiro, Dilip. 2003. The Essential Middle East: A Comprehensive Guide. New York: Carroll
& Graf.
Hiro, Dilip. 2013. A Comprehensive Dictionary of the Middle East. Northampton, MA:
Olive Branch.
Hirst, David. 1994. “Algerian PM Quits as IMF Deal Signed.” April 12. The Guardian.
Accessed May 13, 2019: www.dowjones.com/products/factiva
Hoagland, Jim. 1988. “Tunisia, Used to One-Man Rule, Adjusts to Measure of Pluralism.”
February 18.
The Washington Post.
Accessed March 11, 2020:
www.dowjones.com/products/factiva
Hobbs, William R. and Margaret E. Roberts. 2018. “How Sudden Censorship Can Increase
Access to Information.” American Political Science Review 112(3): 621—636.
Horrace, William C. and Ronald L. Oaxaca. “Results on the bias and inconsistency of
ordinary least squares for the linear probability model.” Economics Letters 90: 321—
327.
341
Howard, Philip N. and Muzammil M. Hussain. 2011. “The Upheavals in Egypt and Tunisia:
The Role of Digital Media.” Journal of Democracy 22(3): 35-48.
Huang, Haifeng, Serra Boranbay-Akan, and Ling Huang. 2019. “Media, Protest Diffusion,
and Authoritarian Resilience.” Political Science Research and Methods 7(1): 23—42.
Hudson, Michael C. 1991. “After the Gulf War: Prospects for Democratization in the Arab
World.” The Middle East Journal 45: 407-426.
Human Rights Watch.
www.hrw.org
1992.
“World Report 1992.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1993.
“World Report 1993.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1994.
“World Report 1994.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1995.
“World Report 1995.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1996 “World Report 1996.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1997a.
“World Report 1997.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1997b.
“Freedom of the Press.” Accessed August 5, 2019:
Human Rights Watch.
www.hrw.org
1998.
“World Report 1998.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
1999.
“World Report 1999.” Accessed September 20, 2019:
Human Rights Watch.
2000.
“World Report 2000.” Accessed September 20, 2019:
342
www.hrw.org
Human Rights Watch.
www.hrw.org
2001.
“World Report 2001.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2002.
“World Report 2002.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2003.
“World Report 2003.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2005.
“World Report 2005.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2006.
“World Report 2006.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2007.
“World Report 2007.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2008.
“World Report 2008.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2009.
“World Report 2009.” Accessed September 20, 2019:
Human Rights Watch. 2010a. “Iraq’s 2010 National Elections: A Human Rights Platform
for Candidates.” February 25. Accessed January 25, 2019: https://www.hrw.org
Human Rights Watch.
www.hrw.org
2010b.
“World Report 2010.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2011.
“World Report 2011.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2012.
“World Report 2012.” Accessed September 20, 2019:
343
Human Rights Watch.
www.hrw.org
2013.
“World Report 2013.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2014.
“World Report 2014.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2015.
“World Report 2015.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2016.
“World Report 2016.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2017.
“World Report 2017.” Accessed September 20, 2019:
Human Rights Watch.
www.hrw.org
2018.
“World Report 2018.” Accessed September 20, 2019:
Human Rights Watch. 2019. “Biography of Hassan al Turabi.” Accessed November 11, 2019:
www.hrw.org
Huntington, Samuel P. 1991. “Democracy’s Third Wave.” Journal of Democracy 2(2): 12-34.
Ibrahim, Youssef M. 1990. “Islamic Party in Algeria Defeats Ruling Groupin Local
Elections.” June 14.
New York Times.
Accessed March 11, 2020:
www.dowjones.com/products/factiva
IHS Markit. 2014. “High-profile corruption allegations spur Kuwait opposition demands
for parliamentary system.” July 18.
Accessed May 28,
2019:
www.dowjones.com/products/factiva
Immigration and Refugee Board of Canada. 2000. “Algeria: Newspaper ‘El-Hiwar’ (19962000).” February 22. Accessed August 23, 2019: www.refworld.org
Inclán, Marı́a de la Luz. 2009. “Repressive Threats, Procedural Concessions, and the
Zapatista Cycle of Protests, 1994—2003.” Journal of Conflict Resolution 53(5): 794-819.
344
Independence Party. 2019. “Parti de l’Istiqlal.” Internet Archive. Accessed October 16,
2019: http://web.archive.org/web/∗ /istiqlal.ma
Index on Censorship. 1992. “Index.” 21(2): 34-41.
Index on Censorship. 1996. “Index.” 25(5): 78-107.
Index on Censorship. 2002. “Index.” 31(4): 108-135.
Ingram, Jamie. 2011. “Protest Groups Call for Rally in Kuwait on Friday.” September 15.
IHS Markit. Accessed May 31, 2019: www.dowjones.com/products/factiva
Ingram, Jamie.
2012.
“Islamist Group Slams UAE Government for Continued
Persecution.” September 13.
IHS Markit.
Accessed March 23, 2019:
www.dowjones.com/products/factiva
Ingram, Jamie. 2013. “Protests continue in Kuwait as convicted ex-MP refuses arrest.”
April 17. Accessed July 27, 2019: www.dowjones.com/products/factiva
Inter-Parliamentary Union. 2019. “PARLINE database on national parliaments.” Accessed
April 3, 2019: http://archive.ipu.org
International Crisis Group. 2011. “Popular Protests in North Africa and the Middle East
(III): The Bahrain Revolt.” April 6. Middle East/North Africa Report No. 105.
International Crisis Group. 2018. “Collapse of the Houthi-Saleh Alliance and the Future of
Yemen’s War.” January 11. Accessed July 27, 2019: www.crisisgroup.org
International Foundation for Electoral Systems. 2019. “Election Guide.” Accessed May 23,
2019: http://www.electionguide.org
International Research & Exchanges Board. 2006. Media Sustainability Index — Middle
East and North Africa. Accessed September 23, 2019: www.irex.org
International Research & Exchanges Board. 2008. Media Sustainability Index 2006/2007 —
The Development of Sustainable and Independent Media in the Middle East and North
Africa. Accessed September 23, 2019: www.irex.org
345
International Research & Exchanges Board. 2009. Media Sustainability Index 2008 —
Development of Sustainable and Independent Media in the Middle East and North Africa.
Accessed September 23, 2019: www.irex.org
International Research & Exchanges Board. 2010. Media Sustainability Index 2009:
Development of Sustainable Independent Media in Africa. Accessed September 22,
2019: www.irex.org
International Research & Exchanges Board. 2011a. “Media Sustainability Index 2010:
Development of Sustainable and Independent Media in the Middle East and North
Africa. Accessed September 23, 2019: www.irex.org
International Research & Exchanges Board. 2011X. Media Sustainability Index 2010:
Development of Sustainable Independent Media in Africa. Accessed September 22,
2019: www.irex.org
International Research & Exchanges Board. 2013. Media Sustainability Index 2012: The
Development of Sustainable Independent Media in Africa. Accessed November 16, 2019:
www.irex.org
International Telecommunication Union. 2020. “Statistics.” Accessed February 14, 2020:
https://www.itu.int/en/ITU-D/Statistics/Pages/stat
Islamic Action Front. 2019. “ ú×CB @
k H Qk” Internet Archive. Accessed October
ÉÒªË@ éîD
.. .
21, 2019: http://web.archive.org/web/∗ /jabha.net
Jadaliyya. 2011. “Democratic Alliance for Egypt.” November 18. Accessed June 25, 2019:
www.jadaliyya.com
Jakes, Lara. 2010. “Iraqi sheik threatens boycott over ballot purge.” January 30. Associated
Press. Accessed June 22, 2019: www.dowjones.com/products/factiva
Jamaleddine, Ahmad. 2014. “Sisi brings back Egypt’s police state with a vengeance.”
December 20. Al Akhbar. Accessed June 25, 2019: www.dowjones.com/products/factiva
Joffe, Lawrence. 2016. “Hassan al-Turabi obituary.” March 11. The Guardian. Accessed
November 11, 2019: www.theguardian.com
346
Jones, Derek. 2001. Censorship: A World Encyclopedia. New York: Routledge.
Jones, Marc. 2019. “History (Bahrain).” In Europa World Online. Routledge. Accessed
June 7, 2019: http://europaworld.com
Jordan Times. 2011. “Muslim Brotherhood website hacked.” August 14. Accessed October
19, 2019: www.dowjones.com/products/factiva
Jordan Times. 2014. “Over 60% of news in major Arabic dailies taken from Petra, press
releases — report.” September 22.
Accessed August 10,
2019:
www.dowjones.com/products/factiva
Jordan Times. 2015a. “Al Sabeel chief editor released on bail after brief detention.” August
18. Accessed August 10, 2019: www.dowjones.com/products/factiva
Jordan Times. 2015b. “Authorities close Yarmouk satellite channel’s broadcast studios.”
August 31. Accessed August 10, 2019: www.dowjones.com/products/factiva
Jordan Times. 2016. “Brotherhood offers dovish leader top position.” April 21. Accessed
August 10, 2019: www.dowjones.com/products/factiva
Jordan Times. 2017. “Media watchdog records 172 ‘violations’ against journalists.”
December 5. Accessed August 10, 2019: www.dowjones.com/products/factiva
Justice and Spirituality. 2019a. “ àAkB @ð
ÈYªË@” Internet Archive.
Accessed October 10,
2019: http://web.archive.org/web/∗ /aljamaa.org
Justice and Spirituality. 2019b. “ àAkB @ð
ÈYªË@” Internet Archive.
Accessed October 10,
2019: http://web.archive.org/web/∗ /aljamaa.com
Justice and Spirituality. 2019c. “ àAkB @ð
ÈYªË@” Internet Archive.
Accessed October 10,
2019: http://web.archive.org/web/∗ /aljamaa.net
Justice and Spirituality. 2019d. “ àAkB @ð
ÈYªË@” Internet Archive.
2019: http://web.archive.org/web/∗ /aljamaa.info
347
Accessed October 10,
Kabyle Universel.
2019.
http://kabyleuniversel.com
“Kabyle
Universel.”
Accessed
August
22,
2019:
Kamal, Ahmad M. 2018.
“The democratic peak of the authoritarian swing:
Post-revolutionary Egypt’s politics, media, and public.” In Authoritarian and Populist
Influences in the New Media. Sai Felicia Krishna-Hensel, editor. New York: Routledge.
Pages 39-64.
Karam, Souhail. 2003. “Moroccan Islamists could help thwart bombers.” May 21. Reuters.
Accessed April 7, 2019: www.dowjones.com/products/factiva
Keesing’s Record of World Events. 1988. “Retirement of President Bourguiba.” March.
34(3): 35801.
Kendall-Taylor, Andrea and Erica Frantz. 2014. “How Autocracies Fall.” Washington
Quarterly 37(1): 35-47.
Kenyon, Peter. 2010. “Tense Political Climate Proceeds Election in Bahrain.” October 12.
National
Public
Radio.
Accessed
October
18,
2019:
www.dowjones.com/products/factiva
Khalaf, Roula. 1997. “Algeria Poll Disputed by Opposition.” June 7. Financial Times.
Accessed July 13, 2019: 2019: www.dowjones.com/products/factiva
Khalaf, Roula. 2000. “Arrests setback for Morocco human rights.” December 12. Financial
Times. Accessed October 6, 2019: www.dowjones.com/products/factiva
Khazen, Jihad. 1999. “Censorship and State Control of the Press in the Arab World.”
Press/Politics 4(3): 87-92.
King, Gary, Jennifer Pan, and Margaret E. Roberts. 2013. “How Censorship in China Allows
Government Criticism but Silences Collective Expression.” American Political Science
Review 107(2): 326-343.
King, Gary, Jennifer Pan, and Margaret E. Roberts. 2017. “How the Chinese Government
Fabricates Social Media Posts for Strategic Distraction, not Engaged Argument.”
American Political Science Review 111(3): 484-501.
348
Kirkpatrick, David D. 2012. “Abduction and Hotel Attack Spoltight Libya’s Disarray.”
March 25. The New York Times. Accessed January 26, 2019: https://www.nytimes.com
Kirkpatrick, David D. 2014. “As Moderate Islamists Retreat, Extremists Surge Unchecked.”
Jun 18. The New York Times. Accessed April 3, 2019: www.nytimes.com
Klmty.net.
2019.
” Internet Archive.
“ úæÒÊ¿
Accessed September 27, 2019:
http://web.archive.org/web/∗ /klmty.net
Knight, Brian and Ana Tribin. 2019. “Opposition Media, State Censorship, and Political
Accountability: Evidence from Chavez’s Venezuela.” Manuscript.
Kokan, Jane. 1994. “Riot Deaths as Hassan Faces Challenge.” February 20. The Sunday
Times. Accessed April 4, 2019: www.dowjones.com/products/factiva
Korany, Bahgat. 2010. “Arab Media over the Past Twenty Years: Opportunities and
Challenges.” In The Changing Middle East: A New Look at Regional Dynamics. Bahgat
Korany, editor. Cairo: American University of Cairo Press. 59-83.
Korbani, Agnes. 1995. The Political Dictionary of Modern Middle East. New York:
University Press of America.
Korn, David A. 1993. “The Middle East: Islam vs. the Established Order.” In Freedom in
the World: The Annual Survey of Political Rights and Civil Liberties, 1992-1993. R.
Bruce McColm, Dale Bricker, James Finn, Charles Graybow, Jonathan D. Karl, Douglas
W. Payne, Joseph E. Ryan, and George Zarycky, editors. New York: Freedom House.
Krämer, Gudrun. 1992. “Liberalization and Democracy in the Arab World.” Middle East
Report 174: 22-25, 35.
Krämer, Gudrun. 1994. “The Integration of the Integrists: a comparative study of Egypt,
Jordan, and Tunisia.” In Democracy Without Democrats?: Renewal of Politics in the
Muslim World. Ghassan Salame, editor. London: I.B. Tauris. 200-226.
Kramer, Robert S., Richard A. Lobban Jr., and Carolyn Fluehr-Lobban. 2013. Historical
Dictionary of the Sudan. Fourth Edition. Lanham, MD: Scarecrow.
349
Krieger, Joel. 2003. The Oxford Companion to Politics of the World. New York: Oxford
University Press.
Kuran, Timur. 1991. “Now Out of Never: The Element of Surprise in the East European
Revolution of 1989.” World Politics 44(1): 7-48.
Kwak, Do Won, Robert S. Martin, and Jeffrey M. Woolridge. 2018. “The robustness of
conditional logit for binary response panel data models with serial correlation.” August.
U.S. Bureau of Labor Statistics Working Paper 502.
L’Expression. 2011. “Qui est Foued Mebazaa?” January 16. Accessed November 15, 2019:
www.dowjones.com/products/factiva
L’Expression. 2012. “Première radio privée lancée en Mauritanie.” May 7. Accessed August
16, 2019: www.dowjones.com/products/factiva
La Libre. 2013. “Il lutte contre la justification religieuse de ce crime: il est passible de la
peine de mort.” June 10. Accessed May 23, 2019: https://www.lalibre.be
LADDH. 2019. Algerian Human Rights League. Accessed May 17, 2019: https://laddhalgerie.org
Lansford, Tom. 2012. Political Handbook of the World 2012. Thousand Oaks, CA: CQ
Press. Accessed March 23, 2019: http://library.cqpress.com/phw/index.php
Lansford, Tom. 2017. Political Handbook of the World 2016-2017. Thousand Oaks, CA:
CQ Press. Accessed March 25, 2019: http://library.cqpress.com/phw/index.php
Laremont, Ricardo. 2009. “Justice and Benevolence Party.” In The Oxford Encyclopedia
of Islam and Politics.
John L. Esposito, editor.
Accessed April 8, 2019:
www.oxfordislamicstudies.com
Layachi, Azzedine. 2009. “Islam and Politics in North Africa.” In The Oxford Handbook
of Islam and Politics.
John L. Esposito, editor.
Accessed April 7, 2019:
www.oxfordislamicstudies.com
Legum, Colin. 1987. “Tunisia.” In Africa Contemporary Record, XX 1986-1987. Africana
350
Publishing.
Legum, Colin. 1988. “Tunisia.” In Africa Contemporary Record, XXI 1987-1988. Africana
Publishing.
Legum, Colin. 1992a. “Tunisia.” In Africa Contemporary Record, XXIII 1990-1992.
Africana Publishing.
Legum, Colin. 1992b. “Morocco.” In Africa Contemporary Record, XXIII 1990-1992.
Africana Publishing.
Legum, Colin. 1996. “Algeria.” In Africa Contemporary Record, XXV 1994-1996. Africana
Publishing.
Legum, Colin. 2000. “Morocco.” In Africa Contemporary Record, XXVII 1998-2000.
Africana Publishing.
Levinson, Charlse. 2011. “Coalition Looks Set to Form in Tunisia.” The Wall Street Journal.
October 28. Accessed September 10, 2019: www.dowjones.com/products/factiva
Lewis, Ioan M. and J´’org H.A. Janzen. 2018. “Somalia.” Encyclopaedia Britannica.
Accessed January 23, 2019: https://www.britannica.com
Lim, Merlyna. 2012. “Clicks, Cabs, and Coffee Houses: Social Media and Oppositional
Movements in Egypt, 2004—2011.” Journal of Communication 62: 231—248.
Little, Andrew T. 2016. “Communication Technology and Protest.” Journal of Politics 78(1):
152—66.
Lohmann, Suzanne. 1994. “The Dynamics of Informational Cascades: The Monday
Demonstrations in Leipzig, East Germany, 1989-91.” World Politics 47(1): 42-101.
Lorentzen, Peter. 2014. “China’s Strategic Censorship.” American Journal of Political
Science 58(2): 402—414.
Luck, Taylor. 2017. “How Trump move on Jerusalem throws a lifeline to Muslim
Brotherhood.” December 18. Christian Science Monitor. Accessed March 22, 2019:
351
www.dowjones.com/products/factiva
Lulu, Tahiyya. 2011. “The real story of Bahrain’s divided society.” March 3. The Guardian.
Accessed March 16, 2020: www.theguardian.com
Lust-Okar, Ellen. 2005. Structuring Conflict in the Arab World: Incumbents, Opponents,
and Institutions. New York: Cambridge University Press.
Lynch, Marc. 2016. “Is the Muslim Brotherhood a terrorist organization or a firewall against
violent extremism?” March 7. Washington Post. Monkey Cage blog. Accessed March
28, 2019: https://www.washingtonpost.com/news/monkey-cage/
MacDonald, Susan. 1988. “Ben Ali sheds old guard; Cabinet reshuffle in Tunisia.” August 5.
The Times of London. Accessed March 11, 2020: www.dowjones.com/products/factiva
MacFarquhar, Neil. 2012. “Trying to Mold a Post-Assad Syria From Abroad.” The New
York Times. May 5. Accessed March 30, 2019: www.nytimes.com
Mada Masr. 2015. “Who speaks for the Brotherhood?” October 1. Accessed September 24,
2019: https://madamasr.com/en
Maghraoui, Abdeslam M. 2001. “Monarchy and Political Reform in Morocco.” Journal of
Democracy 12 (1): 73-86.
Maghreb Confidential. 2011. “Ennahda & UGTT profit from political vacuum?” April 14.
Accessed August 17, 2019: www.dowjones.com/products/factiva
Maimone, Giuseppe. 2019. “Recent History (Mauritania).” In Europa World Online.
Routledge. Accessed May 21, 2019: http://europaworld.com
Makedhi, Madjid. 2013. “Sofiane Djilali opposé à un 4e mandate pour Bouteflika.” March
3. El Watan. Accessed May 18, 2019: www.dowjones.com/products/factiva
Masbah, Mohammed. 2014. “Morocco’s loyalists reinvent themselves.” January 31. Daily
Star. Accessed April 8, 2019: www.dowjones.com/products/factiva
Mattar, Philip. 2004. Encyclopedia of the Modern Middle East and North Africa, Volume
352
3. Second Edition. New York: Thomson Gale.
Mellor, Noah. 2017. Voice of the Muslim Brotherhood: Da’wa, Discourse, and Political
Communication. New York: Taylor & Francis.
Menon, Kesava. 2002. “Bahrain allows a slice of democracy.” May 29. The Hindu. Accessed
June 7, 2019: www.dowjones.com/products/factiva
Michael, Maggie. 2014. “Amid turmoil, Libya holds parliament elections.” Associated Press.
June 25. Accessed September 10, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1988. “Government of Tunisia.” August 19. Accessed March
11, 2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1990a. “Formation of the Yemeni Unionist Party.” January
19. Accessed June 18, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1990b. “Copies of Tunis Weekly Confiscated.” June 29.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1990c. “A government reshuffle was announced on 3 March.”
March 16. Accessed March 11, 2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1990d. “Islamic Radicals Demonstrate.” October 12.
Accessed March 10, 2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1990e. “Nahda Members Arrested.” November 30. Accessed
February 25, 2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1991a. “A critical year for the kingdom.” December 27.
Accessed March 29, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1991b. “Riots hit Sanaa as tensions mount — shooting dead
of traffic policeman by army colonel acts as catalyst.” November 1. Accessed June 18,
2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1991c. “Return to reform — Determination needed to
353
stimulate sufficient investment to revive economy.” September 6. Accessed March 9,
2020: www.dowjones.com/products/factiva
Middle East Economic Digest. 1992c. “An uneasy balance of political forces.” July 31.
Accessed March 23, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1992d. “Front de Liberation Nationale deeply split over
cancellation of elections.” February 7.
Accessed May 11,
2019:
www.dowjones.com/products/factiva
Middle East Economic Digest. 1992e. “Islamist leader released after three years of house
arrest.” October 30. Accessed April 4, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 1992f. “Political Crisis Reaches Climax with Resignation of
President
Chadli.”
January
24.
Accessed
November
10,
2019:
www.dowjones.com/products/factiva
Middle East Economic Digest. 1997. “Presidential aides dominate new government.” July
7. Accessed April 22, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest. 2008. “Mohammed al-Olaim.” May 23. Accessed May 28,
2019: www.dowjones.com/products/factiva
Middle East Economic Digest.1993a. “Military Shake-Up Strengthens Hard Line.” July 23.
Accessed November 10, 2019: www.dowjones.com/products/factiva
Middle East Economic Digest.1993b. “Nezzar to Withdraw from Political Life When
Mandate of Higher Council of State Expires.” October 11. Accessed November 10,
2019: www.dowjones.com/products/factiva
Middle East Monitor. 2016. “Yarmouk TV to stop transmitting from Jordan.” April 19.
Accessed August 10, 2019: www.middleeastmonitor.com
Middle East North Africa Financial Network. 2015a. “Rebels kidnap journalists.” April 20.
Accessed September 29, 2019: www.dowjones.com/products/factiva
Middle East North Africa Financial Network. 2015b. “Sudanese Pres. announces new
354
cabinet lineup.” June 7. Accessed April 1, 2019: www.dowjones.com/products/factiva
Middle East Reporter, The. 2012. “Morocco Politics.” January 6. Accessed March 29, 2019:
www.dowjones.com/products/factiva
Møller, Jørgen and Svend-Erik Skaaning. 2013. “The Third Wave: Inside the Numbers.”
Journal of Democracy 24(4): 97-109.
Montabes Pereira, Juan and Marı́a A. Parejo Fernández. 2003. “Morocco.” In Elections in
Africa: A Data Handbook. Dieter Nohlen, Bernard Thibaut, and Michael Krennerich,
editors.
Oxford Scholarship Online.
Accessed June 14,
2019:
http://www.oxfordscholarship.com
Montabes Pereira, Juan. 2003. “Tunisia.” In Elections in Africa: A Data Handbook. Dieter
Nohlen, Bernard Thibaut, and Michael Krennerich, editors. Oxford Scholarship Online.
Accessed March 23, 2019: http://www.oxfordscholarship.com
Morgan, Marwa. 2017. “Mauritania Cracks Down on Critical Press After Referendum.”
October 18. All Africa. Accessed August 16, 2019: www.dowjones.com/products/factiva
Mostafa, Amr. 2019. “Even as Sisi frees prisoners, opponents unite against him.” June 20.
Al-Monitor. Accessed June 25, 2019: www.al-monitor.com
Mostafa, Mahmoud. 2015. “The rules of the game: Who joins last, takes over.” July 1.
Daily News Egypt. Accessed June 25, 2019: www.dowjones.com/products/factiva
Mostyn, Trevor and Albert Hourani, edtiors. 1988. The Cambridge Encyclopedia of the
Middle East and North Africa. New York: Cambridge University Press.
Motaouakal, Abdelouahad. 2014. “Al-Adl wal-Ihsan: An Explanation of its Rise and its
Strategy for Social and Political Reform in Morocco.” Dissertation: University of Exter.
Accessed March 29, 2019: https://ore.exeter.ac.uk/repository/
Mounassar, Hammoud. 2012. “Yemen’s Saleh formally steps down after 33 years.”
February 27.
Agence France Presse.
Accessed November 11, 2019:
www.dowjones.com/products/factiva
355
Moussa, Jedjib Sidi. 2015. “Occultation étatique et Réappropriations Privées: Tensions
Mémorielles Autor de la Question Messaliste en Algérie (1964-1994).” Matériaux pour
l’histoire de notre temps 117-118 (3): 31-38.
Mummolo, Jonathan and Erik Peterson. 2018. “Improving the Interpretation of Fixed
Effects Regression Results.” Political Science Research and Methods 6(4): 829—835.
Murphy, Brian. 2011. “Kuwait next in line for protests but no stranger to political
battles.” March 6.
Associated Press.
Accessed May 31, 2019:
www.dowjones.com/products/factiva
Murphy, Emma C. 1998. “Legitimacy and economic reform in the Arab world.” Journal of
North African Studies 3(3): 71-92.
Murphy, Emma C. 1999. Economic and Political Change in Tunisia: From Bourguiba to
Ben Ali. New York: St. Martin’s.
Murphy, Emma C. 2013. “The Tunisian elections of October 2011: a democratic consensus.”
The Journal of North African Studies, 18(2): 231-247.
Murphy, Kim. 1992. “Media : Muslim Voices Joining Mideast’s Press.” August 11. Los
Angeles Times. Accessed August 5, 2019: www.latimes.com
Muslih, Mohammad and Michaelle Browers. 2014. “Democracy.” In The Oxford
Encyclopedia of Islam and Politics. Emad El-Din Shahin, editor. Oxford University
Press. Accessed February 13, 2020: www.oxfordreference.com
Muslim Brotherhood 2019e. “ àXPB@
g”
áÒÊÖÏ @ à@ñkB @ é«AÔ
.
Internet Archive. Accessed
October 21, 2019: http://web.archive.org/web/∗ /ikhwanjo.com
Muslim Brotherhood. 2019a. “ àñÒÊÖÏ @
à@ñkB@” Internet Archive.
Accessed September 27,
2019: http://web.archive.org/web/∗ /ikhwanonline.net
Muslim Brotherhood. 2019b. “ àñÒÊÖÏ @
à@ñkB@” Internet Archive.
2019: http://web.archive.org/web/∗ /ikhwanonline.com
356
Accessed September 27,
Muslim Brotherhood.
2019c.
“ àñÒÊÖÏ @
à@ñkB@”
Accessed September 27, 2019:
https://ikhwanonline.com
Muslim Brotherhood. 2019d. “Ikhwan Web: The Muslim Brotherhood’s Official English web
site.” Accessed September 27, 2019: http://web.archive.org/web/∗ /ikhwanweb.com
Naoum, Abdelfattah. 2017. “Morocco’s PAM Under Pressure.” September 8. Carnegie
Endowment
for
International
Peace.
Accessed
April
8,
2019:
https://carnegieendowment.org
Nasser, Afrah. 2017. “The Yemen War, Media, and Propaganda.” May 3. MENASource
blog. Accessed September 29, 2019: https://www.atlanticcouncil.org/blogs/menasource
Nation’s Dignity, A. 2019. “ á£ð
Q» èQÓ” Internet Archive.
éÓ@
Accessed October 4, 2019:
http://web.archive.org/web/∗ /karametwatan.wordpress.com
National Democratic Institute. 2011. “Final Report on the Tunisian National Constituent
Assembly Elections October 23, 2011.” Accessed January 23, 2019: https://www.ndi.org
National Rally for Reform and Development.
Internet Archive.
/tewassoul.org
Accessed September 21, 2019:
National Rally for Reform and Development.
Internet Archive.
/tewassoul.mr
2019a.
2019b.
Accessed September 21, 2019:
“ éJÒJJË@ð
hCCË úæ£ñË@ ©Òj.JË@”
http://web.archive.org/web/∗
“ éJÒJJË@ð
hCCË úæ£ñË@ ©Òj.JË@”
http://web.archive.org/web/∗
Naylor, Phillip C. 2015. Historical Dictionary of Algeria. Fourth edition. New York:
Rowman & Littlefield.
Nicholson, Mark. 1991. “Al Sabahs keeping their hold on power: Kuwait’s new cabinet still
includes no opposition members.” April 22. Financial Times. Accessed May 27, 2019:
www.dowjones.com/products/factiva
Niethammer, Katja. 2014. “Persian Gulf States.” In The Middle East. Thirteenth Edition.
Ellen Lust, editor. Washington, DC: Sage and CQ Press. Pages 717-745.
357
No to Military Trials for Civilians. 2019.
September 27, 2019: www.nomiltrials.org
“ áJKYÒÊË
AjÒÊË B”
éK QºªË@ HAÒ»
Accessed
Noman, Helmi. 2007. “Testing for Internet filtering during parliamentary elections in
Jordan.” November 25.
OpenNet Initaitive.
Accessed August 10, 2019:
https://opennet.net
Noman, Helmi. 2010. “Algeria joins the Internet censors club.” January 2. OpenNet
Initiative. Accessed September 20, 2019: https://opennet.net
Nordenson, Jon. 2017. Online Activism in the Middle East:Â Political Power and
Authoritarian Governments from Egypt to Kuwait. London: I. B. Tauris.
Omar, Ali. 2014. “Cairo University SAC march dispersal leaves 2 critically wounded, 1
reported dead.” April 14.
Daily News Egypt.
Accessed June 25, 2019:
www.dowjones.com/products/factiva
OpenNet Initiative. 2005. “Internet Filtering in Bahrain in 2004-2005: A Country Study.”
February. Accessed November 6, 2019: https://opennet.net
OpenNet Initiative.
2009a.
https://opennet.net
“Bahrain.” August 6.
Accessed November 6, 2019:
OpenNet Initiative.
2009b.
https://opennet.net
“Egypt.” August 6.
OpenNet Initiative.
2009c.
https://opennet.net
“Jordan.” August 6.
Accessed October 21, 2019:
OpenNet Initiative.
2009d.
https://opennet.net
“Sudan.” August 7.
Accessed October 18, 2019:
Accessed September 27, 2019:
OpenNet Initiative. 2019a. “Internet Filtering in Bahrain in 2006-2007.” Accessed November
6, 2019: https://opennet.net
OpenNet Initiative. 2019b. “Internet Filtering in Egypt in 2006-2007.” August 6. Accessed
September 27, 2019: https://opennet.net
358
OpenNet Initiative. 2019c. “Internet Filtering in Jordan in 2006-2007.” Accessed 21, 2019:
https://opennet.net
OpenNet Initiative. 2019d. “Internet Filtering in Sudan in 2006-2007.” Accessed October
18, 2019: https://opennet.net
OpenNet Initiative. 2019e. “Yemen.” Accessed October 3, 2019: https://opennet.net
OpenNet Initiative.
https://opennet.net
2020.
“About Filtering.” Accessed February 14,
2020:
Ottaway, Marina and Amr Hamzawy. 2011. “Protest Movements and Political Change in
the Arab World.” January 28. Carnegie Endowment of International Peace. Accessed
February 15, 2020: https://carnegieendowment.org
Ould Sadi, Hademine. 2009. “Mauritania’s junta leader steps down to run for president.”
April 16.
Agence France Presse.
Accessed November 10, 2019:
www.dowjones.com/products/factiva
Owen, Roger. 1981. “The Arab economies in the 1970’s.” Middle East Research and
Information Project 100:Â 3—13.
Oxford Islamic Studies Online. 2019. “Muslim Brotherhood.” The Islamic World: Past and
Present. John L. Esposito, M. A. Abdel Haleem, A.J. Arberry, and Hanna E. Kassis,
editors. Accessed October 31, 2019: www.oxfordislamicstudies.com
PANAPRESS. 2014a. “Mauritania: Opposition protests ahead of 21 June election.” June 5.
Accessed May 22, 2019: www.dowjones.com/products/factiva
PANAPRESS. 2014b. “Mauritania: Radical opposition, civil society demand direct dialogue
with gov’t.” April 2. Accessed May 22, 2019: www.dowjones.com/products/factiva
Parks, Michael. 1986. “Blank Spaces in S. Africa Papers Protest Censorship.” June 18. Los
Angeles Times. Accessed August 9, 2019: www.latimes.com
Parsons, Nigel. 2019. “Recent History (Palestinian Territories),” in Europa World. London:
Routledge. Accessed January 20, 2019: http://www.europaworld.com
359
Partrick, Neill. 2019a. “History (Morocco).” In Europa World Online. Routledge. Accessed
April 8, 2019: http://europaworld.com
Partrick, Neill. 2019b. “History (Tunisia).” In Europa World Online. Routledge. Accessed
March 23, 2019: http://europaworld.com
Pazzanita, Anthony G. 2008. Historical Dictionary of Mauritania. Third Edition. Lanham,
MD: Scarecrow.
Pearce, Katy. 2014. “Two can play at that game: Social media opportunities in Azerbaijan
for government and opposition.” Demokratizatsiya: The Journal of Post-Soviet
Democratization 22: 39—66.
Peck, Malcolm C. 1995. “Eastern Arabian States: Kuwait, Bahrain, Qatar, United Arab
Emirates, and Oman.” In The Government and Politics of the Middle East and North
Africa. David E. Long and Bernard Reich, editors. San Francisco: Westview. Pages
113-149.
Peck, Malcolm C. 2008. Historical Dictionary of the Gulf Arab States, Second Edition.
Lanham, MD: Scarecrow.
Perkins, Kenneth J. 2016a. A History of Modern Tunisia. New York: Cambridge University
Press.
Perkins, Kenneth J. 2016b. Historical Dictionary of Tunisia. Third Edition. New York:
Rowman & Littlefield.
Peterson, J.E. and Gerd Nonneman. 2019. “History (Saudi Arabia)” in Europa World.
London: Routledge. Accessed January 21, 2019: http://europaworld.com
Petrova, Maria. 2008. “Inequality and media capture.” Journal of Public Economics 92(1-2):
183-212.
Phythian, Nicholas. 1996. “New party seeks Mauritania opposition breakthrough.” October
16. Reuters. Accessed May 23, 2019: www.dowjones.com/products/factiva
Pineau, Carol. 2000. “Sudan’s influential Umma Party quits opposition.” March 16. Agence
360
France Presse. Accessed April 1, 2019: www.dowjones.com/products/factiva
Pitman, Todd.
2009.
“Mauritania holds presidential vote with coup-leader as
front-runner.” July 18.
Associated Press.
Accessed November 10, 2019:
www.dowjones.com/products/factiva
Plotkin Boghardt, Lori. 2013. “The Muslim Brotherhood in the Gulf: Prospects for
Agitation.” June 10. The Washington Institute for Near East Policy. Accessed March
26, 2019: www.washingtoninstitute.org
Popular Action Bloc. 2019. “ úæªË@
.
ÉÒªË@ éÊ K” [Popular Action Bloc] Accessed October 4,
2019: https://www.facebook.com/pages/category/Politician/% D9% 83% D8% AA%
D9% 84% D8% A9-% D8% A7% D9% 84% D8% B9% D9% 85% D9% 84-% D8% A7%
D9% 84% D8% B4% D8% B9% D8% A8% D9% 8A-124214454255246
[https://www.facebook.com/pages/category/Politician/124214454255246- úæªËCÒªË@
éÊ J»]
.
Public Broadcasting Service. 2019. “April 6 Youth Movement.” Accessed June 25, 2019:
www.pbs.org
Purkiss, Alice. 2011. “Egypt: Journalists Strike Against Censorship.” October 6. Index on
Censorship. Accessed February 4, 2020: www.indexoncensorship.org
Radio Free Europe. 2011. “Egypt Cracks Down as Antigovernment Protests Continue.”
January 26. Accessed July 27, 2019: www.dowjones.com/products/factiva
Rally for Culture and Democracy. 2019a. “Rassemblement pour la Culture et la
Démocratie.”
Internet
Archive.
Accessed
August
30,
2019:
http://web.archive.org/web/∗ /http://rcd-algerie.net/
Rally for Culture and Democracy. 2019b. “Rassemblement pour la Culture et la
Démocratie.”
Internet
Archive.
Accessed
August
30,
2019:
http://web.archive.org/web/∗ /www.rcd-algerie.org
Rally of Democratic Forces. 2019. “Rassemblement des Forces Démocratiques.” Internet
Archive.
Accessed September 20, 2019:
http://web.archive.org/web/∗
/www.rfd-mauritanie.org
361
Rama, Martin. 1998. “How Bad Is Unemployment in Tunisia? Assessing Labor Market
Efficiency in a Developing Country.” The World Bank Research Observer 13(1): 59-77.
Ranko, Annette. 2015. The Muslim Brotherhood and its Quest for Hegemony in Egypt
State-Discourse and Islamist Counter-Discourse. Wiesbaden, Germany: Springer VS.
Ranko, Annette. 2015. The Muslim Brotherhood and its Quest for Hegemony in Egypt:
State-Discourse and Islamist Counter-Discourse. Wiesbaden: Springer Fachmedien
Wiesbaden. Accessed April 3, 2019: www.springer.com
Rasler, Karen. 1996. “Concessions, Repression, and Political Protest in the Iranian
Revolution.” American Sociological Review 61(1): 132-152.
Rauch. 2019. “Muhammad VI.” Encyclopedia Britannica. Accessed November 15, 2019:
www.britannica.com
Recknagel, Charles. 2012. “Is Libya Sliding Toward Civil War?” January 5. Radio Free
Europe. Accessed June 19, 2019: www.rferl.org
Redden, Jack. 1993. “Islamic Party loses ground in Jordan election.” November 9. Reuters.
Accessed February 4, 2020: www.dowjones.com/products/factiva
Reporters Sans Frontières. 2008. “Press law amendments hailed but journalists still face jail
and websites risk closure.” July 3. Accessed November 6, 2019: https://rsf.org
Reporters Sans Frontières. 2011a. “Bloodbath in Yemen, violence throughout the region.”
September 27. Updated January 20, 2016. Accessed September 30, 2019: https://rsf.org
Reporters Sans Frontières. 2011b. “Egypt.” March 11. Accessed September 27, 2019:
https://rsf.org
Reporters Sans Frontières. 2011c. “Upheaval in the Arab world: Media as Key Witnesses
and Political Pawns.” November. Accessed September 30, 2019: https://rsf.org
Reporters Sans Frontières. 2012. “Bahrein.” March 12. Accessed November 16, 2019:
https://rsf.org
362
Reporters Sans Frontières. 2014. “Houthi rebels bombard state broadcaster in Sanaa.”
September 23. Accessed October 3, 2019: https://rsf.org
Reporters Sans Frontières. 2017. . “Egyptian websites try to resist blocking.” July 13.
Accessed February 14, 2020: https://rsf.org
Reporters Sans Frontières. 2019. “Reporters Sans Frontières.” Accessed Accessed September
20, 2019: https://rsf.org
Reuters. 1987a. “Hedi Baccouche Becomes New Tunisian Prime Minister.” November 7.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Reuters. 1987b. “Bourguiba Aides Expelled from Tunisian Party Leadership.” November
18. Accessed March 11, 2020: www.dowjones.com/products/factiva
Reuters. 1987c. “Tunisian President Sacks Ruling Party’s Old Guard.” December 8.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Reuters. 1988. “Tunisian President Reshuffles Ruling Party Leadership.” August 1.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Reuters. 1989a. “Algerian Berbers Take First Step Toward Political Party.” February 12.
Accessed July 13, 2019: www.dowjones.com/products/factiva
Reuters. 1989b. “The New Cabinet, Formed on April 11, 1989, is as Follows.” April 12.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Reuters. 1990a. “About 50,000 Berbers Demonstrated in Central Algiers.” January 25.
Accessed July 13, 2019: www.dowjones.com/products/factiva
Reuters. 1990b. “Algerian Court Ends Seizure of Ben Bella Party Magazine.” July 14.
Accessed July 24, 2019: www.dowjones.com/products/factiva
Reuters. 1990c. “Algerians Demonstrate to Support Ex-President Ben Bella.” June 20.
Accessed May 6, 2019: www.dowjones.com/products/factiva
Reuters. 1990d. “Ben Bella Supporters Stage Hunger Strike in Algeria.” July 7. Accessed
363
July 24, 2019: www.dowjones.com/products/factiva
Reuters. 1990e. “Ben Bella’s Party Launches Newspaper in Algeria.” July 12. Accessed
July 24, 2019: www.dowjones.com/products/factiva
Reuters. 1990f. “Tunisian Authorities Gives [sic] Islamists Right to Publish.” January 8.
Accessed July 16, 2019: www.dowjones.com/products/factiva
Reuters. 1990g. “Tunisian Moslems Complain of Press Obstacles.” March 13. Accessed July
16, 2019: www.dowjones.com/products/factiva
Reuters. 1990h. “Tunisia Suspends Islamist Newspaper for Three Months.” June 22.
Accessed July 16, 2019: www.dowjones.com/products/factiva
Reuters. 1991a. “500 Somali Refugees Try Their Luck in Aden.” March 11. Accessed
September 30, 2019: www.dowjones.com/products/factiva
Reuters. 1991b. “Algerian Military Bans Two Fundamentalist Publications.” August 19.
Accessed August 5, 2019: www.dowjones.com/products/factiva
Reuters. 1991c. “Morocco Bans Mass Pro-Iraqi Demonstration in Casablanca.” February
14. Accessed July 26, 2019: www.dowjones.com/products/factiva
Reuters. 1991d. “Morocco Offers to Legalise Moslem Party on Certain Conditions.”
September 10. Accessed March 29, 2019: www.dowjones.com/products/factiva
Reuters. 1991e. “Moslem Fundamentalists March Through Casablanca Streets.” August 19.
Accessed July 26, 2019: www.dowjones.com/products/factiva
Reuters. 1991f. “Islamic Militants Organise Protests in Central Tunis.” January 2. Accessed
March 10, 2020: www.dowjones.com/products/factiva
Reuters. 1992. “Algeria’s Former Rulers Told to Return Buildings.” May 5. Accessed May
11, 2019: www.dowjones.com/products/factiva
Reuters. 1993. “Algerian Leader Escapes Apparent Assassination Bid.” February 13.
Accessed November 10, 2019: www.dowjones.com/products/factiva
364
Reuters. 1994a. “Algerian FLN leader to launch daily newspaper.” April 27. Accessed
August 26, 2019: www.dowjones.com/products/factiva
Reuters. 1994b. “Berbers march in Algeria, towns paralysed.” April 20. Accessed June 29,
2019: www.dowjones.com/products/factiva
Reuters. 1994c. “Deux Autres Journaux Suspendus en Algérie.” November 16. Accessed
August 7, 2019: www.dowjones.com/products/factiva
Reuters. 1994d. “Islamic fundamentalist groups active in Morocco.” Accessed March 29,
2019: www.dowjones.com/products/factiva
Reuters. 1994e. “Kuwaiti magazine urges Moslems to boycott Russia.” December 26.
Accessed August 11, 2019: www.dowjones.com/products/factiva
Reuters. 1995a. “Algeria’s Saadi opposes Moslem militants, old guard.” November 15.
Accessed June 29, 2019: www.dowjones.com/products/factiva
Reuters. 1995b. “Press Digest — Morocco.” May 23.
www.dowjones.com/products/factiva
Accessed August 17, 2019:
Reuters. 1995c. “Yemen opposition party says members being arrested.” April 13. Accessed
September 30, 2019: www.dowjones.com/products/factiva
Reuters. 1996a. “Divisions surface on Algeria’s ‘unity pact’.” September 16. Accessed
August 7, 2019: www.dowjones.com/products/factiva
Reuters. 1996b. “Yemeni tribesmen release kidnapped oilmen — paper.” January 5.
Accessed September 30, 2019: www.dowjones.com/products/factiva
Reuters. 1997a. “Algerian opposition vows more protests.” November 3. Accessed May 6
2019: www.dowjones.com/products/factiva
Reuters. 1997b. “Mauritanian opposition stages protest march.” October 31. Accessed May
23, 2019: www.dowjones.com/products/factiva
Reuters. 1997c. “Press Digest — Morocco.” October 29. Accessed August 17, 2019:
365
www.dowjones.com/products/factiva
Reuters. 1998. “Des manifestations dégénèrent à Tizi-Ouzou - Un mort.” June 27. Accessed
July 24, 2019: www.dowjones.com/products/factiva
Reuters. 2001a. “Algérie — Après la violence, le RCD quitte le gouvernement.” May 1.
Accessed July 13, 2019: www.dowjones.com/products/factiva
Reuters. 2001b. “Manifestation à Alger, troubles en petite Kabylie.” May 3. Accessed July
24, 2019: www.dowjones.com/products/factiva
Reuters. 2001c. “Morocco bans access to radical Islamists’ websites.” April 16. Accessed
October 16, 2019: www.dowjones.com/products/factiva
Reuters. 2002. “Arab tourists must avoid corrupt West—Kuwait group.” June 28. Accessed
August 11, 2019: www.dowjones.com/products/factiva
Reuters. 2005. “Sudan’s opposition parties form national alliance.” June 2. Accessed April
1, 2019: www.dowjones.com/products/factiva
Reuters. 2009. “Boycott to gift Aziz win in tainted Mauritania vote.” April 23. Accessed
November 10, 2019: www.dowjones.com/products/factiva
Reuters. 2010. “Iraq election officials confirm Sunni candidate ban.” February 13. Accessed
January 25, 2019: https://in.reuters.com
Reuters. 2011a. “Factbox: Key players in Yemen power struggle.” September 23. Accessed
September 28, 2019: www.reuters.com
Reuters. 2011b. “Kuwaiti protests on Tuesday aim to remove PM.” March 7. Accessed May
31, 2019: www.dowjones.com/products/factiva
Reuters. 2011c. “Yemen Government List.” December 18. Accessed March 24, 2019:
www.dowjones.com/products/factiva
Reuters. 2011d. “Saleh pits Yemenis against each other in bid to stay on.” March 7. Accessed
January 5, 2020: www.dowjones.com/products/factiva
366
Reuters. 2013. “Somalia’s Puntland breaks off relations with central government.” August
5. Accessed January 23, 2019: https://www.reuters.com
Reuters. 2014. “What next for Algeria after new Bouteflika hospital visit?” January 16.
Accessed December 16, 2019: www.dowjones.com/products/factiva
Reuters. 2015. “Yemen president quits, throwing country deeper into chaos.” January 22.
Accessed November 11, 2019: www.dowjones.com/products/factiva
Reuters. 2016. “Morocco’s Islamists test gains in parliamentary election.” October 6.
Accessed April 8, 2019: www.dowjones.com/products/factiva
Riani, Gala. 2010. “Libya Frees Islamists.” September 2. HIS Markit. Accessed April 2,
2019: www.dowjones.com/products/factiva
Rich, Jeremy. 2009. “Nezzar, Khaled.” Oxford Islamic Studies Online. John L. Esposito,
editor. Accessed November 15, 2019: www.oxfordislamicstudies.com
Riedel, Bruce. 2017. “Who are the Houthis, and why are we at war with them?” December
18.
Markaz blog.
Brookings Institution.
Accessed June 18, 2019:
https://www.brookings.edu/blog/markaz
Roberts, David B. 2017. “Qatar and the UAE: Exploring Divergent Responses to the Arab
Spring.” Middle East Journal 71(4): 544-562.
Rød, Espen Geelmuyden and Nils B. Weidmann. 2015. “Empowering activists or
autocrats? The Internet in authoritarian regimes.” Journal of Peace Research 52(3):
338-351.
Rohan, Brian. 2017. “Egypt censors websites as el-Sissi seeks investors.” June 13. Associated
Press. Accessed February 14, 2020: www.dowjones.com/products/factiva
Ross, Michael. 1986. “New Pressures Imperil Mubarak’s Thaw of Egyptian Media
Censorship: ‘The Official Press Does Not Report a Lot of the News, While the
Opposition Press Makes Up a Lot of It’.” April 20. Los Angeles Times. Accessed
February 4, 2020: www.latimes.com
367
Rowland, Jacky. 1992. “New Prime Minister picked after resignation.” July 9.
Guardian. Accessed May 11, 2019: www.dowjones.com/products/factiva
The
Rubin, Barry M. 2010. Guide to Islamist Movements. Armonk, NY: M. E. Sharpe.
Rugh, William A. 2004. Arab Mass Media: Newspapers, Radio, and Television in Arab
Politics. Westport, CT: Praeger.
Ryan, Curtis R. 2002. Jordan in Transition: From Hussein to Abdullah. Boulder, CO:
Lynne Rienner.
Ryan, Missy and Arshad Mohammed. 2014. “As Libya teeters near chaos, U.S. keeps
hands-off policy.” Reuters.
June 24.
Accessed September 10, 2019:
www.dowjones.com/products/factiva
Sabbagh, Rana. 1991. “Jordan Cracks Down on Militants.” October 9. Reuters. Accessed
August 9, 2019: www.dowjones.com/products/factiva
Sadiki, Larbi. 2000. “Popular Uprisings and Arab Democratization.” International Journal
of Middle East Studies 32(1): 71-95.
Sagar, D.J. 2009. Political Parties of the World. Seventh edition. London: John Harper.
Sakr, Naomi. 2003. “Freedom of Expression, Accountability and Development in the Arab
Region.” Journal of Human Development 4(1): 29-46.
Sakr, Naomi. 2013. Transformations in Egyptian Journalism. London: I. B. Tauris.
Sakthivel, Vish. 2014. “Al-Adl wal-Ihsan: Inside Morocco’s Islamist Challenge.” August.
The Washington Institute for Near East Policy.
Accessed March 29, 2019:
www.washingtoninstitute.org
Salafist Front. 2019a. “ éJ®ÊË@
mÌ '@” [Salafist Front] Internet Archive.
éîD
..
Accessed September
27, 2019: http://web.archive.org/web/∗ /GabhaSalafia.com
Salafist Front. 2019b. Personal communication with member of Salafist Front. September
368
27, 2019.
@ XAÖÏ”
Salafist Front. 2019c. “ AêËAg. Pð éJ®ÊË@ éîD.m.Ì '@ áÓ Ê®Ë@
Salafist Front of Egypt.
Facebook.
https://www.facebook.com/gabhasalafia
April 6. Official Page of the
Accessed October 1, 2019:
Saleh, Heba. 2008. “Algerian president’s power bid angers opposition.” October 30.
Accessed July 13, 2019: www.dowjones.com/products/factiva
Salisbury, Peter. 2015. “Yemen’s former president behind rise of Houthis.” March 26.
Financial Times. Accessed October 3, 2019: www.dowjones.com/products/factiva
Salmoni, Barak A., Bryce Loidolt, and Madeleine Wells. 2010. Regime and Periphery in
Northern Yemen: The Huthi Phenomenon. Santa Monica, CA: RAND.
Saouli, Adham. “History (Lebanon)” in Europa World. London: Routledge. Accessed
January 28, 2019: http://www.europaworld.com
Schedler, Andreas. 2002. “Elections Without Democracy: The Menu of Manipulation.”
Journal of Democracy 13(2): 36-50.
Schemm, Paul. 2012. “Tens of thousands at funeral for Moroccan religious opposition
leader fills capital.” December 14. Associated Press. Accessed March 29, 2019:
www.dowjones.com/products/factiva
Schemm, Paul. 2013. “Morocco’s largest Islamist opposition movement warns of unrest
without reform.” February 5.
Associated Press.
Accessed March 29, 2019:
www.dowjones.com/products/factiva
Schmidmayr, Michael. 2010. “Islamist Engagement in Contentious Politics: Kuwait and
Bahrain.” In Contentious Politics in the Middle East: Political Opposition under
Authoritarianism. Holger Albrecht, editor. Gainesville, FL: University Press of Florida.
Schmitter, Philippe C. and Nadine Sika. 2017. “Democratization in the Middle East and
North Africa: A More Ambidextrous Process?” Mediterranean Politics 22(4): 443-463.
Schmitz, Charles and Robert D. Burrowes. 2018. Historical Dictionary of Yemen. Third
369
Edition. New York: Rowman & Littlefield.
Schreck, Adam. 2013. “Nearly 10 years after US invasion of Iraq, Sunni protesters dig
in as familiar tensions flare.” February 19. Associated Press. Accessed June 22, 2019:
www.dowjones.com/products/factiva
Schwedler, Jillian. 2006. Faith in Moderation: Islamist Parties in Jordan and Yemen.
Cambridge: Cambridge University Press.
Schwedler, Jillian. 2015. “Jordan: The Quiescent Opposition.” August 27. The Wilson
Center. Accessed March 22, 2019: www.wilsoncenter.org
Seddon, David. 1986. “Riot and Rebellion: Political Responses to Economic Crisis in North
Africa, Tunisia, Morocco and Sudan.” School of Development Studies, University of East
Anglia. Discussion Paper 196.
Seddon, David. 2004. A Political and Economic Dictionary of the Middle East. New York:
Europa.
Sela, Avraham. 2002. The Continuum Encyclopedia of the Middle East. New York:
Continuum.
Selvik, Kjetil. 2011. “Elite Rivalry in a Semi-Democracy: The Kuwaiti Press Scene.” Middle
Eastern Studies 47(3): 477-496.
Shadid, Anthony. 2011. “Coalition of Factions from the Streets Fuels a New Opposition in
Syria.” July 1. Accessed June 21, 2019: : www.dowjones.com/products/factiva
Shadmehr, Mehdi and Dan Bernhardt. 2015. “State Censorship.” American Economic
Journal: Microeconomics 7(2): 280-307.
Shaib, Abdelkhalig. 2019. Vice President, Arab Banking Corporation.
communication. November 18.
Personal
Shehadi, Philip. 1989. “Algerian Fundamentalists Seek Dialogue to Avert New Violence.”
October 4. Accessed August 5, 2019: www.dowjones.com/products/factiva
370
Shehata, Samer. 2015. “Egypt: The Founders.” In “Egypt: The New Founders.” August
27. Wilson Center. Accessed June 25, 2019: www.wilsoncenter.org
Shenker, Jack. 2011. “Egyptians call for day of action to revive their ’stifled’ revolution.”
November 3. The Guardian. Accessed September 27, 2019: www.theguardian.com
Shepherd, Melinda C. 2019.
www.britannica.com
“Chadli Benjedid.” Accessed November 15, 2019:
Siddique, Haroon. 2012. “Mohamed Morsi supporters and opponents clash in Cairo.”
December
6.
The
Guardian.
Accessed
June
24,
2019:
www.dowjones.com/products/factiva
ñk B@
Sisters of the Hereafter. 2019. “ H@
èQk B@” Internet Archive. Accessed October 8, 2019:
http://web.archive.org/web/∗ /mouminate.net
Slackman, Michael. 2005. “With No Status as a Party, Egyptian Group Wields Power.”
August 16.
New York Times.
Accessed February 16,
2020:
www.dowjones.com/products/factiva
Smith, Benjamin. 2004. “Oil Wealth and Regime Survival in the Developing World, 1960—
1999.” American Journal of Political Science 48(2): 232—246.
Smith, Craig S. 2004. “Islam and Democracy: Algerians Try to Blaze a Trail.” The New
York Times. April 14. Accessed May 9, 2019: www.dowjones.com/products/factiva
Sobolev, Anton. 2019. “Can Independent Media Help Autocrats Suppress Collective
Action?” Manuscript.
Social Reform Society. 2019. “ ú«AÒJk. B@
hCB @ éJ ªÔg. @” Internet Archive. Accessed October
7, 2019: http://web.archive.org/web/∗ /Eslah.com
Socialist Union of Popular Forces. 2019a. “Libération.” Internet Archive. Accessed October
16, 2019: http://web.archive.org/web/∗ /liberation.press.ma
Socialist Union of Popular Forces. 2019b. “Union Socialiste des Forces Populaires.” Internet
Archive. Accessed October 16, 2019: http://web.archive.org/web/∗ /usfp.ma
371
St John, Ronald Bruce. 2014. Historical Dictionary of Libya. Fifth edition. New York:
Rowman & Littlefield.
St John, Ronald Bruce. 2019. “History (Libya). In Europa World Online. Routledge.
Accessed April 2, 2019: http://europaworld.com
Stein, Elizabeth A. 2008. “Leading the Way: The Media and the Struggle against
Authoritarian Rule.” Doctoral Dissertation. University of California, Los Angeles.
Stein, Elizabeth Ann. 2016. “Information and Civil Unrest in Dictatorships.” In The Oxford
Research Encylopedia of Politics. William R. Thompson, editor. New York: Oxford
University Press.
Stockmann, Daniela and Mary E. Gallagher. 2011. “Remote Control: How the Media
Sustain Authoritarian Rule in China.” Comparative Political Studies 44(4): 436—67.
Studies 17(2): 145—66.
Sudan Tribune. 2014. “Opposition, rebel forces sign joint declaration for peace and
democracy in Sudan.” December 4.
Accessed April 1,
2019:
www.dowjones.com/products/factiva
Sudan Tribune. 2018. “Al-Bashir warns Sudan’s political opposition against alliance with
armed groups.” March 23. Accessed April 1, 2019: www.dowjones.com/products/factiva
Suhail Net.
2019.
K
“I
ÉJîD ”
Internet Archive.
Accessed October 2, 2019:
http://web.archive.org/web/∗ / suhail.net
Suwaed, Muhammad. 2016. An Historical Dictionary of Political Parties in the Middle East
and North Africa. Lewiston, NY: Edwin Mellen.
Svolik, Milan M. 2012. The Politics of Authoritarian Rule. New York: Cambridge University
Press.
Sweis, Rana F. 2015. “Jordan Gives Prison Term for Criticism on Facebook.” February 15.
New York Times. Accessed February 4, 2020: www.nytimes.com
372
Swiney, Christie Flournoy. 2009. “Islamic Constitutional Movement.” In The Oxford
Encyclopedia of the Islamic World. John L. Esposito, editor. Accessed May 28, 2019:
www.oxfordislamicstudies.com
Sylla, Ibrahima. 1996. “New party wins token opposition seat in Mauritania.” October 19.
Reuters. Accessed May 23, 2019: www.dowjones.com/products/factiva
Tachau, Frank. 1994. Political Parties of the Middle East and North Africa. Westport, CT:
Greenwood.
Targeted News Service. 2010. “Egypt Urged to Protect Peaceful Demonstrators.” April 15.
Accessed July 26, 2019: www.dowjones.com/products/factiva
Tarrow, Sidney. 1994. Power in Movement: Social Movements and Contentious Politics.
New York: Cambridge University Press.
Tessler, Mark A., John P. Entelis, and Gregory W. White. 1995. “Kingdom of Morocco.”
In The Government and Politics of the Middle East and North Africa. David E. Long
and Bernard Reich, editors. San Francisco: Westview. Pages 369-393.
Thielmann, Jörn. 2014. “Islamic Salvation Front.” In The Oxford Encyclopedia of Islam
and Politics. Emad El-Din Shahin, editor. Oxford University Press. Accessed March 11,
2020: www.oxfordreference.com
Thorne, John. 2006. “Islamists making inroads in Morocco’s universities.” May 31.
Associated Press. Accessed April 7, 2019: www.dowjones.com/products/factiva
Toorani, Noor and Sandeep Singh Grewal. 2010. “Newsletter Ban for Groups.” September
21. Gulf Daily News. Accessed August 8, 2019: www.dowjones.com/products/factiva
Touati, Abdemalek. 1999. “FOCUS-Ex finance minister heads new Algerian govt.”
December 23. Reuters. Accessed May 11, 2019: www.dowjones.com/products/factiva
Trager, Eric. 2011. “The Unbreakable Muslim Brotherhood: Grim Prospects for a Liberal
Egypt.” Foreign Affairs (September-October).
Accessed March 28, 2019:
www.washingtoninstitute.org
373
Trager, Eric. 2013. “How Egypt’s Muslim Brotherhood Can Bounce Back.” The Atlantic.
November 5. Accessed March 28, 2019: www.dowjones.com/products/factiva
Trager, Eric. 2016. Arab Fall: How the Muslim Brotherhood Won and Lost Egypt in 891
Days. Washington, DC: Georgetown University Press.
Trend News Agency. 2008. “New Kuwait cabinet to meet amid Islamist anger.” May 29.
Accessed May 27, 2019: www.dowjones.com/products/factiva
Tsfati, Yariv and Gal Ariely. 2014. “Individual and Contextual Correlates of Trust in Media
Across 44 Countries.” Communication Research 41(6): 760-782.
Tufekci, Zeynep and Christopher Wilson. 2012. “Social Media and the Decision to
Participate in Political Protest: Observations From Tahrir Square.” Journal of
Communication 62: 363—379.
Tunis Afrique Presse. 2011. “Tunis plays host to international conference on trade unions
in
democratic
transition.”
May
8.
Accessed
June
26,
2019:
www.dowjones.com/products/factiva
Turner, Barry. 2015. “Oman.” In The Statesman’s Yearbook 2015: The Politics, Culture
and Economies of the World. New York: Palgrave Macmillan.
U.S. Bureau of Economic Analysis. 2019. “Gross Domestic Product: Implicit Price
Deflator [GDPDEF].” Federal Reserve Bank of St. Louis. Accessed December 8, 2019:
https://fred.stlouisfed.org
U.S. Central Intelligence Agency. 2019a. “Bahrain.” In The World Factbook. Accessed June
7, 2019: https://www.cia.gov/library/publications/the-world-factbook
U.S. Central Intelligence Agency. 2019b. “Yemen.” In The World Factbook. Accessed April
3, 2019: https://www.cia.gov/library/publications/the-world-factbook
U.S. Central Intelligence Agency. 2019c. “Mauritania.” In The World Factbook. Accessed
August 16, 2019: https://www.cia.gov/library/publications/the-world-factbook
U.S. Central Intelligence Agency. 2019d. “Syria” in The World Factbook. Accessed January
374
23, 2019: https://www.cia.gov/library/publications/the-world-factbook
U.S. Department of State. 1991. “Country Reports on Human Rights Practices for 1990.”
February. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1992. “Country Reports on Human Rights Practices for 1991.”
February. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1993. “Country Reports on Human Rights Practices for 1992.”
February. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1994. “Country Reports on Human Rights Practices for 1994.”
February. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1995. “Country Reports on Human Rights Practices for 1994.”
February. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1996. “Country Reports on Human Rights Practices for 1995.”
April. Hathi Trust. Accessed October 8, 2019: https://catalog.hathitrust.org
U.S. Department of State. 1997. “Country Reports on Human Rights Practices for 1996.”
January 30. Accessed August 26, 2019: https://1997-2001.state.gov
U.S. Department of State. 1998. “Country Reports on Human Rights Practices for 1997.”
January 30. Accessed August 26, 2019: https://1997-2001.state.gov
U.S. Department of State. 1999. “Country Reports on Human Rights Practices for 1998.”
February 26. Accessed August 26, 2019: https://1997-2001.state.gov
U.S. Department of State. 2000. “Country Reports on Human Rights Practices for 1999.”
February 25. Accessed August 26, 2019: https://1997-2001.state.gov
U.S. Department of State. 2001. “Country Reports on Human Rights Practices for 2000.”
February 23. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2002. “Country Reports on Human Rights Practices for 2001.”
March 4. Accessed August 26, 2019: https://2001-2009.state.gov
375
U.S. Department of State. 2003. “Country Reports on Human Rights Practices for 2002.”
March 31. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2004. “Country Reports on Human Rights Practices for 2003.”
February 25. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2005. “Country Reports on Human Rights Practices for 2003.”
February 28. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2006. “Country Reports on Human Rights Practices for 2005.”
March 8. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2007. “Country Reports on Human Rights Practices for 2006.”
March 6. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2008. “Country Reports on Human Rights Practices for 2007.”
March 11. Accessed August 26, 2019: https://2001-2009.state.gov
U.S. Department of State. 2009. “Country Reports on Human Rights Practices for 2008.”
February 25. Accessed August 28, 2019: https://2009-2017.state.gov
U.S. Department of State. 2010. “Country Reports on Human Rights Practices for 2009.”
March 11. Accessed August 28, 2019: https://2009-2017.state.gov
U.S. Department of State. 2011a. “Background Notes: Sudan.” July 7. Accessed June 17,
2019: www.dowjones.com/products/factiva
U.S. Department of State. 2011b. “Country Reports on Human Rights Practices for 2010.”
April 8. Accessed August 28, 2019: https://2009-2017.state.gov
U.S. Department of State. 2012. “Country Reports on Human Rights Practices for 2011.”
May 24. Accessed August 29, 2019: https://2009-2017.state.gov
U.S. Department of State. 2013. “Country Reports on Human Rights Practices for 2012.”
April 19. Accessed August 29, 2019: https://2009-2017.state.gov
U.S. Department of State. 2014. “Country Reports on Human Rights Practices for 2013.”
376
February 27. Accessed August 29, 2019: https://2009-2017.state.gov
U.S. Department of State. 2015. “Country Reports on Human Rights Practices for 2014.”
June 25. Accessed August 29, 2019: https://2009-2017.state.gov
U.S. Department of State. 2016. “Country Reports on Human Rights Practices for 2015.”
April 13. Accessed August 29, 2019: https://2009-2017.state.gov
U.S. Department of State. 2017. “Country Reports on Human Rights Practices for 2016.”
April 20. Accessed August 29, 2019: www.state.gov
U.S. Department of State. 2018. “Country Reports on Human Rights Practices for 2017.”
March 3. Accessed August 29, 2019: www.state.gov Unitary National Democratic
Assemblage. 2019.
U.S. Energy Information Administration.
2011.
“Global natural gas prices vary
considerably.” September 30. Accessed February 20, 2020: www.eia.gov
U.S. Energy Information Administration. 2019. “Frequently Asked Questions.” Accessed
December 8, 2019: www.eia.gov
Umma National Party. 2019. “ úG@XñË@
éÓ B@ H Qk”
ú×ñ®Ë@
.
Internet Archive. Accessed
October 18, 2019: http://web.archive.org/web/∗ /umma.org
Valeri, Marc. 2018. “Islamist political societies in Bahrain: collateral victims of the 2011
Popular Uprising.” In Islamists and the Politics of the Arab Uprisings. Hendrik
Kraetzschmar and Paola Rivetti, editors. Edinburgh: Edinburgh University Press.
Valeri, Marc. 2019. “History (Oman).” In Europa World Online. Routledge. Accessed
March 26, 2019: http://europaworld.com
Vanderhill, Rachel. 2018. “The Internet, Regime Breakdown, and Democratization: Lessons
from Tunisia.” June 4. Journal of International Affairs blog. Columbia University.
Accessed June 26, 2019: https://jia.sipa.columbia.edu
Varadarajan, Tunku. 1996. “Islamic hardliners test tolerance of Morocco.” December 23.
Accessed March 29, 2019: www.dowjones.com/products/factiva
377
Voice of America. 2008. “Repression of Pro-Democracy Demonstrations Raises Tension in
Mauritania.” October 8. Accessed May 21, 2019: www.dowjones.com/products/factiva
Voice of America. 2009a. “Mauritania Military and Politicians Agree to Postpone Saturday’s
Election.” June 3. Accessed November 10, 2019: www.dowjones.com/products/factiva
Voice of America. 2009b. “Power-Sharing Deal Signed in Mauritania.” June 27. Accessed
November 10, 2019: www.dowjones.com/products/factiva
Walgrave, Stefaan and Jan Manssens. 2000. “The Making of the White March: The Mass
Media as a Mobilizing Alternative to Movement Organizations.” Mobilization: An
International Journal 5(2): 217-239.
Waltz, Susan and Joseph A. Kéchichian. 2009. “Hizb al-Nahda.” In The Oxford
Encyclopedia of the Islamic World. John L. Esposito, editor. Accessed March 24, 2019:
www.oxfordislamicstudies.com
Waltz, Susan E. 1995. Human Rights and Reform: Changing the Face of North African
Politics. Berkeley, CA: University of California Press.
Wantchekon, Leonard. 2002. “Why Do Resource Abundant Countries Have Authoritarian
Governments?” Journal of African Finance and Economic Development 2: 57—77.
Wedeman, Ben. 2012. “Salafis call for Islamic law in Egypt protest.” November 9. CNN.
Accessed June 25, 2019: https://edition.cnn.com
Wegemund, Regina. 2003. “Mauritania.” In Elections in Africa: A Data Handbook. Dieter
Nohlen, Bernard Thibaut, and Michael Krennerich, editors. Oxford Scholarship Online.
Accessed May 23, 2019: http://www.oxfordscholarship.com
Whitten-Woodring, Jenifer and Douglas A. Van Belle. 2014. Historical Guide to World
Media Freedom: A Country-by-Country Analysis. Washington, DC: CQ Press.
Whitten-Woodring, Jenifer and Douglas A. Van Belle. 2015. “The Correlates of Media
Freedom: An Introduction of the Global Media Freedom Dataset.” Â Political Science
Research and Methods, FirstView Article, December.
378
Whois. 2019a. “al-islah.net” Accessed October 3, 2019: www.whois.com
Whois. 2019b. “aljamaa.com” Accessed October 8, 2019: www.whois.com
Whois. 2019c. “aljamaa.info” Accessed October 8, 2019: www.whois.com
Whois. 2019d. “aljamaa.net” Accessed October 8, 2019: www.whois.com
Whois. 2019e. “aljamaa.org” Accessed October 13, 2019: www.whois.com
Whois. 2019f. “alsahwa-yemen.net” Accessed September 30, 2019: www.whois.com
Whois. 2019g. “chahed.tv” Accessed October 16, 2019: www.whois.com
Whois. 2019h. “elmourabiton.tv” Accessed September 21, 2019: www.whois.com
Whois. 2019i. “eslah.com” Accessed October 7, 2019: www.whois.com
Whois. 2019j. “ffs.dz” Accessed August 22, 2019: www.whois.com
Whois. 2019k. “kabyleuniversel.com” Accessed August 22, 2019: www.whois.com
Whois. 2019l. “mouminate.net” Accessed October 16, 2019: www.whois.com
Whois. 2019m. “nadiayassine.net” Accessed October 16, 2019: www.whois.com
Whois. 2019n. “umma.org” Accessed October 18, 2019: www.whois.com
Whois. 2019o. “yassine.net” Accessed October 16, 2019: www.whois.com
Whois. 2019p. “yemenshabab.net” Accessed October 2, 2019: www.whois.com
Willis, Michael. 1999. “Between alternance and the Makzhen: At-Tawhid wa Al-Islah’s
entry into Moroccan Politics.” The Journal of North African Studies 4(3): 45-80.
379
Willis, Michael. 1999. The Islamist Challenge in Algeria: A Political History. New York:
New York University Press.
Willis, Michael. 2007. “Justice and development or justice and spirituality? : the challenge
of Morocco’s nonviolent Islamist movements.” In The Maghrib in the new century:
identity, religion, and politics. Bruce Maddy-Weitzman and Daniel Zisenwine, editors.
Gainesville, FL: University Press of Florida. 150-174.
Willis, Michael. 2009. “Islamic Movements in Algeria, Morocco, and Tunisia.” In The
Oxford Handbook of Islam and Politics. John L. Esposito, editor. Accessed April 7,
2019: www.oxfordislamicstudies.com
Willis, Michael. 2012. Power and Politics in the Maghreb: Algeria, Tunisia, and Morocco
from Independence to the Arab Spring. London: Hurst & Company.
Wintrobe, Donald. 1998. The Political Economy of Dictatorship. New York: Cambridge
University Press.
Woodside, Duncan. 2019. “Recent History (Sudan).” In Europa World Online. Routledge.
Accessed April 1, 2019: http://europaworld.com
Woodward, Peter. 2009. “Turabi, Hasan al-.” In The Oxford Encyclopedia of the Islamic
World.
John L. Esposito, editor.
Accessed November 11, 2019:
www.oxfordislamicstudies.com
World Association of Newspapers and News Publishers (WAN-IFRA). 2010. World Press
Trends 2010. WAN-IFRA/ZenithOptimedia. Accessed February 14, 2020: www.wanifra.org
World Bank, The. 2019. “World Development Indicators.” Accessed November 28, 2019:
https://databank.worldbank.org
Wright, Jonathan. 1989. “Tunisian Moslems Clear First Election Hurdle.” March 11.
Accessed March 11, 2020: www.dowjones.com/products/factiva
Wright, Jonathan. 1990a. “Tunis Government Attacked by Students, Workers and Parties.”
March 1. Reuters News. Accessed March 9, 2020: www.dowjones.com/products/factiva
380
Wright, Jonathan. 1990b. “Tunisian Islamists Bring Out Newspaper After Long Fight.”
April 21. Reuters News. Accessed July 16, 2019: www.dowjones.com/products/factiva
Wright, Jonathan. 1990c. “Tunisian Leftists Announce Common Manifesto for Broad
Front.”
April
16.
Reuters.
Accessed
March
11,
2020:
www.dowjones.com/products/factiva
Wright, Joseph, Erica Frantz, and Barbara Geddes. 2013. “Oil and Autocratic Regime
Survival.” British Journal of Political Science 45: 287-306.
Yadav, Stacey Philbrick. 2014. “The limits of the ‘sectarian’ framing in Yemen.”
September 25.
Monkey Cage blog.
Accessed September 28, 2019:
www.washingtonpost.com/news/monkey-cage
Yassine, Abdesslam. 2019. “ áAK
” [School of the Imam of
ÐCË@ YJ.« XYj.ÖÏ @ ÐAÓB @ éPYÓ
Renewal Abdesslam Yassine] Internet Archive.
http://web.archive.org/web/∗ /yassine.net
Accessed October 9,
2019:
Yassine, Nadia. 2019. “Nadia Yassine.” Internet Archive. Accessed October 16, 2019:
http://web.archive.org/web/∗ /nadiayassine.net
Yemen Post. 2011. “Government Raids Suhail TV Station and Newspaper.” May 26.
Accessed October 2, 2019: http://yemenpost.net
Yemen Shabab. 2019. “ H
. AJ.
áÖß èAJ¯” [Yemen Shabab (Youth) Channel] Internet Archive.
Accessed October 2, 2019: http://web.archive.org/web/∗ /yemenshabab.net
Yemen Times. 2011. “Demonstrations and protests in Taiz.” January 27. Accessed July 27,
2019: www.dowjones.com/products/factiva
Yemen Times. 2015. “Al-Arabiya and Al-Jazeera added to list of banned websites.” April
8. Accessed October 2, 2019: www.dowjones.com/products/factiva
Youssef, Maamoun. 1993. “Yemenis scorn cleaning jobs in garbage-heaped Sanaa.” May 4.
Reuters News. Accessed September 30, 2019: www.dowjones.com/products/factiva
Zaccara, Luciano, Courtney Freer, and Hendrik Kraetzschmar. 2018. “Kuwait’s Islamist
381
proto-parties and the Arab uprisings: between opposition, pragmatism and the pursuit
of cross-ideological cooperation.” In Islamists and the Politics of the Arab Uprisings.
Hendrik Kraetzschmar and Paola Rivetti, editors. Edinburgh: Edinburgh University
Press.
Zahid, Mohammed. 2012. The Muslim Brotherhood and Egypt’s Succession Crisis: The
Politics of Liberalisation and Reform in the Middle East. London: I.B. Tauris.
Zartman, I. William. 1988. “Opposition as Support of the State.” In Beyond coercion : the
durability of the Arab state. Adeed Dawisha, editor. London: Croom Helm.
Zhang, Chuchu. 2019. Islamist Party Mobilization: Tunisia’s Ennahda and Algeria’s HMS
Compared, 1989-2014. New York: Springer.
Zisser, Eyal. 2019. “History (Syrian Arab Republic)” in Europa World. London: Routledge.
Accessed January 23, 2019: http://www.europaworld.com
Zweiri, Mahjoob. 2007. “The Victory of Al Wefaq: The Rise of Shiite Politics in Bahrain.”
April. Research Institute for European and American Studies. Research Paper No. 108.
382