Democracy in Hard Places Scott

Mainwaring
Visit to download the full and correct content document:
https://ebookmass.com/product/democracy-in-hard-places-scott-mainwaring/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Democracy Inside: Participatory Innovation in Unlikely

Places Albert W. Dzur

https://ebookmass.com/product/democracy-inside-participatory-
innovation-in-unlikely-places-albert-w-dzur/

Hard Evidence: Case Studies in Forensic Anthropology

https://ebookmass.com/product/hard-evidence-case-studies-in-
forensic-anthropology/

Shaping American Democracy: Landscapes and Urban Design

1st Edition Scott M. Roulier (Auth.)

https://ebookmass.com/product/shaping-american-democracy-
landscapes-and-urban-design-1st-edition-scott-m-roulier-auth/

Hard Pass Tb Mann

https://ebookmass.com/product/hard-pass-tb-mann/
Destinations in Mind: Portraying Places on the Roman
Empire's Souvenirs Kimberly Cassibry

https://ebookmass.com/product/destinations-in-mind-portraying-
places-on-the-roman-empires-souvenirs-kimberly-cassibry/

Human Geography: Places and Regions in Global Context –

Ebook PDF Version

https://ebookmass.com/product/human-geography-places-and-regions-
in-global-context-ebook-pdf-version/

Night in Passchendaele Scott Bennett

https://ebookmass.com/product/night-in-passchendaele-scott-
bennett/

Hard Candy Heartthrob Lorelei M. Hart

https://ebookmass.com/product/hard-candy-heartthrob-lorelei-m-
hart/

Innovation in Real Places: Strategies for Prosperity in

an Unforgiving World 1st Edition Breznitz

https://ebookmass.com/product/innovation-in-real-places-
strategies-for-prosperity-in-an-unforgiving-world-1st-edition-
breznitz/
Democracy in Hard Places
Democracy in Hard Places
Edited by
SCOTT MAINWARING AND TAREK MASOUD
 Oxford University Press is a department of the University of Oxford. It furthers the
University’s objective of excellence in research, scholarship, and education by publishing
worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and
certain other countries.
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2022

All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, without the prior permission in
writing of Oxford University Press, or as expressly permitted by law, by license, or under
terms agreed with the appropriate reproduction rights organization. Inquiries concerning
reproduction outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same
condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Mainwaring, Scott, 1954– editor. | Masoud, Tarek E., editor.
Title: Democracy in hard places / [edited by] Scott Mainwaring, Tarek Masoud.
Description: New York, NY : Oxford University Press, 2022. |
Includes bibliographical references and index.
Identifiers: LCCN 2022006297 (print) | LCCN 2022006298 (ebook) |
ISBN 9780197598764 (paperback) | ISBN 9780197598757 (hardback) |
ISBN 9780197598788 (epub)
Subjects: LCSH: Democracy—Case studies. | Democratization—Case studies. |
World politics—21st century—Case studies.
Classification: LCC JC423 .D381319 2022 (print) | LCC JC423 (ebook) |
DDC 321.8—dc23/eng/20220322
LC record available at https://lccn.loc.gov/2022006297
LC ebook record available at https://lccn.loc.gov/2022006298
DOI: 10.1093/oso/9780197598757.001.0001
 Contents

List of Figures
List of Tables
Acknowledgments
Contributors
Abbreviations

1. Introduction: Democracy in Hard Places

Tarek Masoud and Scott Mainwaring
2. India’s Democratic Longevity and Its Troubled Trajectory
Ashutosh Varshney
3. The Politics of Permanent Pitfalls: Historical Inheritances and
Indonesia’s Democratic Survival
Dan Slater
4. Africa’s Democratic Outliers: Success amid Challenges in Benin
and South Africa
Rachel Beatty Riedl
5. Georgia, Moldova, and Ukraine: Democratic Moments in the
Former Soviet Union
Lucan Ahmad Way
6. The Puzzle of Timor-Leste
Nancy Bermeo
7. Economic Crises, Military Rebellions, and Democratic Survival:
Argentina, 1983–2021
Scott Mainwaring and Emilia Simison
8. Why Democracies Survive in Hard Places
Scott Mainwaring
References
Index
 List of Figures

1.1. Democracy in the world, 1900–present

1.2. Liberal democracy scores of the nine cases in this volume,
1974–2017
1.3. Marginal effects of continuous predictors
1.4. Marginal effects of qualitative predictors
1.5. India (1978–2017), probability of democratic breakdown
(benchmarked against Spain)
1.6. Indonesia (2000–2017), probability of democratic breakdown
(benchmarked against Spain)
1.7. Benin (1992–2017), probability of democratic breakdown
(benchmarked against Spain)
1.8. South Africa (1996–2017), probability of democratic
breakdown (benchmarked against Spain)
1.9. Georgia (2005–2017), probability of democratic breakdown
(benchmarked against Spain)
1.10. Ukraine (1995–1997, 2007–2013), probability of democratic
breakdown (benchmarked against Spain)
1.11. Moldova (1995–2004, 2010–2017), probability of democratic
breakdown (benchmarked against Spain)
1.12. Timor-Leste (2002–2017), probability of democratic
breakdown (benchmarked against Spain)
1.13. Argentina (1984–2017), probability of democratic breakdown
(benchmarked against Spain)
2.1. Electoral and liberal democracy index, India 1950–2020
2.2. The electoral-liberal democracy gap—decadal comparison
2.3. Electoral democracy index: India and some world regions,
1950–2020
2.4. Liberal democracy index: India and some world regions,
1950–2020
2.5. Electoral democracy index: India compared with selected
countries, 1950–2020
2.6. Liberal democracy index: India compared with selected
countries, 1950–2020
3.1. Indonesian economic growth rates, 1961–2019
3.2. Indonesian economic development levels, 1960–2019
3.3. Indonesia’s liberal democracy score over time, 1953–2019
4.1. Regime transition pathways
4.2. Pre-transition economic decline in South Africa and Benin,
1980–1994
4.3. V-Dem liberal democracy index rankings in South Africa and
Benin, 1960–2019
6.1. Transition year GDP vs. mean for existing democracies, 1945–
2015
6.2. Polity scores for former Portuguese colonies, 1975–2015
6.3. Polity scores for post-1989 Asian democracies, 1990–2017
6.4. Civil liberties for Southeast Asia and India, 1990–2017
6.5. Liberal democracy for Southeast Asia and India, 1990–2017
6.6. Conflict history and democratic durability, Kaplan–Meier Plot
6.7. Property rights in Timor-Leste vs. Indonesia, 1975–2015
7.1. Graphic summary of the argument
7.2. Argentina, V-Dem liberal democracy and Freedom House
scores, 1983–2020
7.3. V-Dem illiberalism scores—FPV-PJ, UCR, and PRO, 1983–2019
8.1. Governing party illiberalism scores
8.2. Percentage of national assembly seats controlled by the
governing party and coalition
 List of Tables

1.1. Regression Analysis of Correlates of Democratic Breakdown

1.2. Values for Each Case on Key Independent Variables
(Socioeconomic Indicators Averaged over Democratic
Lifespan)
6.1. Aid, Peacekeeping, and Democratic Durability
8.1. Illiberalism Scores of Governing Parties and Regime Outcomes
8.2. Weighted Legislative Powers Scores for Nine Countries
8.3. Expert Perception of Judicial Capacity to Constrain
Governments, V-Dem (2019)
8.4. World Bank Governance Indicators: Perceptions of Control of
Corruption and Rule of Law in Nine Hard Cases, 2019
8.5. Perception of Government Effectiveness and Per Capita GDP
Growth in Nine Hard Cases
 Acknowledgments

This book is about why democracy sometimes survives for a long

time in difficult conditions. When we began planning the conference
that led to the book, the COVID-19 pandemic and the January 6
insurrection at the United States Capitol would have been
unthinkable to all but the most determined pessimists, but it was
already a time of grave worry about the state of democracy in the
world. We hope that this book not only enriches our understanding
of what makes democracy in hard places possible, but also that it
might inspire belief during these hard times that democracy can
persist in the face of grave challenges.
The idea for this volume grew out of many conversations we had
from 2016 to 2019 at Harvard University. We had neighboring
offices, co-taught a course on “Getting and Keeping Democracy” in
2017 and in 2018 at Harvard’s John F. Kennedy School of
Government, and co-directed a program on “Democracy in Hard
Places” at the Kennedy School’s Ash Center for Democratic
Governance and Innovation. During these years of working together,
we often disagreed with one another—but in so doing, we always
pushed each other to new understandings and insights. We also
accrued a number of debts that we are pleased to acknowledge
here.
Our first debt is to Tony Saich, the director of the Ash Center.
Tony and the Center generously supported our work from the
beginning. This support made possible the eponymous conference in
May 2019 that gave rise to this book, and also enabled us to secure
the editorial and administrative assistance necessary to bring the
book to fruition. Special thanks must be rendered to Melissa D’Anello
and Maureen Griffin of the Ash Center, without whom neither the
conference nor the book would have happened. We also thank our
assistants at the Kennedy School, Juanne Zhao and Sari Betancourt.
Our debts were not just financial and administrative, but
intellectual. We, and the contributors to this volume, were fortunate
to benefit from the insights of an outstanding group of
commentators and participants in the May 2019 conference on
Democracy in Hard Places, including Eva Bellin, Fernando Bizzarro,
Melani Cammett, Timothy Colton, Steven Fish, Candelaria Garay,
Frances Hagopian, Sarah Hummel, Steven Levitsky, Pia Raffler, and
Deborah Yashar. We are also grateful to Harvard colleagues Daniel
Ziblatt, Richard Zeckhauser, Gautam Nair, Moshik Temkin, Alex
Keyssar, and Archon Fung for informal conversations that shaped our
thinking about democracy and its survival. Ashutosh Varshney, the
author of the chapter on India, deserves special thanks for being an
invigorating intellectual presence at the Ash Center during a critical
phase of this book’s development.
We thank David McBride of Oxford University Press for his
enthusiastic support for the project from the outset. And we record
our thanks to María Victoria De Negri who, with characteristic talent
and attention to detail, prepared the book for publication and
compiled the index. Fernando Bizzarro provided helpful research.
Finally, Scott thanks his wonderful partner, Sue Elfin, for gracing
his life. Tarek thanks his long-suffering partner, Kristin Alcorn
Masoud, for tolerating his frequent failures to bring grace to hers.
The authors dedicate this book to the memory of Alfred C.
Stepan, whose fingerprints are on every chapter.
 Contributors

Nancy Bermeo (PhD Yale University) is currently a Nuffield Senior

Research Fellow at Oxford University and a PIIRS Senior Scholar at
Princeton University. She writes on the causes and consequences of
political mobilization and regime change as well as the quality of
democracy. Her books include an award-winning study of the
breakdown of democracy titled Ordinary People in Extraordinary
Times (Princeton University Press), and, with Deborah Yashar,
Parties, Movements and Democracy in the Developing World
(Cambridge University Press), plus Mass Politics in Tough Times with
Larry Bartels (Oxford University Press), and Coping with Crisis:
Government Reactions to the Great Recession with Jonas Pontusson
(Russell Sage Foundation). Her latest book, Democracy after War, is
forthcoming from Princeton University Press.
Scott Mainwaring (PhD, Stanford) is the Conley Professor of
Political Science at the University of Notre Dame. His research and
teaching focus on democratization, party systems, and Latin
American politics. His book with Aníbal Pérez-Liñán, Democracies
and Dictatorships in Latin America: Emergence, Survival, and Fall
(Cambridge University Press, 2013) won the Best Book Prizes of the
Democracy and Autocracy section of the American Political Science
Association and the Political Institutions section of the Latin
American Studies Association. Mainwaring was elected to the
American Academy of Arts and Sciences in 2010. In April 2019, PS:
Political Science and Politics listed him as one of the fifty most cited
political scientists in the world. He served as the Jorge Paulo Lemann
Professor for Brazil Studies and as faculty co-chair of the Brazil
Studies program at Harvard University from 2016 to 2019.
Tarek Masoud (PhD Yale University) is the Ford Foundation
Professor of Democracy and Governance at Harvard University’s John
F. Kennedy School of Government. He is the co-Editor of the Journal
of Democracy, the director of the Kennedy School’s Middle East
Initiative, the Initiative on Democracy in Hard Places, and the co-
author of, among other works, The Arab Spring: Pathways of
Repression and Reform (Oxford University Press, 2015).
Rachel Beatty Riedl (PhD Princeton University) is the John S.
Knight Professor of International Studies, Director of the Einaudi
Center for International Studies, and a professor in the Department
of Government at Cornell University. Her research interests include
institutional development in new democracies, local governance and
decentralization policy, authoritarian regime legacies, and religion
and politics, with a regional focus in Africa. She is the author of
award-winning Authoritarian Origins of Democratic Party Systems in
Africa (Cambridge University Press, 2014) and co-author with
Gwyneth McClendon of From Pews to Politics: Religious Sermons and
Political Participation in Africa (Cambridge University Press, 2019).
Riedl is co-host of the podcast Ufahamu Africa, featuring weekly
episodes of news highlights and interviews about life and politics on
the African continent.
Emilia Simison (MA Torcuato Di Tella University) is a PhD
candidate in Political Science at the Massachusetts Institute of
Technology. Her research focuses on the comparative political
economy of policymaking and policy change, especially on how
political institutions in democratic and authoritarian regimes shape
the extent to which citizens and interest groups influence policy.
Prior to MIT, she was a PhD fellow at the Argentine National
Scientific and Technical Research Council (CONICET) working at Gino
Germani Research Institute, and she taught at the University of
Buenos Aires and Torcuato Di Tella University.
Dan Slater (PhD Emory University) is Weiser Professor of Emerging
Democracies and Director of the Weiser Center for Emerging
Democracies (WCED) at the University of Michigan. He specializes in
the politics and history of enduring dictatorships and emerging
democracies, with a regional focus on Southeast Asia. He previously
served as Director of the Center for International Social Science
Research (CISSR), Associate Professor in the Department of Political
Science, and associate member in the Department of Sociology of
the University of Chicago. He is the author of Ordering Power:
Contentious Politics and Authoritarian Leviathans in Southeast Asia
(Cambridge University Press, 2010) and co-author of Coercive
Distribution (Cambridge University Press, 2018).
Ashutosh Varshney (PhD Massachusetts Institute of Technology)
is Sol Goldman Professor of International Studies and the Social
Sciences and Professor of Political Science at Brown University,
where he also directs the Center for Contemporary South Asia.
Previously, he taught at Harvard, Notre Dame, and the University of
Michigan, Ann Arbor. His books include Battles Half Won: India’s
Improbable Democracy (Penguin Books); Collective Violence in
Indonesia (Lynne Rienner Publishers); Ethnic Conflict and Civic Life:
Hindus and Muslims in India (Yale University Press); India in the Era
of Economic Reforms (with Jeffrey Sachs and Nirupam Bajpai,
Oxford University Press); and Democracy, Development, and the
Countryside: Urban–Rural Struggles in India (Cambridge University
Press).
Lucan Ahmad Way (PhD University of California, Berkeley) is
Professor of Political Science at the University of Toronto. Way’s
research focuses on democratization and authoritarianism in the
former Soviet Union and the developing world. His most recent book
(with Steven Levitsky), Social Revolution and Authoritarian Durability
in the Modern World (forthcoming from Princeton University Press),
provides a comparative historical explanation for the extraordinary
durability of autocracies born of violent social revolution. Professor
Way’s solo authored book, Pluralism by Default: Weak Autocrats and
the Rise of Competitive Politics (Johns Hopkins University Press,
2015), examines the sources of political competition in the former
Soviet Union. His book Competitive Authoritarianism: Hybrid
Regimes after the Cold War (with Steven Levitsky), was published in
2010 by Cambridge University Press. Way’s work on competitive
authoritarianism has been cited thousands of times and helped
stimulate new and wide-ranging research into the dynamics of
hybrid democratic-authoritarian rule.
 Abbreviations

ADP Agrarian Democratic Party

AEI Alliance for European Integration
AITI Association for the Integration of Timor in Indonesia
ANC African National Congress
APODETI Popular Democratic Association of Timor
ASDT Timorese Social Democratic Association
BDPM Bloc for a Democratic and Prosperous Moldova
BJP Indian People’s Party
BPP European Solidarity / Petro Poroshenko Bloc
BYuT Yulia Tymoshenko Bloc
CA Constituent Assembly (India)
CAA Citizenship Amendment Act
CAC Argentine Chamber of Commerce
CDM Electoral Bloc Democratic Convention of Moldova
CEMIDA Center of Military Members for the Argentine Democracy
CGT General Confederation of Labor
CNRT National Congress for Timorese Reconstruction
CONADEP National Commission on the Disappearance of Persons
COSATU Congress of South African Trade Unions
CRIET Court of Punishment of Economic Crimes and Terrorism
DPM Democratic Party of Moldova
ECLAC United Nations Economic Commission for Latin America and the
Caribbean
ENM United National Movement
ESMA School of Mechanics of the Navy
EU European Union
F-FDTL Timor-Leste Defense Force
FALINTIL Armed Forces for the National Liberation of Timor-Leste
FARD Action Front for Renewal and Development
FCBE Cowry Forces for an Emerging Benin
FPV-PJ Front for Victory-Peronist Party
FRELIMO Mozambique Liberation Front
FRETILIN Revolutionary Front for an Independent Timor-Leste
FRTLI Revolutionary Front for an Independent East Timor
GD Georgian Dream
GDP Gross Domestic Product
Gerindra Great Indonesia Movement Party
GNI Gross National Income
GNU Government of National Unity
GOLKAR Party of Functional Groups
IMF International Monetary Fund
INC Indian National Congress
JD Janata Dal
JNP Janata Party
KGB Committee for State Security
KOTA Association of Timorese Heroes
MODIN Movement for Dignity and Independence
MPLA Popular Movement for the Liberation of Angola
NDPU People’s Democratic Party of Ukraine
NF People’s Front
NGO nongovernmental organization
NP National Party
NRC National Registry of Citizens
NU Nahdlatul Ulama
NUNS Our Ukraine‚ People’s Self-Defense Bloc
OAS Organization of American States
OIC Organization of Islamic Cooperation
OECD Organization for Economic Co-operation and Development
OPEC Organization of the Petroleum Exporting Countries
PAN National Mandate Party
PAS Party of Action and Solidarity
PCRM Party of Communists of the Republic of Moldova
PD Democratic Party (Indonesia)
PD Democratic Party (Timor-Leste)
PDAM Agrarian Democratic Party of Moldova
PDI Indonesian Democratic Party
PDIP Indonesian Democratic Party of Struggle
PDM Democratic Party of Moldova
PIL Public Interest Litigation
PJ Peronist Party
PKB National Awakening Party
PKI Indonesian Communist Party
PLDM Liberal Democratic Party of Moldova
PLP People’s Liberation Party
PNI Indonesian Nationalist Party
PNTL National Police of Timor-Leste
PPP United Development Party (Indonesia)
PPP purchasing power parity
PR Party of Regions
PRD Democratic Renewal Party
PRO Republican Proposal
PRPB Peoples’ Revolutionary Party of Benin
PSD Social Democrat Party
PU Progressive Union
RB Renaissance Party of Benin
RENETIL National Resistance of East Timorese Students
RUKH People’s Movement of Ukraine
SACP South African Communist Party
SICONAR Ship Technicians’ Union
A
SJP(R) Samajwadi Janata Party (Rashtriya)
SN Servant of the People
SRA Argentine Rural Society
TMC Trinamool Congress
TRC Truth and Reconciliation Commission
UAE United Arab Emirates
UCEDE Union of the Democratic Center
UCR Radical Civic Union
UDT Timorese Democratic Union
UIA Argentine Industrial Union
UN United Nations
UNAMET United Nations Mission to East Timor
UNDERTI National Union of the Timorese Resistance
M
UNM United National Movement
UNSD United Nations Statistics Division
UNTAET United Nations Transitional Administration in East Timor
UP Uttar Pradesh
USAID United States Agency for International Development
ZYU For a United Ukraine!
 1
Introduction
Democracy in Hard Places
Tarek Masoud and Scott Mainwaring

If recent political events have taught us anything, it is that democracy is often fragile.
Throughout the world, in such places as Hungary, Poland, India, and Brazil, democratic regimes
now find themselves imperiled by the rise of ultra-nationalist and populist leaders who pay a
steady lip service to the will of the people while daily undermining freedom, pluralism, and the
rule of law. Not even the wealthiest and most powerful of the world’s democracies—the United
States of America—has proven immune. The one-time “arsenal of democracy” is now
sometimes held up as a candidate for democratic backsliding. In their 2018 bestseller, How
Democracies Die, Steven Levitsky and Daniel Ziblatt testify to an “epidemic of norm breaking
that now challenges our democracy” and warn of an American future in which no-holds-barred
partisan warfare leads to either a perpetual state of crisis or the inauguration of a full-blown,
one-party regime. As if in agreement, Freedom House now ranks the United States sixty-first
out of 210 countries in terms of its level of freedom, behind much younger democracies such
as the Czech Republic, Slovakia, Lithuania, Greece, Chile, and Taiwan (Freedom House, n.d.).
Although we do not agree with this judgment, the Polity V project assigns the United States a
score of 8 out of 10 for the year 2016, lower than its score from 1809–50 (Marshall and
Jaggers 2020). And in 2015, the Economist Intelligence Unit, which maintains its own
“Democracy Index,” downgraded the United States to a “flawed democracy,” a category
reserved for countries with free and fair elections and basic civil liberties but with “problems in
governance, an underdeveloped political culture and low levels of political participation” (The
Economist Intelligence Unit 2020, 53).
An illustration of how much democracy has lapsed, even in places where we would not
have expected it to, can be seen in Figure 1.1, which plots one hundred years of global and
OECD average scores on the “liberal democracy” index compiled by the University of
Gothenburg’s Varieties of Democracy project (Coppedge et al. 2021). Since peaking around
2011, average scores on that index—which captures the extent of civil liberties, rule of law,
judicial independence, checks on executive power, and electoral integrity—have now declined
to levels not seen since the end of the Cold War, and the trend appears to point downward.
“Nor,” as Larry Diamond has written, “do the numbers capture the full extent of the danger.”
According to Diamond, “China, Russia, and their admirers are making headway with a new
global narrative, hailing strongman rule—not government by the people—as the way forward in
difficult times.”1 The current global pandemic threatens to make a bad situation worse, with
increasing unemployment and, in many countries, shrinking GDPs fueling popular anger and
testing the limits of mass and elite faith in democracy in ways that appeared to accelerate
democratic regressions around the world.
Figure 1.1. Democracy in the world, 1900–present
Source: Coppedge et al. 2021.

These new doubts about democracy’s survival in some of the world’s most prosperous
countries lend new relevance to studies of its survival in places less blessed by abundance.
After all, if American democracy can be compromised by heightened polarization and the
willingness of some political elites to undermine long-standing norms in efforts to gain electoral
advantages, then democracy’s persistence in large, poor, ethnically diverse countries such as
India and Indonesia, with their low rates of educational attainment, rickety governing
apparatuses, and frequent economic difficulties, is nothing short of miraculous. Moreover,
taking stock of democracy’s record across time and space, we find several such instances of
democratic survival in the face of seemingly long odds. Since 1983, Argentines have held onto
their democracy despite three protracted economic depressions, military mutinies, and a prior
history of repeated democratic failures. In the 1990s, South Africans built one of Africa’s
longest-lived democracies out of the ruins of a system of white supremacy that dehumanized
the great majority of citizens and pitted the races against each other. And in the early 2000s,
the people of Timor-Leste erected a multi-ethnic democracy after a decades-long civil war that
laid waste to countless lives and livelihoods. These cases demand our attention, both because
they too face ever-present risks of democratic decay and downfall and because we may be able
to derive from them lessons about how democracy can be fortified in times of challenge.
This, then, is a book about how democracy persists when all signs suggest that it should
not. It puts front and center cases of what we call “democracy in hard places”: countries that
lack the structural factors and exist outside of the contexts that scholars have long associated
with democracy’s emergence and endurance. Democracies in hard places overcome
underdevelopment, ethnolinguistic diversity, state weakness, and patriarchal cultural norms.
They tame grasping, politically ambitious militaries; transcend influences and pressures from
autocratic neighbors; and cope with polarized political parties. Without denying that democracy
is easier to build and hold onto in societies that are free of such hurdles, this book asks what
we can learn about strengthening democracy from those that managed to leap over them. By
theorizing about democratic survival from such cases—which, in the parlance of social science,
lie “off the regression line”—we capture what Michael Coppedge identifies as “the greatest
potential to innovate and challenge old ways of thinking” (Coppedge 2002, 16). Are
democracies in hard places the equivalent of lottery winners—dramatic exceptions to
fundamental rules? Or is there something systematic that can be gleaned from such cases
about how democracy can be erected and upheld around the world?
To answer these questions, this book presents nine case studies—written by leading experts
in the discipline—of episodes in which democracy emerged and survived against long odds. The
cases are drawn from almost every region of the world that formed part of what Samuel
Huntington called the “third wave” of democracy, which began in southern Europe in the mid-
1970s, spread to Latin America in the late 1970s and 1980s, and to Eastern Europe and sub-
Saharan Africa in the 1990s. Six of the cases are ones of long-term democratic survival—
Argentina (1983–present), Benin (1991–2019), India (1977–present), Indonesia (1999–
present), South Africa (1995–present), and Timor-Leste (2002–present). The other three have
more mixed democratic records—Georgia (2005–present), Moldova (1995–2005, 2010–
present), and Ukraine (1995–98, 2007–14). In each case, many of the conditions
conventionally associated with durable democracy were either attenuated or absent. Each case
study details the constellation of obstacles to democracy faced by a given country, describes
the major political actors with the potential to impact regime trajectories, and explains how the
threat of democratic breakdown was staved off or averted.
Figure 1.2 plots the V-Dem “liberal democracy” scores of the cases in this book from 1974
to 2017 (the most recent year available).

Figure 1.2. Liberal democracy scores of the nine cases in this volume, 1974–2017
Source: Coppedge et al. 2018.

Although the case studies presented in this book do not offer a unified answer to the
question of how democracy survives in inauspicious conditions, readers will find in them
powerful rejoinders to structural accounts of democracy’s emergence and survival. This is
perhaps to be expected, given that the cases were selected based on their want of democracy’s
hypothesized structural causes. But, as the editors of this volume, we find in the narratives of
democratic survival offered in this book striking evidence of the importance of political actors,
their normative beliefs, and how these beliefs shape their choices. Twenty-five years ago, Adam
Przeworski and Fernando Limongi (1997, 176–77) complained that, in structural accounts of
democratization, “no one does anything to bring democracy about; it is secreted by economic
development and the corollary social transformations.” In reality, they argued, “democracy was
an outcome of actions, not just of conditions.” The cases in this volume validate that
observation.

Agents, Interests, and Regime Preferences

Agent-centric accounts of political phenomena have sometimes taken the form of “great
person” narratives in which individual leaders operate with considerable autonomy, and in
which outcomes are the result of individual decision-making. Without denying that individual
leaders occasionally make a decisive difference in outcomes, this book eschews that tendency.
Its focus is not on individuals but on key collective actors who command political resources and
influence political competition (Mainwaring and Pérez-Liñán 2013, 5). These include chief
executives and their ruling parties, other political parties, militaries and coercive agencies, labor
unions, social movements, and civil society groups (Bermeo and Yashar 2016a). These are the
agents whose preferences and choices determine whether a democracy endures, or joins the
long list of democracies undone by military takeovers, auto-golpes, palace coups, and civil
wars. Second, each chapter takes seriously the constraints under which actors operate. These
constraints may take the form of structural or cultural conditions, formal institutions (such as
constitutional structures or electoral rules), international actors, or other domestic actors (such
as courts, militaries, or political parties). In addition to these “external” constraints, many of
the chapters also highlight “internal” constraints that have hitherto received short shrift in
accounts of democratic emergence and survival—in particular, the existence of deeply rooted
normative commitments that shape the behavior of collective actors and take certain courses of
action off the table.
Thus, if one goal of this book is to refocus the study of democratic survival from structural
and external contextual conditions to the role played by political actors, another is to contribute
to the debate about what drives those actors. What causes, for instance, a chief executive to
forego an opportunity to expand his power at the expense of democratic legitimacy? Why
would a powerful military in a fledgling democracy stick with a messy democratic system
instead of answering calls to enter the fray in the name of stability? Why would an incumbent
facing electoral defeat accept that defeat instead of tampering with the election or simply
abrogating it outright? Assuming that democratic survival can reasonably be thought of as a
function of such choices, what explains those choices?
The literature has offered two broad answers to this question. The first holds that actors’
preferences over regime type are driven purely by considerations of self-interest. That is, they
weigh the power and income they can expect to derive from different governing arrangements
and choose accordingly. Thus, Adam Przeworski (1991) argues that democracy survives (i.e., is
an equilibrium) when the losers in a particular election believe that they have enough of a
chance of winning in the future to make sticking with democracy an attractive proposition.2
Such accounts assume, as Przeworski (2019, 172) does, that “[t]he dream of all politicians is to
remain forever in office and to use their tenure to do whatever they want.” In this telling,
political actors accept democracy and the associated constraints on power only because it
serves their interests, or because they have no alternative.
The second answer to the question of why some political actors sustain democracy despite
pressures to the contrary holds that self-interest or external constraints are insufficient to
explain most instances of democratic persistence. In an earlier work, one of the authors of the
present chapter points out that parties and movements have often upheld democratic regimes
despite having plausibly better prospects outside of them: “Because actors believe in the
system, they are willing to make concessions to abide by the rules of the game” (Mainwaring
1989). Without denying that actors also have instrumental goals, this alternative view locates
the sources of actors’ democracy-sustaining decisions in their values and normative
commitments. Canonical works that share this view include Berman (1998), Bermeo (1992),
Dahl (1971, 124–89), Hofstadter (1969), Lepsius (1978), Levitsky and Ziblatt (2018), Linz
(1978a, 1978b), and McFaul (2002). Garrard’s (2002) account of the emergence of democracy
in Great Britain similarly emphasizes the acceptance of liberal values by key collective actors.
Along related lines, an extensive body of literature on populism and the threat it poses to
democracy hinges explicitly on populist leaders’ hostility to the liberal aspects of democracy
(Müller 2016b; Weyland 2020; Weyland and Madrid 2019). In short, these works conclude that
democracy survives when the principal political actors who are in a position to maintain a
regime or tear it down bear a preference for democracy that extends beyond self-interest and
usually leads them to prioritize it over other important desiderata.3
The chapters in this volume take different perspectives on this issue, and this book is
unlikely to resolve such a long-standing debate. At the same time, much is to be gained by
placing this question firmly on the agenda for both scholars and practitioners. The urgency of
understanding how political actors can be made to sustain democracy in hard places cannot be
exaggerated. It derives not just from the fact that democracy around the world is under threat,
but from the realization that the greatest potential for democratic expansion lies in hard places
similar to the cases explored here. A recent survey by Mainwaring and Bizzarro identified 103
transitions to democracy between 1975 and 2012, many of which were in hard places, and
almost half of which (46) broke down, with an average democratic duration among the
breakdown cases of only eight years (Mainwaring and Bizarro 2020). If this statistic is to be
improved, scholars and practitioners must understand the factors that have enabled a small but
not insignificant number of hard places to transcend the constraints of structure and achieve
and maintain decent, representative government.
The remainder of this chapter proceeds as follows. First, we offer a review of some of the
most important structural and contextual factors that have been hypothesized to determine a
country’s democratic prospects. We then identify the nine cases whose democratic trajectories
this book examines, explain why we consider them hard places, and discuss how each of them
fits into the book. We end the chapter with the volume’s conclusions and contributions to the
theory and practice of democratization.

The Effects of Context

Democracy’s survival in “hard” places, and its looming retrenchment in comparatively “easy”
ones, are two of the most powerful empirical rejoinders to the structural and contextual
accounts of democratic transition and democratic stability. However, there is a reason those
accounts became nearly hegemonic in the field of comparative democratization—they have
been validated by a wealth of cross-national, statistical analyses. Accordingly, this project is
predicated on the well-grounded idea that democracy is more likely to survive under some
conditions than others. In this section, we discuss seven of these facilitating conditions: (1)
economic development, (2) macroeconomic performance, (3) economic inequality (or, more
properly, the lack thereof), (4) state capacity, (5) ethnic homogeneity, (6) democratic cultures,
and (7) a pro-democratic international environment. We then introduce the cases that
demonstrate the possibility for democratic survival even in the absence of some or all of these
hypothesized conditions.
Of all the hypothesized enabling conditions for democracy, perhaps the one on which there
is the broadest agreement is the achievement of a certain kind of prosperity, built on a shift
from agriculture to industry. More than sixty years ago, Seymour Martin Lipset (1959, 75)
famously found that “the average wealth, degree of industrialization and urbanization, and level
of education is much higher for the more democratic countries” and concluded that “the more
well-to-do a nation, the greater the chances that it will sustain democracy.” In the intervening
decades, scholars confirmed Lipset’s initial empirical finding and offered a variety of
explanations for it (Boix 2011; Boix and Stokes 2003; Epstein et al. 2006; Przeworski et al.
2000).4 Some have argued that affluent societies are more capable of sustaining democracy
because they have more educated citizenries, who are thus more rational and tolerant and less
receptive to the radical ideologies and emotional appeals of would-be autocrats. For others, the
correlation between economic development and democracy is largely due to the growth, in
prosperous societies, of autonomous civic groups that provide citizens with the wherewithal to
demand greater liberties and to deter or beat back encroachments upon them. For still others,
democracy follows from development because the relative prosperity of the average citizen in a
wealthy society means that economic elites possess less reason to fear expropriation if they
grant or expand the franchise (Acemoglu and Robinson 2006; Boix 2003). Without
development, the reasoning goes, these and other beneficial mechanisms would not operate,
and political life would regress to a repressive and conflict-ridden mean. As Alemán and Yang
(2011, 1143) concluded in a study of democratic transition and breakdown from 1970 to 1999,
“the greatest threat to democratic survival is a low level of income.”
Other scholars have noted that sustained periods of poor economic performance, in which
GDP per capita either declines or stagnates, can seriously lower the probability of democratic
survival (Gasiorowski 1995; Przeworski et al. 2000; Sing 2010; Svolik 2015). “Like war and
revolution,” Wibbels and Roberts (2010, 384) wrote, economic crises are “periods of severe
disequilibria” that can “transform a political landscape in profound ways,” and “shake the
foundations of nations.” The breakdown of Germany’s Weimar Republic in 1933 is the most
famous example in which widespread immiseration resulting from economic crisis contributed
to the end of a democratic regime. In their study of democracy in twenty-six developing
countries, Diamond, Lipset, and Linz (1987, 8) declared “it is clear that, over the long term in
particular, the effectiveness of democratic regimes in satisfying people’s wants heavily affects
their stability” and find that the most successful democracies “have generally experienced
relatively steady economic growth, which in turn has strengthened their legitimacy.” It makes
deductive sense that citizens would care more about economic security and less about
democracy in moments of economic hardship. For this reason, Przeworski (1991, 136) argued
that there were steep tradeoffs between economic reform and democracy in the post-
communist countries and Latin America. Economic reforms, he wrote, “are socially costly and
politically risky. . . . [T]hey hurt large social groups and evoke opposition from important
political forces. And if that happens, democracy may be undermined or reforms abandoned, or
both.”
Yet another economic factor broadly associated with democracy’s emergence and survival is
the level of economic inequality. Carles Boix (2003) argues that democracy emerges when
inequality decreases: “As the distribution of income becomes more equal among individuals,
redistributive pressures from the poorest social sectors on the well-off voters diminish.
Accordingly, the relative costs [to the wealthy] of tolerating a mass democracy decline” (Boix
2003, 10). Related arguments have been made by Lipset (1959, 83) and Acemoglu and
Robinson (2006).5 In contrast, Ansell and Samuels (2010, 1545) predict a positive relationship
between democratization and rising income inequality, as democratic reforms are often
demanded by “newly wealthy economic groups” that “accumulate an increasing share of
national income” that they seek to defend against state predation. However, an empirical
analysis by Haggard and Kaufman (2012) of third-wave democratic transitions and reversals
finds little support for the existence of a systematic relationship between inequality and
democratic emergence and survival.
A fourth, non-economic structural factor sometimes identified as important to sustaining
democracy is the strength of the state apparatus. At the extreme, any regime—democratic or
not—that cannot maintain what Max Weber called “a monopoly over the legitimate use of
violence” is unlikely to survive for long. But even states that can manage this basic, minimal
function of statehood may prove unable to hang onto democracy if they lack the infrastructural
power (Mann 1984) to carry out essential government functions of law enforcement, taxation,
and service delivery. O’Donnell (1993) describes the danger to new democracies of “brown
areas” in which state incapacity leads to a functional truncation of citizenship rights. Slater,
Smith, and Nair (2014, 354) argue that democracies with ineffectual governing apparatuses
quickly find themselves the objects of mass discontent and highly susceptible to military
intervention: “[A]dministrative incapacity means that recurrent crises of governability will
repeatedly tempt and enable military intervention to restore political stability. Meanwhile,
democracy’s chronic failure to ‘deliver the goods’ in weak-state settings will give the poor
majority little reason to defend democracy against its enemies.”
A fifth contextual underpinning of democratic survival identified by scholars is the presence
of relative ethnic, linguistic, and religious homogeneity (Dahl 1971; Horowitz 2000; Rabushka
and Shepsle 1972; Welsh 1993). Societies riven by identity-based cleavages are thought to lack
the comity that helps to keep democratic competition from spilling over into bloodshed. As
Arend Lijphart wrote more than forty years ago, the difficulty of “achiev[ing] and maintain[ing]
stable democratic government in a plural society is a well-established proposition in political
science” (1977, 1). Although recent analyses of cross-national data have attenuated the link
between diversity and democratic survival (Fish and Brooks 2004; Teorell 2010), there
nonetheless remain strong theoretical reasons why we might expect ethnic diversity to inhibit
democratic survival. An early expression of these reasons was offered by John Stuart Mill, who,
writing in the middle of the nineteenth century, declared that “free institutions are next to
impossible in a country made up of different nationalities.” Divided by language and culture, he
reasoned, citizens in such societies eye each other with suspicion, as “one section does not
know what opinions, or what instigations, are circulating in another.” Instead of uniting to hold
national leaders accountable and constrain their power, said Mill, the diverse citizenry’s “mutual
antipathies are generally much stronger than dislike of the government,” which, presumably, is
able to divide and conquer (Mill 1861, 289). A practitioner’s corroboration of the importance of
homogeneity to democracy was offered by the early twentieth-century British prime minister,
Arthur Balfour, who is supposed to have said, “[O]ur whole political machinery presupposes a
people so fundamentally at one that they can safely afford to bicker” (quoted in Friedrich 1939,
571). In the absence of that salutary oneness, the reasoning goes, paper stones would soon be
converted into real ones. Scholars have found this danger to be particularly in a democracy’s
early stages, when political entrepreneurs might be most tempted to mobilize ethnic sentiments
in order to acquire power (Horowitz 2000; Snyder 2000).
 A sixth theorized requisite of democracy is the existence in a given society of what might be
called a democratic “culture.” As Welzel and Inglehart (2009) put it, “[M]ass beliefs are of
critical importance for a country’s chances to become and remain democratic.” Perhaps not
surprisingly, for many scholars, the culture most conducive to democracy is that of the Western
countries in which modern forms of mass, democratic government first emerged. Thus, Almond
and Verba (1963, 5), argue that a “democratic form of participatory political system requires as
well a political culture consistent with it” and lament the difficulty of transferring “the political
culture of the Western democratic states to the emerging nations.” In his study of
democratization in Britain, Garrard (2002) describes a gradual process in which previously
disenfranchised groups acquired the right to participate upon their achievement of a certain
level of what he calls “political fitness,” defined largely in terms of their attachment to
classically liberal values held among the British elite.
Democracy’s early emergence and survival in Northern European countries caused some to
implicate the distinctive religion of that part of the world, Protestant Christianity. Woodberry
(2012, 245) notes that “by World War I every independent, predominantly Protestant country
was a stable democracy—with the possible exception of Germany,” a fact he attributes to
Protestantism’s emphasis on mass education and religious liberty. Other forms of Christianity, in
contrast, were deemed more likely to uphold traditional, autocratic politics. For instance, Lipset
(1959, 93) declared that Roman Catholicism’s insistence of a monopoly on truth was at odds
with democracy, “which requires, as part of its basic value system, the belief that ‘good’ is
served best through conflict among opposing beliefs.” Orthodox Christianity’s hospitality to
democracy has similarly been thrown into doubt (Marsh 2005), and Prodromou (2004, 62) has
pointed out that “Orthodox churches often display a certain ambivalence about key elements of
the pluralism that characterizes democratic regimes.” If the existence today of a large number
of Catholic-majority democracies appears to falsify Lipset’s hypothesis, democracy continues to
be weak in the Eastern Orthodox world (which is largely coterminous with the former Soviet
Union).
Perhaps no religious tradition has been scrutinized for its compatibility with democracy as
has Islam—a 1,400-year-old faith practiced by almost two billion people. More than 250 years
ago, Montesquieu ([1748] 1900) famously argued that “a moderate government is most
agreeable to the Christian religion, and a despotic government to the Mohammedan.” Alexis de
Tocqueville similarly thought that Islam was inimical to democracy, arguing that Islam’s fusion
of religion with politics meant that it could “never long predominate in a cultivated and
democratic age” (quoted in Hashemi 2009, 118). More recent testimonials to this supposed
incompatibility of Islam and democracy were offered by the late English historian Elie Kedourie
(1994, 1), who wrote that “the idea of democracy is quite alien to the mind-set of Islam,” and
the American political scientist, Samuel Huntington (1991, 307), who contended that, despite
containing some pro-democratic elements, “Islamic concepts of politics differ from and
contradict the premises of democratic politics.” Repeated cross-national studies by Fish (2002),
Donno and Russett (2004), and Teorell (2010) seem to validate this skepticism, finding a robust
and negative correlation between Islam and democracy, albeit one driven largely by the durable
autocracies of the Muslim world’s Arabic-speaking core (Stepan and Robertson 2003).6
Although the persistence of Indonesian democracy since 1999 and Tunisia’s existence as a
relatively stable democracy between 2011 and 2021 could be thought to constitute rejoinders
to arguments about Islam’s incompatibility with democracy, the decay of previously
consolidated, Muslim-majority democracies like Turkey and Mali give the hypothesis continued
relevance.
A seventh contextual factor long thought to shape the possibility for democratic survival is
the international environment confronting a given regime. More than fifty years ago, Rustow
(1970, 348) recognized the existence of “cases where the major impulse to democratization
came from the outside,” although he left their analysis to scholars of international relations.
Huntington (1991, 13) argued that “the occurrence and the timing of the third-wave transitions
to democracy” was shaped in part by “changes in the policies of external actors, most notably
the European Community, the United States, and the Soviet Union,” and the “demonstration
effect of transitions earlier in the third wave in stimulating and providing models for subsequent
efforts at democratization.” Brinks and Coppedge (2006, 464) point to the importance of
superpower influence on the regime types of their clients, as well as to what they call
“neighbor emulation,” which they define as “a tendency for neighboring countries to converge
toward a shared level of democracy or nondemocracy.” In their study of democratization in
Latin America, Mainwaring and Pérez-Liñán testify to the importance of neighbor influence,
arguing that a “favorable regional political environment, characterized by the existence of many
democracies in Latin America, increase[d] the likelihood of transitions from authoritarian rule to
competitive regimes and diminishe[d] the likelihood of breakdowns of existing competitive
regimes” (2013, 6). Observers of Eastern Europe, East Asia, and the Middle East have noted
the ways in which regional powers exert pressures on their neighbors that lead to a kind of
institutional isomorphism (Burnell 2010; Coppedge et al. 2022; Darwich 2017; Lee 2018).
Table 1.1 contains the results of a time series, cross-sectional logistic regression analysis of
the correlates of democratic breakdown for the eighty-three countries that transitioned to
democracy between 1974 and 2012.7 The purpose of this exercise is to identify the factors
most associated with democratic longevity and breakdown, and to subsequently calculate the
breakdown risk for each of the cases in this volume. The dependent variable, democratic
breakdown, is a dichotomous variable that captures a shift from electoral democracy to
autocracy, measured up to 2017. Specifically, we used Mainwaring and Bizzarro’s (2020)
modified version of Lührmann et al.’s (2018) “Regimes of the World Dataset,” which is based on
the Varieties of Democracy (V-Dem) project. Lührmann et al. consider a regime minimally
democratic if (1) it meets minimum standards of free and fair elections; (2) it allows for some
multiparty competition; and (3) V-Dem’s electoral democracy score (which ranges between 0
and 1) is at least 0.50. Mainwaring and Bizzarro modified these coding rules to exclude as
either a transition or breakdown small, momentary increases above or below the 0.50 electoral
democracy threshold that were followed by a quick reversion to the prior regime state. (For
instance, a country whose democracy score of 0.49 jumps in one year to 0.51 before settling
back down at 0.49 would not be considered to have undergone a transition followed by a
breakdown.)
The key independent variables are:
 Economic development: This is measured as the natural log of the GDP per capita (in 2010 constant US doll
sourced from the World Bank’s World Development Indicators.
Economic inequality: This is the Gini index, which measures income inequality on a scale from 0 to 1, whe
corresponds to greatest inequality. The source for this variable is the V-Dem dataset, which in turn draws on
collected by the United Nations University’s World Institute for Development Economics Research (Coppedge e
2018).
Economic growth: This is the percentage change in GDP per capita for each country year.
Neighborhood effects: This is the share of countries in each geopolitical region (post-communist, Latin Ame
Middle East, sub-Saharan Africa, Western Europe and North America, East Asia, Southeast Asia, South Asia,
Pacific, and the Caribbean) that score 0.5 or above on the Varieties of Democracy project’s electoral democ
index.
Years democratic: Calculated as the number of years since the last democratic transition.
Ethnic fractionalization: This is the probability that any two randomly selected individuals from a given country
given year will be from different ethnic groups, calculated by Drazanova (2019) from data gathered by
Composition of Religious and Ethnic Groups project at the University of Illinois at Urbana-Champaign. The in
ranges (theoretically) from 0, indicating that every citizen is from the same ethnic group, to 1, where there ar
many ethnic groups as individuals.
State capacity: This is a measure of state infrastructural power devised by Hanson and Sigman (2013)
captures three dimensions of state capacity (extractive, administrative, and coercive). Values range from –3 to

Finally, cultural and religious influences are captured with dummy variables for membership
in the Organization of Islamic Cooperation (Albania, Bangladesh, Benin, Comoros, Guyana,
Indonesia, Iraq, Lebanon, Libya, Maldives, Mali, Niger, Nigeria, Senegal, Sierra Leone,
Suriname, Turkey, and Tunisia), the Arab League (Comoros, Iraq, Lebanon, Libya, Mauritania,
and Tunisia), and for being situated in Eastern Europe (Albania, Armenia, Belarus, Bulgaria,
Croatia, Czech Republic, Estonia, Georgia, Hungary, Latvia, Lithuania, Macedonia, Moldova,
Poland, Romania, Russia, Serbia, Slovakia, Slovenia, and Ukraine). Summary statistics for each
of these variables are presented in Table A1.1 in the appendix.

Table 1.1. Regression Analysis of Correlates of Democratic Breakdown

Random effects logit Penalized logit
(1) (2) (3) (4) (5) (6)
GDP (logged) −0.589
−0.744** −0.891* −1.372** −0.720 ** −0.981**
(0.373)
* (0.506) * * *
(0.207) (0.420) (0.186) (0.265)
GDP growth (%) −4.045 −4.361 −5.193 −5.982
−5.160* −5.194**
(4.254) (4.432) (4.720) (4.671)
(3.047) (2.533)
Share of neighboring countries democratic −0.755 −0.704
−6.911* −5.590* −4.822* −3.583*
(1.577) (1.427)
* (3.070) * (2.062)
(3.179) (2.194)
Years since last transition −0.00623 0.116 −0.00725 0.0537
0.137* 0.0783*
(0.0310) (0.0713) (0.0235) (0.0441)
(0.0716) (0.0457)
Organization of Islamic Cooperation −0.253 −1.967 −0.206
−2.168* −1.228* −1.332**
(0.479) (1.214) (0.423)
(1.218) (0.652) (0.637)
Arab League *** 4.246
2.266 5.242** 2.213*** 2.845* 3.875***
(3.451)
(0.822) (2.524) (0.671) (1.474) (1.234)
Eastern Europe
0.928* 2.262** 1.821** 0.910** 1.756** 1.353**
(0.499) (0.997) (0.804) (0.441) (0.757) (0.607)
Ethnic fractionalization index −1.462 −1.209
(1.103) (0.913)
 Random effects logit Penalized logit
(1) (2) (3) (4) (5) (6)
State capacity
−1.043* −0.890*
(0.548) (0.505)
Gini index 0.0309 0.0291
(0.0317) (0.0275)
Constant 3.838 1.643
2.492* 8.066** 2.373* 5.188***
(4.447) (3.108)
(1.460) (3.132) (1.301) (1.788)
Observations 1,675 1,026 1,026 1,675 1,026 1,026
Number of clusters (countries) 82 72 72
Note: Cells are coefficient estimates of regression models predicting the probability of democratic breakdown in a
given year. Models 1, 2, and 3 are random effects logits using Stata’s xtreg command, with standard errors
clustered by country. Models 4, 5, and 6 are penalized logits using the user-generated command firthlogit
(Coveney 2021). Standard errors in parentheses.
* p < .1;
** p < .05;
*** p < .01

Due to missingness in the data for ethnic fractionalization, state capacity, and economic
inequality, column 1 of Table 1.1 presents the results of a random effects logit without those
variables included. Column 2 presents the regression results with all variables included, at the
expense of losing nearly 40 percent of observations. In order to determine whether the
differences in the sign, significance, and magnitudes of the coefficients in columns 1 and 2 are
due to the inclusion of the additional variables in Model 2 or to the loss of observations, column
3 reports the results of a regression with only the smaller number variables in column 1 and
only the smaller number of cases in column 2. Columns 4, 5, and 6 repeat the analyses in
columns 1, 2, and 3 using a penalized logit estimator to account for the rarity of democratic
breakdowns in the data (forty-five in 1,725 regime-years). The results of these latter
regressions are substantially similar to the regressions without rare-events adjustments,
although effect sizes and standard errors are generally smaller.
As expected, the relationships between democratic breakdown and GDP, GDP growth,
regional regime characteristics, and state capacity are negative in most of the models—
meaning that democratic breakdown is less likely in countries that are richer, growing
economically, situated among other democracies, and which possess high state capacity.
However, none of these variables is statistically significant across all six models reported in
Table 1.1, which testifies to the uncertain relationship between structural factors and
democratic survival.
Figures 1.3a through 1.3g plot the marginal effects of our key continuous predictors (GDP
per capita, GDP growth, regional share of democracies, years democratic, ethnic
fractionalization, state capacity, and income inequality) on the probability of democratic
breakdown.8 Figure 1.3a shows that, holding all else equal, a country whose logged GDP per
capita is in the 75th percentile (approximately USD 8,500) has a 1.2 percent chance of
experiencing a democratic breakdown in a given year, which is about a quarter of the
probability that democracy will break down in a country whose GDP is in the 25th percentile
(approximately USD 1,500). Figure 1.3b displays the effect of economic growth on the
probability of democratic breakdown. A country in the 75th percentile of economic growth
experiences a probability of breakdown (0.28 percent) that is ten times as small as the
probability of breakdown for a country that is in the 25th growth percentile. The effect of
neighbor emulation, displayed in Figure 1.3c, is relatively flat (which is to be expected, given
that the coefficient on this variable is insignificant in Model 1). Based on the simulation for
Model 1, a democracy in a neighborhood that is 55 percent democratic (corresponding to the
75th percentile of regions on this measure) has, all else equal, a 2.5 percent chance of
breaking down, whereas one in a neighborhood that is 45 percent democratic (corresponding
to the 25th percentile), has a 2.7 percent chance of breaking down. State capacity is, as
expected, negatively associated with breakdown, as can be seen in Figure 1.3f. A democracy
with around the 25th percentile of state capacity has a 3.2 percent chance of breaking down in
a given year, while one in the 75th percentile of state capacity has a chance of breaking down
of around 1 percent.

Figure 1.3. Marginal effects of continuous predictors

Sources: For (a) and (b), World Development Indicators, n.d.; for (c), (d), and (g), Coppedge et al. 2018; for (e),
Drazanova 2019; and for (f), Hanson and Sigman 2013.

The effect of the years since democratic transition is sensitive to specification, with sign and
significance changing depending on which variables are included. The Gini index and ethnic
fractionalization do not achieve significance.9
The effects of the dichotomous regional predictors on democratic breakdown are presented
in Figures 1.4a through 1.4c. Here, the results offer mixed support for the conventional
wisdom. Membership in the Organization of Islamic Cooperation, which is our rough proxy for a
country’s Muslim majority status, is negatively associated with democratic breakdown—
significantly so in Model 5 when controls for state capacity, economic inequality, and ethnic
fractionalization are included. However, the indicator for membership in the Arab League is a
positive and almost always significant correlate of democratic breakdown (Model 2 is the only
exception). A non-Arab democracy has a 2.4 percent chance of breakdown in a given year,
while a democracy that is also an Arab League member has a 17 percent chance of breaking
down in a given year. Being situated in Eastern Europe is consistently a risk factor for
democratic breakdown, although the substantive impact is smaller than being an Arab League
member: the average Eastern European democracy has a 5.4 percent chance of breaking down
in a given year, while the average non–Eastern European democracy’s chance of breaking down
is around 2.3 percent.

Figure 1.4. Marginal effects of qualitative predictors

Note: The Organization of Islamic Cooperation includes Albania, Bangladesh, Benin, Comoros, Guyana, Indonesia,
Iraq, Lebanon, Libya, Maldives, Mali, Niger, Nigeria, Senegal, Sierra Leone, Suriname, Turkey, and Tunisia; the
Arab League includes Comoros, Iraq, Lebanon, Libya, Mauritania, and Tunisia; and Eastern European countries are
Albania, Armenia, Belarus, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Latvia, Lithuania,
Macedonia, Moldova, Poland, Romania, Russia, Serbia, Slovakia, Slovenia, and Ukraine.

Overall, the results reported in Table 1.1 provide mixed support for long-theorized
arguments about the relationship between contextual factors and countries’ democratic
prospects. Moreover, the instability in the magnitudes of the coefficients and their significance
indicate the need for caution about strong claims about the influence of these contextual
variables. The simple fact of the matter is that these variables cannot explain all of the
observed variation in democratic breakdown. Thus, although the chapters that follow all clearly
situate the case studies in the context of these macro variables, they also emphasize how
actors have maneuvered around them to keep democracy alive despite the odds.

The Cases
We have identified several factors that scholars have traditionally deemed important to a
country’s chances of getting and keeping democracy: economic development; the absence of
stark economic inequality; economic growth (or, at the very least, the absence of a prolonged
economic crisis that leads to mass disillusionment with democratic government); ethnic
homogeneity and the absence of deep identity cleavages; strong states capable of maintaining
order and governing effectively; mass cultures compatible with democratic norms of tolerance
and pluralism; and neighboring governments and foreign patrons that are themselves
democratic and supportive of democratic rule. Although these factors have some explanatory
power, they are nonetheless collectively unable to account for the durability of the many
democracies around the world that are poor, ethnically fragmented, with traditional cultures,
situated amid a sea of autocracies, and with troubled economies that are often stagnant or in
crisis.
To explore how democracies survive under such conditions, the book analyzes the regime
trajectories of nine countries that experienced considerable (if varying) democratic spells
despite facing great challenges: Argentina, Benin, Georgia, India, Indonesia, Moldova, South
Africa, Timor-Leste, and Ukraine. Two criteria guided our selection of cases. The first is that a
country should have survived as a democracy for a long period of time (as of 2017, when the
project was launched). Using the data compiled by Bizzarro and Mainwaring (2020), we
calculated that the average duration of a third-wave democracy (as of 2017) was 12.3 years.
We therefore established this duration as the initial cutoff for inclusion as a core case in the
volume. Of the nine cases in this volume, seven meet this criterion: Argentina, Benin, Georgia,
India, Indonesia, Timor-Leste, and South Africa. Two additional cases, Moldova and Ukraine,
have oscillated between democratic and autocratic spells and do not qualify as long-lasting
democracies, but they are among the most democratic of the twelve former Soviet Republics
outside of the Baltics.
The second criterion for case selection was that the cases be deemed improbable by the
lights of structural accounts of democratic survival. In all of these countries—save possibly for
Argentina—the factors that facilitate democratic survival were largely absent. Most experienced
low levels of economic development and educational attainment. Many inherited enfeebled
administrative apparatuses and coup-prone militaries. Some confronted repeated economic
crises. And some were located in parts of the world in which democracies were few, or were
embedded in cultures and religions very different from those in which modern democracy was
incubated.
Table 1.2 presents the average score for each country on each of the predictors analyzed in
Table 1.1, for the entirety of the period for which it appears in the data. Table 1.2 highlights
some of the daunting challenges the countries in this volume faced in building long-lasting
democracies. Most of the countries are poor—Benin and India strikingly so. Argentina, South
Africa, and Ukraine have had poor economic performance as measured by per capita GDP
growth. India, Benin, and South Africa are situated in regions dominated by authoritarian
regimes. Benin, Timor-Leste, India, and South Africa have great ethnic and linguistic diversity.
Benin and Timor-Leste have shaky state capacity. And South Africa features some of the
starkest income inequalities on the planet.

Table 1.2. Values for Each Case on Key Independent Variables (Socioeconomic Indicators Averaged over
Democratic Lifespan)
Countr GDP GDP Regional Years Ethnic State Gin Islam Ar Eastern
y per growth democra democrati fractionalizati capaci i ic ab Europe
capita (%, cy share on ty an
ca
geometr (%)
ic
mean)
Argenti 8,511. 1.01 55.74 34 0.13 0.20 45. 0 0 0
na 71 88
Benin 726.89 2.04 41.22 26 0.77 −0.93 36. 1 0 0
91
Timor- 3,087. 4.18 50.79 16 0.83 −1.42 32. 0 0 0
Leste 56 5
Georgia 3,515. 5.67 54.39 13 0.38 0.38 40. 0 0 1
77 77
India 879.18 3.96 38.36 41 MISSING −0.03 33. 0 0 0
5
Indones 3,006. 3.89 49.32 18 0.8 0.13 35. 1 0 0
ia 01 3
Moldov 1,713. 3.29 52.16 19 0.48 −0.31 34. 0 0 0
a 23 77
South 6,797. 1.31 43.81 22 0.86 0.72 59. 0 0 1
Africa 02 75
Ukraine 2,665. 1.69 52.97 12 0.39 −0.16 29. 0 0 1
59 91
Sources: Coppedge et al. 2018; Drazanova 2019; Hanson and Sigman 2013; World Development Indicators, n.d.
a According to coding scheme in Mainwaring and Bizzarro (2020).

Figures 1.5 through 1.13 display the breakdown probabilities—drawn from Model 1 in Table
1.1—of each of the cases in this volume. In order to enable the reader to more easily
comprehend the outsized magnitude of these (objectively small) breakdown probabilities in a
given year, we include in each graph the breakdown probabilities for Spain—one of the first
third-wave democracies—over the life of its democratic regime. According to these simulations,
among our six cases of long-lasting democracy, India in the years just after redemocratization
in 1977 faced particularly long odds. Moldova and Ukraine also had years in which the
predicted odds of breakdown were vastly higher than those for Spain. For all of these cases,
the probability of breakdown in any particular year was not startlingly high—although it was
much higher than Spain’s—but the cumulative probability of survival over the life span of these
democracies ranges from poor to moderate.
The first case in the book is India, which is perhaps the quintessential democracy in a hard
place, having long been identified by scholars as a stubbornly persistent rejoinder to most
theories of democratization. (The annual breakdown probabilities for India, based on our
model, are presented in Figure 1.5.) Poor, ethnically heterogeneous, with a caste system that
seems incompatible with democracy’s assumptions of formal political equality, and located in a
neighborhood with precious few democracies to speak of, India seems almost entirely bereft of
the things that scholars believe to be conducive to democracy. And yet, as contributor Ashutosh
Varshney points out, India has been a democracy for seventy-two of the last seventy-four
years. Although India has always fallen short of the liberal ideal and currently faces a populist
challenge not unlike those faced by some other established democracies, it is nonetheless
perhaps democracy’s most remarkable success story.

Figure 1.5. India (1978–2017), probability of democratic breakdown (benchmarked against Spain)

Varshney’s explanation for how India has eluded the democratic collapse predicted for it by
most theories of democratization emphasizes the importance of the beliefs and values of that
country’s modern founders, and, in particular, its first prime minister, Jawaharlal Nehru.
Possessed of a genuine belief in popular sovereignty, and in the equality of India’s people
regardless of religion, caste, or class, Nehru and his allies wrote a constitution that guaranteed
the right to political organization and emphasized the inviolability of democratic procedure. But
although Varshney attributes the birth of Indian democracy to the values of the country’s
founders, he attributes democratic persistence to the fact that India’s political elites have come
to see democracy as being in their interests. In particular, the proliferation of political parties at
all levels of the country has ensured that “there is enough countervailing power available in the
system to oppose the violation” of democratic norms and procedures. Thus, when Indira
Gandhi suspended democracy in 1975 on the pretense of a national emergency, she was only
able to hold the line for nineteen months, eventually giving in to pressure for new elections.
Once democratic political life was resumed, Gandhi was voted out, and Indian political elites
amended the constitution to make future ant-idemocratic trespasses difficult to repeat. Courts
and an independent election commission have helped keep democracy intact, and the
dispersion of power across India’s states and union territories has given many parties a stake in
preserving the regime. The story of India, then, is one of founders motivated by a commitment
to democracy who created a system that eventually gave rise to a highly pluralistic federal
landscape in which democracy was, until recently, by far the dominant game in town. Varshney
notes, however, that the rise of the distinctly illiberal, Hindu nationalist Bharatiya Janata Party
(BJP) has introduced new threats to Indian democracy, and serves as a dramatic illustration of
the extent to which democracy’s survival hinges on political parties’ normative commitments.
Indonesia, the world’s third most populous democracy after India and the United States,
also faced daunting challenges (see Figure 1.6 for Indonesia’s predicted probabilities of
democratic breakdown). That country’s per capita GDP of USD 4,284 may be approximately
double that of India, but it remains in the bottom half of countries. Its population of 267 million
is about one fifth of India’s, but Indonesians speak more than seven hundred languages, live
on one thousand islands scattered across the world’s largest archipelago, and mostly practice
Islam, a faith that some cross-national studies of regime type find to be negatively associated
with democracy. On top of all of this, Indonesia’s democracy, inaugurated in 1998, came after
thirty-two years of a brutal, military-backed dictatorship that had been baptized in the blood of
more than half a million Indonesian communists.

Figure 1.6. Indonesia (2000–2017), probability of democratic breakdown (benchmarked against Spain)

Given these characteristics, argues Dan Slater, Indonesia is vulnerable to four modes of
democratic collapse: a failure of the state, a military takeover, a transition to electoral
authoritarianism, or the rise of a majoritarian, illiberal “democracy” in which minority rights are
trampled. That Indonesia has managed to avoid these pitfalls is, he argues, a function of the
legacies of the pre-authoritarian period. If the story of India is one of democratic legacies
bequeathed by committed democrats, which could only briefly be undone, the story of
Indonesia is one of autocrats who unwittingly erect the scaffolding that later helps to hold up
(an admittedly imperfect) democratic edifice. Slater emphasizes four inheritances, three from
the authoritarian period known as the New Order (1966–98). The first is an encompassing
national identity that helped to instill a sense of belonging among the country’s diverse peoples
and to stave off ethnic polarization and secessionism. The second is the emergence of an
independent, technocratic bureaucracy that would prove capable of governing even after the
removal of the dictator, thus precluding the kind of administrative nonperformance that
routinely sours citizens on new democracies. The third was an autocratic “ruling party” that
remained intact through the transition and which kept former regime satraps invested in the
new democracy by providing them with a channel for participating in it. And finally, the
autocrat’s strategy of dividing and ruling over the military apparatus left it without the ability to
act collectively to abrogate democracy even if it had wanted to.
Slater’s emphasis on how authoritarianism can structure the political landscape in ways that
are later conducive to democratic survival offers a counterpoint to Varshney’s emphasis on the
values and beliefs of political leaders at founding moments. The differing emphases of these
two contributions lay bare a key point of contention that comes up time and again in this
book’s narratives of democratic survival: Though most accounts in this book agree that actors
either keep democracy alive or kill it, they differ in the extent to which they believe that their
democracy-sustaining (or democracy-destroying) behaviors are the result of beliefs versus a
narrowly instrumental calculation of costs and benefits.
In her analysis of democracy’s emergence and survival in Benin and South Africa,
contributor Rachel Riedl takes the latter position. Both countries are hard places for democracy,
as evidenced by the annual breakdown probabilities displayed in Figures 1.7 and 1.8. Benin
(Figure 1.7) is impoverished (with a current GDP per capita of USD [2010] 1,200), ethnically
and religiously diverse, and plagued by low state capacity. South Africa (Figure 1.8) is more
prosperous and has a more modern, effective state, but it grapples with extreme economic
inequality stemming from a long history of white domination of the country’s black majority and
other non-white minorities. Both countries are located in a part of the world that has few long-
lived democracies, further complicating the likelihood of getting and keeping democratic
government.

Figure 1.7. Benin (1992–2017), probability of democratic breakdown (benchmarked against Spain)
Figure 1.8. South Africa (1996–2017), probability of democratic breakdown (benchmarked against Spain)

For Riedl, the emergence and survival of democracy in these two hard places was not a
function of ideological commitments to democratic government on the part of political leaders.
In both places, if incumbent elites had been able to maintain autocracy, or if rising challengers
had proven able to erect autocracies of their own, there is reason to expect they would have.
But in both places, Riedl argues that democracy emerged and was sustained as the result of
pacts by incumbents who understood that the authoritarian status quo was unsustainable and
opponents who realized that economic and social stability required compromise with the ancien
régime. As in Indonesia, democracy survived in South Africa and Benin because the old elite
was given a place in the new political system and thus disincentivized from engaging in
subversion.
Chapter 5, by Lucan Way, explores the fate of democracy in three of the former Soviet
Republics: Georgia, Ukraine, and Moldova (breakdown probabilities displayed in Figures 1.9,
1.10 and 1.11, respectively). According to Way, these cases are instances not so much of
democratic survival amid challenge as of democratic “moments” embedded within otherwise
troubled regime trajectories. Each case counts as a “hard” one by the lights of this book. Even
the case with arguably the best democratic record, Georgia, is a hard place on a number of
dimensions: its per capita GDP of less than USD 5,000 places it in the bottom half of countries;
83 percent of the population is Orthodox Christian and another 11 percent are Muslims, two
religions that scholars have identified as being inconducive to democracy; and around 13
percent of the country is made up of ethnic minorities, including Azeris (6 percent), Armenians
(4 percent), and Russians (1.2 percent) (CIA 2021).
Figure 1.9. Georgia (2005–2017), probability of democratic breakdown (benchmarked against Spain)

Figure 1.10. Ukraine (1995–1997, 2007–2013), probability of democratic breakdown (benchmarked against
Spain)
Figure 1.11. Moldova (1995–2004, 2010–2017), probability of democratic breakdown (benchmarked against
Spain)

In Way’s account, the democratic moments experienced by these three cases are the
results of three factors that act to constrain would-be autocrats: The first is the relative
weakness of the state apparatus, which results in what he calls “pluralism by default.” The
second is the inheritance from the authoritarian period of political parties and independent
media organizations that act to check the power of new regimes. And the third is the looming
specter of Russian influence, which has rendered some of the region’s anti-Russian political
leaders—who might have indulged their authoritarian tendencies—receptive to external
pressures to minimally uphold democracy and the rule of law in order to maintain US support.
One of the strengths of the accounts that attribute democratic survival to the constraints
faced by political elites is that the origins of those constraints are relatively transparent,
generally residing in conditions inherited from the authoritarian past. (Often less clear in these
accounts is why the foundational democratic pacts persist over time, when they do.) In
contrast, those who emphasize the importance of normative commitments to democracy often
do not explain where those commitments came from. The final two chapters in this volume
take this task head-on, offering accounts of democratic survival that pay close attention to the
processes that convert political leaders into committed democrats.
The first of the chapters explains how democracy took root and survived in Timor-Leste, a
small island nation that was occupied successively by Portugal (1702–1975) and Indonesia
(1975–99), against which it fought a decades-long war of independence that claimed more
than one hundred thousand lives. Of all of the cases in this volume, Timor-Leste is one of the
hardest—in addition to its long history of internal war, it remains a poor country with a very
weak state and a high degree of ethnolinguistic diversity. Timor-Leste’s relatively high
probabilities of democratic breakdown since the onset of its democracy in 2002 are displayed in
Figure 1.12.
Figure 1.12. Timor-Leste (2002–2017), probability of democratic breakdown (benchmarked against Spain)

Nancy Bermeo argues that Timor-Leste’s democracy has survived these challenges as a
result of changes to the polity and its leaders that took place over the course of its long war for
independence from Indonesia. In the waning years of the Portuguese colonial period, Timor-
Leste’s political and economic elite were deeply divided over their identities and their
preferences for the country’s future. Some wanted independence, others wanted union with
Indonesia, still others sought union with Australia. Some were communists, others monarchists,
still others Christian conservatives. These cleavages meant that Timor’s political elites were
unable to erect an independent government in the wake of Portugal’s withdrawal in 1975, and
their resulting infighting opened the door for Indonesia’s invasion shortly thereafter.
However, according to Bermeo, the Indonesian occupation and the war to end it set in
motion several processes that ultimately endowed Timor-Leste with the ingredients for its
present democracy. First, the country’s anti-democratic political and economic elites were so
brutalized by the occupation that they either disappeared from the scene or gave themselves
over to a new, democratic mindset. Second, the independence struggle forged a unified and
inclusive national identity among the people of Timor-Leste, enabling them to overcome the
ethnic and ideological cleavages that had contributed to their prior vulnerability to Indonesian
aggression. Third, the war gave rise to a number of political leaders with good reputations and
large constituencies who could convert their renown into votes, resulting in a post-
independence political landscape that was pluralistic rather than dominated by a single group.
Fourth, and finally, the exigencies of the national struggle had forced the militias to coalesce
into a national army divorced from any one political party and subordinate to the authority of
democratically elected leaders.
If Timor-Leste is one of the hardest places for democracy explored in this volume, the final
case study, of Argentina, might appear at first glance to be the easiest. With the highest per
capita GDP among our cases, a relatively capable state, limited ethnic fragmentation, and, by
the 1990s, a mostly favorable neighborhood, that country would appear to be free of some of
the challenges that confront the other countries in this volume. And yet, Scott Mainwaring and
Emilia Simison point out, Argentina’s democracy suffered five breakdowns in the twentieth
century. In the years since democracy was re-established in 1983, it confronted three punishing
economic crises, a dramatic increase in inequality and poverty, several military rebellions, and a
tumultuous period in the close of 2001 in which the country had five presidents in thirteen
days. These economic difficulties are reflected in the country’s elevated probability of
democratic breakdown (compared to Spain) since its transition in 1983 (see Figure 1.13). For
those worried about democratic collapse in established democracies of the Western world, the
case of Argentina is therefore instructive.

Figure 1.13. Argentina (1984–2017), probability of democratic breakdown (benchmarked against Spain)

How did Argentina overcome its authoritarian past and hang onto democracy when other
countries that have faced much lesser economic crises experienced economic breakdowns?
Mainwaring and Simison emphasize the importance of normative commitments to democracy
on the part of the country’s political parties, labor and social movements, and business leaders.
They also emphasize the policy moderation of all key actors—a profound contrast to the
situation during the country’s previous democratic experience of 1973–76. And like Bermeo,
they identify the source of these normative commitments and the shift to policy moderation:
Argentina’s experience with a brutal and inept military dictatorship from 1976 to 1983.
According to Mainwaring and Simison, those years were so catastrophic for Argentina, marked
by “torture, ‘disappearances,’ a reckless and failed war,” and repression of labor unions and
political parties, that political leaders across the political spectrum came to value democracy.
They acted to uphold the democratic system even during severe economic downturns,
hyperinflation, sharp increases in inequality, and dramatic increases in poverty and
unemployment. Thus, when groups of officers engaged in military rebellions during a
particularly difficult period for the Argentine economy, masses and elites were unified in
rejecting them. As Guillermo O’Donnell (1986, 15) wrote thirty-six years ago when assessing
the prospects of Latin American democracies that had recently emerged out of brutal autocratic
regimes, “Largely as a consequence of the painful learning induced by the failures of those
regimes and their unprecedented repression and violence, most political and cultural forces of
any weight now attribute high intrinsic value to the achievement and consolidation of political
democracy.”
 The Importance of Normative Commitments
Each of these chapters demonstrates powerfully that actors matter, but not in the manner
assumed by naïve voluntaristic accounts. It is not the case that democracy is made and kept by
great leaders bound only by their skills and passions. Democratic survival is a process marked
by constraint. In each chapter, we see that actors who could have abrogated or weakened
democracy were prevented from doing so—sometimes by “external” constraints in the form of
institutional shackles or the countervailing power of competing actors, but sometimes by
“internal” constraints, in the form of deeply ingrained normative commitments. In the volume’s
concluding chapter, Scott Mainwaring explores the relative importance of these two forms of
constraint for understanding democracy in hard places.
Although Mainwaring argues that both internal and external constraints are at work in most
cases, he nonetheless makes a powerful argument for greater attention to the former. Formal
institutions such as courts and legislatures do sometimes constrain rulers, helping keep
democracy intact against executive encroachments, but these institutions tend to be weak in
democracies in hard places; presidents and prime ministers can often control, manipulate, or
bypass them. A balance of power among partisan actors is helpful to democratic survival,
forcing actors to accept a democratic compromise. But the balances of power that lead to the
establishment of democracy usually do not persist. In the cases in this volume, the African
National Congress has enjoyed unassailable electoral hegemony since 1994. In Benin, Georgia,
India, and South Africa, parties sometimes won landslide electoral victories. They had the
opportunity to steamroll the opposition, and yet, they refrained from doing so (until 2018 in
Benin).
In short, the external constraints highlighted by the literature are likely neither necessary
nor sufficient for democratic survival. The preferences of actors over regime type matter. Some
actors value the rights and procedures that are defining features of democracy. These clearly
help to sustain democracy. At the other end of the spectrum are committed authoritarians who
believe that different forms of dictatorship—communism, fascism, theocracy, populist illiberal
democracy—are the normative ideal. These actors weaken democracy. Likewise, radical actors
who demand wholesale revisions of the status quo—or who defend the status quo at all costs—
raise the stakes of a democratic game and contribute to the possibility of democratic
breakdown.
Surveying the cases in the volume, Mainwaring finds evidence for the importance of
normative commitments to democracy across the cases. Using data from the Varieties of Party
Identity and Organization (V-Party) project, he shows that none of the cases of democratic
survival in our sample featured highly illiberal political parties for the period when they were
undeniably democratic. The emergence of highly illiberal parties (such as the BJP in India) do
not so much constitute a rejoinder to the notion that normative commitments matter as a clear
warning of heightened risk of democratic breakdown (as seen in Benin in 2019, Moldova in
2001 and 2009, and Ukraine in 2010).
But if normative commitments matter, where do they come from? According to Mainwaring,
they do not simply emerge from thin air or the biographical details of particular leaders. The
collective actors who can make or break democracy—and political parties in particular—develop
programmatic commitments to regime type that they inculcate into members and militants and
which often prove durable over time—just as other programmatic commitments do (Berman
1998). In the case of Argentine parties, this commitment to democracy emerged as part of a
process of repudiating an authoritarian past. In the case of India’s Congress Party, the
commitment to democracy and liberal ideals was shaped by an anti-colonial struggle that had
at its center the liberal ideals of individual rights and freedom from domination. In Timor-Leste,
the commitment to democracy took shape under the shadow of Indonesian occupation and a
bloody civil war. There is every reason to expect commitments forged in this manner to endure.
Though we would not go as far as Berman (1998, 207) to say that ideational motivations are
so powerful as to render “political actors . . . insensitive to changes in their environment and
relatively unconcerned with ‘cost-benefit’ calculations,” we do believe that collective actors’
normative commitments to democracy, once formed, constitute important “internal” constraints
on their behavior. In some instances, these internal constraints can prove as important as the
counterbalancing power of competitors or the shackles of institutions.10

Conclusion
We opened this book with testimonials to the peril in which democracy finds itself around the
world, including—and perhaps especially—in wealthy countries where it was once thought
impregnable. The possibility of democratic collapse where we least expected it has added new
urgency to the age-old inquiry into how democracy, once attained, can be made to last. This
book argues that scholars and practitioners interested in this question can learn much from
democratic survivals that were as unexpected as the democratic erosions currently feared in
some corners of the developed world. Just as social scientists long believed that well-
established, Western, educated, industrialized, and rich democracies were immortal, so too did
they assign little chance of democracy to countries that lacked these characteristics. And yet, in
defiance of decades of social science wisdom, over the course of the last half-century, many
countries that were bereft of the hypothesized enabling conditions for democracy not only got
it but kept it, year after year after year. What is the secret of democratic longevity in such hard
places? What has enabled countries such as India, Indonesia, and South Africa to hang onto
democratic government despite poverty, inequality, and ethnic fragmentation? How did Timor-
Leste forge a durable democracy out of the ruins of civil war? How did Argentina manage to
overcome an authoritarian past and keep democracy alive despite repeated economic
depressions? This book brings these and other unlikely cases of democratic survival front and
center in an effort to derive lessons about what makes democracy stick, especially during
moments of tumult and crisis.
Structure is not a prison. As the cases in this book demonstrate, Rustow (1970) was correct
when he observed that there are no background conditions whose presence is absolutely
necessary for democracy to emerge and survive. Instead, democracy is made and upheld by
collective political actors—parties, executives, movements, militaries—that must make difficult
choices structured not just by country-level factors and international contexts, but also by their
own normative commitments. Presidents facing electoral defeat may abide by the results or use
their control over the instruments of power to distort or ignore them entirely. Military officers
chafing under economic hardship can either respect incumbents or throw them out. Opposition
parties observing mass discontent with a regime can either wait until the next election or go
“knocking on the barracks door” (Stepan 1988, 128). The latter outcomes are more likely in
“hard” places than in “easy” ones, but, as the cases in this volume demonstrate, they are not
predetermined. If the accounts that follow demonstrate that democratic potential exists in
places that dominant structural theories would consign to authoritarianism, so too do they
demonstrate that nowhere should its continued existence be taken for granted. If this volume
is to make any contribution to the global fight to preserve democracy, it is to emphasize for
readers that democracy is nowhere assured and nowhere doomed, but rather lives or dies
depending on what political actors believe and do.
 Notes
1. Diamond, Larry. 2019. “The Global Crisis of Democracy.” The Wall Street Journal, May 17, 2019,
https://www.wsj.com/articles/the-global-crisis-of-democracy-11558105463.
2. In a later contribution, Przeworski (2005, 266–67) finds that the prospects of future electoral victory “are
neither sufficient nor necessary for democracy to survive.” In poor countries, he argues, the income gain
from ruling as a dictator might outweigh the income gain to be had from winning democratic elections,
while the cost of being a loser in both circumstances is the same. In rich countries, in contrast, it may be
better to be a perpetual loser under a democratic system than to risk being a loser under an authoritarian
one.
3. Przeworski (2005, 265) admits that actors may have a “preference for democracy, independently of income,”
but he argues that this would only operate at high levels of income (i.e., not in “hard places). He writes,
“as the marginal utility of consumption declines, the preference for democracy (or against dictatorship)
overwhelms the eventual consumption gain from becoming a dictator.”
4. An important exception to this widely shared finding is Acemoglu et al. (2008).
5. As Ansell and Samuels (2010, 1544) note, Boix (2003) and Acemoglu and Robinson (2006) offer different
predictions about the level of income inequality that is most conducive to democratization, but the logic of
their arguments is similar.
6. Andrew March (2011) offers an alternative perspective on Islam’s compatibility with democracy, and
particularly what he calls “liberal citizenship.”
7. For exact details of coding democratic transitions and breakdowns, see Mainwaring and Bizzarro (2020,
1561, note 3).
8. All marginal effects are calculated using the results in column 1 of Table 1.1, except for the marginal effects
of ethnic fractionalization, the Gini index, and state capacity, which are calculated based on the results in
column 2.
9. Although non-significant, the negative correlation between ethnic diversity and democratic breakdown runs
contrary to expectations. It is premature to speculate about the cause of a correlation that has yet to be
consistently demonstrated, but it could suggest that social diversity renders it harder for single parties or
groups to monopolize power and transform competitive regimes into autocratic ones. This does not mean
that the conventional wisdom—which sees ethnic diversity as a challenge to democracy—is wrong, only
that the difficulty that diversity poses could be in getting democracy, and not in keeping it.
10. In their book, Costly Democracy: Peacebuilding and Democratization After War, Zurcher et al. (2013) find
that the success of democratization after civil wars hinges in part on the extent to which political actors
“demand” democracy, which can stem from a combination of self-interest and normative belief.

Appendix
Table A1.1 Summary Statistics
Variable Obs. Mean Std. Dev. Min. Max.
Breakdown 1,731 0.026 0.159 0 1
GDP (logged) 1,682 8.249 1.135 5.777 10.383
GDP growth 1,675 0.023 0.046 −0.407 0.599
Share of neighboring countries democratic 1,731 0.513 0.127 0.103 0.794
Years since last transition 1,731 12.270 8.953 0 42
Organization of Islamic Cooperation 1,731 0.162 0.369 0 1
Arab League 1,731 0.020 0.141 0 1
Eastern Europe 1,731 0.224 0.417 0 1
Ethnic fractionalization index 1,588 0.436 0.253 0.003 0.889
State capacity 1,118 0.209 0.791 −2.377 2.015
Gini index 1,656 41.567 10.267 21.2 74.3
Another random document with
no related content on Scribd:
 She gently placed little Thu Daw on his feet and took up her position
along the side of his chair, drawing his arm around her.

"When you come to think of it, Cyprian, it was very stupid of us not to
guess that we couldn't live without one another. The night I was frightened,
and you came up and told me about Muriel! Do you remember I wanted
you to marry the vicar's daughter in my sublime faith that her home-made
chocolate fudge would utterly console? And you promised that whoever
you married you'd always love me best. And then, I was the first to marry,
eleven years later. Oh, Cyprian, Cyprian! I want to get out—I want to get
out of the cage."

Her sudden sobbing hurt him.

"Don't, dear. We are out. To me, we seem the great exception to every
rule on earth."

"But earth isn't the end of everything," Ferlie gasped, "and you'll see
that too, some day; even if I were to take advantage of your blindness now.
Is there not one corner in all this wide world where you and I can hide
ourselves with John and Thu Daw and live out our friendship in peace?"

"Dear, you've given me the key and I have turned the lock. The bolt is
on the inside. I yielded up some none-too-solid convictions to your
mysticism; cannot you, in return, yield a little of your mysticism to my
common sense? We are in the twentieth century: it is a day of elastic ideas
for all who have imbibed the plain truth that to live in peace we must let our
neighbours live in peace. The minimum observance of Herd Law, and
Civilization is satisfied. As a woman who has divorced her husband—in the
last divorce bill we can find cause for that without touching upon more
personal reasons—you'd be remarried to a man who had had a son by a
dead Burmese girl. The position would, at least, be comprehensible, even to
the narrow-minded. Why are we torturing ourselves with the precepts of a
dead Jewish Teacher, whose dead words are only kept in evidence by those
to whose interest it is to exploit them?"

He spoke with quick diffidence, pricked by the thought that he was

literally attacking her God.
 The circumstances in which they were living might meet with the
censure of the Church's accumulated wisdom and understanding of human
nature, but he would have had her oppose directly the inflexible word of
One who taught all lovers that His Name is Love.

She pulled herself together and stood up, spreading wide her arms as if
to embrace the light which eluded her.

"This much I will yield, Cyprian. Let me try and find some quiet place
where we can be alone and think. I am glad we have been forced to sever all
artificial connection with our fellow-men. Bless Aunt B.! We have enough
money to go where we please, and when I have chosen a refuge where we
can be, figuratively, apart in the desert, when we have had time to forget the
harsh inharmonies of the past weeks; when there is no need any longer for
us to live under the shadow of a lie, then I will promise you to approach the
whole question with an open mind; to make sure that my sense of values is
rightly adjusted. And there we can decide, at rest in one another's trust,
which of us two in these visionary matters, as you describe them, is finally
to follow and which to lead."

Before he could answer John walked into the room; the broad-brimmed
cow-boy hat which he usually wore for shade, pressed flat against the back
of his head like a plate. The Burman boy, who accompanied him daily on
his afternoon jaunts, silently disappeared as he caught sight of the master
and mistress.

"How funny you look, John," said Ferlie. "Pull your hat on properly if
you are going out."

But John demurred.

"The Lord Jesus wears his topee like this," he informed them pleasantly.

And, "O crumbs!" exclaimed Cyprian collapsing. "Yet another Infant

Samuel!"

But his amusement was short-lived. John had a grievance and had come
to report it.
 "Mother," he said, "there was one tea-party in the Gardens and nobody
didn't let me go to it."

"How do you mean, darling?"

"It was Jimbo's birthday this day. And I tooked him the red coal-truck
what Po Sein did make me to-morrow. And I saw's Jimbo going to the party
and his nurse saw'd me and didn't stop, and Jimbo runned back an' she was
werry angry. An' he said he could not take the truck, 'cos his mother said he
wasn't not to play with me any more. An' he said Derrick's mother said he
wasn't not to play with me neither, an' then his nurse comed up and told Po
Sein we couldn't come up that road 'cos it was Mrs. Grey's party and Jimbo
must go away at once."

Ferlie turned to look blankly at Cyprian. The sins of the fathers...

For the first time in her life he swore thoroughly and completely in her
presence, and without apologizing. Then he pushed back his chair and
swung John up on his shoulder.

"We have no time to think of parties now, you and I. Don't you know
that we are going away in the train? And I do believe you've not packed a
thing."

But when the pair of them had vanished down the verandah, shouting,
Ferlie knelt down beside the baby on the floor whence it was surveying her
with the puzzled concentrated gaze of the man she worshipped.

"Little thing, forgive them!" she whispered. "They, who know not what
they do...."

As soon as it was possible the four of them, and Po Sein, boarded the
evening express for Rangoon, and the house on the hill with the frangipani
trees stood forlornly empty for quite a long time.

* * * * * *
 They had decided to account for themselves verbally and not attempt
the written word. Accordingly, they arrived, unwelcomed, to take up their
quarters in the hotel, and it was agreed that Ferlie should send for Peter,
while Cyprian sought an interview with Maddock.

Eventually, the exact opposite transpired. Peter, wearing a Head-of-the-

Family air, presented himself before Cyprian with "my sister" possessively
decorating his lips, and Ferlie ran the old Colonel, accidentally, to earth on
the yacht, in the meantime.

Just as well, perhaps; for, while Cyprian was quite equal to Peter's lofty
dutifulness, Ferlie was much more likely to prove a match for the Colonel.

He had known her all her life but they had not met since her marriage.
She did not mean to make a Father-Confessor of him, but his mellowness
invited confidence. He had outlived all passionate visions of altering his
neighbours' landmarks and had developed, instead, a distinct sense of
humour.

Ferlie imagined Peter to be lacking a little in that commodity.

"Well, young woman!" was the Colonel's greeting as he unbashfully

embraced her. "So you are playing truant and, likewise, leading the future
Lord John Greville-Mainwaring astray from his ancient heritage. Are any of
us to be enlisted as peace-makers?"

"Peace is my present objective," said Ferlie, "but I do not anticipate that

Black Towers will supply it, Uncle Ricky, even at your invitation."

"What's the trouble, Duckie? I've given up trying to fit square pegs into
round holes at my time of life. I'm a lonely old man and the secrets of a
pretty girl would just about rejuvenate me."

"Yes, you're nice and old," she agreed pathetically; "it's the young who
are so cruel."

"The young! Well, I'm... And who has been accusing you of dyeing that
burning bingled bush? Show me the woman, for it was never no lady!"
 "Uncle Ricky! You've asked Cyprian, John and me to join you. There'll
be a Fourth Child too if we come. Will you be quite serious and listen to me
for at least a quarter of an hour?"

He noted the tired shadows under her eyes and drew her arm through
his.

"You come into my cubby-hole," he commanded.

He heard her out over American iced drinks with fruit floating in them.
He was sane and sea-bronzed and unexclamatory.

"Of course, m'dear," he told her in the end, "the position would just
about have killed your poor mother."

"I can't help being glad that I have been spared the hopeless task of
trying to make darling Mother understand, this side of her tombstone,"
owned Ferlie. "But I've always been sure that if Father had not been so ill,
he would have positively forbidden me to marry that particular Catch of the
Season."

"Ill or well, he never knew, any more than the rest of us, that you cared
for—anyone else," he reminded her.

"That was because Anyone Else wouldn't admit that he cared for me—
no, not even to himself. And I couldn't force him to, though I did try. I knew
we 'belonged.' But there was Peter and Mother and Margery Craven and
Lady Cardew and everyone sighing over my hesitation, and at last it seemed
the only thing to do to yield. Right up to the Wedding Voluntary I wondered
if, perhaps, Cyprian might not rouse up and rescue me. But he only sent me
a golden apple, and not a line with it! I began to believe that I'd mistaken
what I knew was the truth about Cyprian and me."

He leant forward and patted her hand.

"What's finished is finished. The question now is to find the shortest cut
to regularizing the affair. Divorce?"
 "I'm a Catholic, you know."

"So? Most short-sighted of you. I thought it was just another dish Peter
wanted to taste. But he, too, is going to set up a row of names, like ninepins,
for me to knock down. Rude names that suit Biblical Royalty but not the
sort of people one knows. Tut! tut! You were always a complicated couple.
What of our self-restrained hero?"

"Cyprian? He—he is against divorce on principle, but..."

"Quite so! Quite so! Circumstances over which he has no control! By

Gad, I'd take that line myself if you were the woman in the case!"

"Uncle Ricky! You've been a Christopher Columbus all your life. Don't
you know one spot on this troublesome earth where Cyprian and I could
have a peaceful holiday with the babies? He's had to retire, and without
work and nowhere to go—Oh, don't you see we can't come to any
conclusion while we are occupying the situation of living targets to
Society's very natural curiosity?"

"Where on earth do you suppose you'd like to be?"

"Somewhere with sea and sands and open sky and trees and warmth and
loveliness and ..."

"Here, Ferlie! Put the brake on. You want a tropical island, fully
supplied with everything but worry."

She jumped at that.

"An island! Yes. You are going to the Andamans. What are they like?"

"Populated where habitable, alas!"

"Don't tease. There must be an odd one among the number, where we
might picnic."

"Under what flag? I mean, with what suitable excuse? You'd be hunted
up and fêted and asked to Government House.... No, but let me think."
 When he next raised his head she saw a solution had struck him.

"It's a daft scheme enough," he muttered. "But you are just a pair of
lunatics at present, so here goes."

He pulled open a drawer and unrolled a couple of maps, selecting one to

push towards her.

"These are the Nicobars; the nearest of 'em is twenty-four hours' journey
from Port Blair of the Andamans. The Nicobarese are a totally different race
from the Andamanese aboriginals, and a solitary missionary lives among
them obeying the letter of the Christian Law with more spiritual optimism
than horse-sense.

A ship takes his mails from Port Blair every three months, and, once a
year, the Chief Commissioner inspects the tin church on bamboos, the
dozen beehive huts and the wooden shanty which constitutes the Settlement
of Car Nicobar. The missionary has a luxurious wooden house of three or
four rooms and a verandah.

He has to flavour his diet with quinine, and so will you if you join him.
Mind, I don't advocate it, I'm merely putting it to you that, as I know him, I
can run down there—it won't alter our course enough to matter—and
introduce you as friends of mine who—well, desire to study the flora and
fauna of the islands!

He'll offer to put you up at the Settlement, overwhelmed by the prospect

of your company, but of course, if you prefer it, I suppose you can build
yourselves huts, or get the converted sheep of his fold to build them for you
——"

"Oh, huts!" cried Ferlie. "Of course, huts!"

He paused to study her transformed face.

"Bless my soul!" said the Colonel. "And I half expected to be dismissed

as an old fool for my pains!"
 * * * * * *

Cyprian and Peter interrupted them making detailed lists of stores and
physic.

Cyprian had been finding Peter a little difficult; Peter had been finding
Cyprian, to use his own desperate expression, "as little open to reason as
Balaam's donkey."

"Which saw the guiding angel and was prepared to follow it, while
Balaam was only likely to bump his nose against a tree," said Cyprian. And
then resorted to brutality.

"Do you think it in the best of taste to criticize Ferlie and myself when
our destiny has been completely blasted in order that yours might flourish?"
he inquired.

"What are you getting at? Ferlie is a Catholic."

"Driven to become one by a marriage to which she was driven by

necessity for money. Has it never struck you that Ferlie and I meant a great
deal to one another before Greville-Mainwaring appeared upon her horizon
—and her family's?"

Peter looked blank.

"The little idiot never said so!"

"No. She did not consider herself at all, Peter."

"All the same, that being true, you gave her up to him pretty coolly, if I
remember rightly. Even Mother remarked that you seemed to have lost
interest in Ferlie."

"I was cowardly enough to leave that particular decision to Ferlie,"

admitted Cyprian. "I bear my share of all now. But I hardly see where you
come in."
 "What you've told me certainly rather slams the door in my face. I
thought Ferlie had cut loose simply because Clifford was impossible; not
because you were the only possible dispenser of her happiness. But as to
Clifford—have you heard anything about my hypnotic experiments?"

"I know the line you've taken up. It must be hellish work."

"I suppose that's the best way to describe dragging folks out of hell,"
said Peter. "It's not always their own fault that they are there, you know. To
my way of thinking, unfortunate products, like Clifford, and the Vane
woman, are less worthy of censure than you with your Burmese kid."

Cyprian made no reply to this. His close-lipped control appealed to

Peter as could have neither anger nor attempted self-justification.

"Never mind," he added, "you're a damned good fellow, Cyprian, and

you'll be relieved to hear that I'm not going to shove my oar in any more.
As Ferlie's only brother I considered I had no choice but to tackle you. It's a
rotten business, and I am more sorry than I can say for you both; but, after
all, I'm not Ferlie's confessor."

"That's all right, old chap," said Cyprian.

So Peter, though tempted to facetiousness, was inclined to be

encouraging about the island.

"We could leave them the small motor-boat," he suggested to Maddock.

"We never use it, Uncle Rick."

"Ferlie has wheedled more than half the contents of the yacht out of me
already," grumbled the Colonel, who was immensely in his element as the
only genuine man-of-the-world in the party untrammelled by Creed or
Convention. "Look round my cabin, Sterne; mark down what suits you.
Don't mind me."

"I wonder," said Peter, "when we call to fetch them, three months hence,
before the monsoon sets in, whether they'll be tattooed all over and chastely
clothed in the Nicobarese Sunday gear of half a coco-nut and an old top-
hat?"

Thus, by a maintained flow of chaff, the sense of incipient strain in the

atmosphere was dispelled.

Cyprian regarded Ferlie ruefully when she first broke the news.

"And you really want to disappear into this incomparably rural retreat?"

"Don't you think it's a perfect plan? We shall, at least, have time to
breathe."

"Yes, there'll be plenty of time—for everything," he agreed.

One matter troubled the Colonel.

"It'll have to be as Mr. and Mrs. Sterne, you know, Ferlie. Not that that
little Jellybrand is of a suspicious nature—they won't want a passport from
him, hereafter, to prove that of such is the Kingdom of Heaven. But, hang it
all, I'm standing sponsor for you, and I desire to take my undeservedly
unblemished reputation to the grave. And you and Cyprian don't look at one
another with either a brotherly or a sisterly regard. Best make it 'Ferlie
Sterne,' and no questions asked."

"As if it were necessary for me to wear Cyprian's hat to indicate that I

have belonged to him since the world began," stormed Ferlie. "All this
bother comes from Civilization's idiotic habit of changing a woman's name.
I am just 'Ferlie Marguerite' in this Evolution, and that's all about it. I'd
rather be 'Ferlie Cyprian' than 'Sterne'; which is only half his name, because
it certainly wasn't his mother's."

But Uncle Ricky was not going to be drawn into any more Ferlie-esque
controversy.

"In whatever capacity you belonged to Cyprian twelve thousand years

ago," he said, "and in whatever capacity you intend to belong to him when
there is no more marrying and giving in marriage, you're landing on Car
Nicobar as his lawful, or unlawful, wife."

CHAPTER XVII

Robinson Crusoe had had, as yet, little time to do more than leave his
footprints on the stretch of white sand sliding into the summer seas of Car
Nicobar, and the tide washed those out very quickly. Wilder, lovelier, more
brilliant were these islands than the Andamans, while the Andamanese and
the Nicobarese proved to be separate races indeed, having no connection
with one another.

Their very huts and canoes were secular in design. The Nicobarese
names, marking the different islands on the map, fascinated Ferlie. Chowra,
Teressa, Bompoka, Trinkat and Katehal of the Central Group, and Great and
Little Nicobar, with their satellites, Kondul and Pulo Milo in the south,
varied as to dialect in a language so primitive that, as the Reverend Gabriel
Jellybrand mournfully complained, there seemed no words to express either
"gratitude" or "forgiveness." Which abstract nouns once eliminated from
the Christian religion, his sermons became uphill work.

For some years he had toiled in seclusion among this semi-civilized

yellowish flock, of whom it was said that, until Christianity came to their
territory, never one had been known to steal or lie.

Originally acknowledged to be of Indo-Chinese extraction, they were

Malayan-Mongolian in type; often above average height, well-made,
simple, lazy and cowardly, but very good-natured and polite to strangers.

This, then, was the people among whom Ferlie and Cyprian came, with
their respective sons, to reflect upon the sequel to Eden's story and decide
what parts they would play in it themselves.
 Gabriel Jellybrand accepted their arrival as a direct answer to prayer, for
he had been sick of a fever lately and very lonely. His weak green eyes,
rounded like marbles behind an enormous pair of sun-glasses, protruded so
unusually that Ferlie expected them to pop out and hit her when she made
clear her ambition to choose the quietest corner of the quietest island and
erect a row of Nicobarese huts.

The padre invariably fought a losing battle with the letter "w."

"But there is no need to do that," he persisted. "The little bungalow w-

which the forest officer uses on inspection is standing empty. There w-won't
be enough beds, but if you say you've brought hammocks—a most original
idea!"

He led the way to the Settlement, clad in a cotton cassock, tied round
with a black cord which swung out and scourged anybody who ventured too
near him when he held up the garment to leap from rock to rock, revealing
an expanse of white cotton sock below a short tussore trouser-leg.

From time to time he would lose his topee, and then a faithful
bodyguard of Baptismal Candidates, palpably prepared for total immersion
at any moment, would hasten to hand him a large khaki umbrella, lined with
green, which they took it by turns to carry again after the topee had been
recovered from the puddle or overhanging branch which had claimed it.

"You might have warned us that he was a Comic Turn," Cyprian told
Haddock reproachfully.

"Describe him to me again at our next meeting," said the Colonel, for
whom Jellybrand betrayed a pathetically guileless admiration.

"Uncle Ricky has always had a number of the queerest friends," Ferlie
whispered. "Generally they are lame dogs who would have perished in
some one of the world's ditches but for him. This one, at any rate, seems too
nearly an imbecile to take any interest in the obstruse riddles of our
existence."
 The jungle road, strewn with bamboo leaves, twisted them out on to a
cleared space among the coco-nut trees, where a tiny church, which would
have lost itself in an ordinary-sized drawing-room, stood on stilts, as
indeed, did everything in the form of a building, so that, to reach the
platforms on which they were built, it was necessary to run up a short step-
ladder.

The interior of the church was fraught with the pathos of primitive
endeavour. There was a crucifix, smothered in fading hibiscus heads; a
great many common candles, some of which were, nakedly, two shortened
corpses stuck one on top of the other to attain the regulation length; a
canopy of bamboo and coloured paper, enshrining a Madonna with a broken
nose imperfectly mended by means of hot wax, which gave her a somewhat
rakish appearance; the crudest of moral-enforcing missionary prints; a
framed St. Paul (after whom the building was named) so literally "in the
pink" as to impress one instantaneously with his possible value in a football
scrum, and—Oh, lift up your heads!—a small shiny harmonium on which
the little padre had, in the course of a year, just learnt to accompany his
flock with one finger, though he had not yet given up hope of bringing the
other nine into requisition somewhen. For the rest, they relied upon a
catechist with a concertina, who knew seven hymns which he performed in
strict rotation.

"But now that you have come..." said Jellybrand, eyeing Ferlie with
eager expectancy. And she had not the heart to erect a barrier of doctrinal
differences, for her protection, between his hungry enthusiasm and the
harmonium stool.

"I see you have started a shop," said Colonel Maddock emerging from a
thatched fastness, wherein lay heaped up some yards of calico, red flannel,
a few tin pots and pans, coloured prints, packets of tea, pounds of sugar and
several bright glass necklaces.

"Oh, dear me, yes. W-would you believe it? Mr. Pell, here—w-where
are you, Mr. Pell?—a most unselfish person, w-who always acts on the
excellent principle that it is more blessed to give than to receive, w-was left
in charge of the shop w-when I w-was over at Nankauri, visiting. And—w-
would you believe it?—he gave every single thing away w-without taking
any coco-nuts in payment! Of course, the people are ready now to let the
Government start any number of shops...."

Mr. Pell, hitching up the nether garments which were the outward and
visible sign of his inward state of grace, beamed urbanely upon the
newcomers.

"This," said Jellybrand, indicating a gentleman with a string of cowrie

shells about his middle and a battered English straw upon his head, "is
Friend-of-England, the chief of the village. He is not yet a member of my
little community though he w-wishes us w-well, I am sure.

"A long w-while ago, he had forty devils extracted from him by the
menluana, or w-witch doctors, and it is hardly surprising that, under their
conscientious exorcism, he nearly died. All the men in the village sat round
him in a circle and on anyone's perceiving a devil he pounced on it, w-
wrapped it in a leaf, and put it in the corner of the house.

"Every year, the Chief Commissioner, w-when he inspects us from Port

Blair, presents him w-with a suit of clothes in virtue of his position; but he
reserves them for festivals and always puts them away w-when it rains.

"The two beside him are Mr. Corney Grain and Mr. Don Juan. The
traders coming to the islands give them those names in fun, but I hope to
baptize them soon and then they w-will be Peter and Paul—much more
suitable."

Ferlie wondered why.

A man with enormous calves supporting a very small body approached

the speaker confidentially and, presently, Jellybrand interpreted.

"This is James Snook. He and Friend-of-England have both, as you see,

got elephantiasis. A lot of them suffer from it—most unfortunate. He w-
wishes to show you his new house."

Encouraged to proceed, James waved the party towards a thatched

beehive supported on four rickety poles, seeming not to possess, at first
sight, either door or window. However, a ladder discovered underneath the
contraption vanished into a yawning hole, and Cyprian and Ferlie braved
this first, to fall gasping on to a palm-plaited floor which bounded like a
spring mattress beneath them. They could not stand upright and there was a
thick warmth of atmosphere and almost total darkness until Mr. Snook
removed a loose lump of thatch from the wall.

In one corner of the room lay his cooking utensils opposite the rag
bundle on which he slept nightly, after blocking out all oxygen.

The walls were covered with works of art cut from any ancient
illustrated paper which had happened to fall into his hands from the padre's
stock. They were hung, for the most part, upside down, and thus Pavlova
waved mocking legs at a Mission print of Christ crucified, into the Figure
of which the enlightened James had brilliantly bethought himself to
hammer real nails.

Above, glared a garish painting on talc of a Hindu deity with six arms.

Truly, Gabriel Jellybrand stood in need of that optimism which

accompanies the faith of all Heaven's "little children."

* * * * * *

The steam-yacht pushed off again with the evening tide, leaving
Cyprian and Ferlie wandering back from the shore, the coco-nut trees
chattering above their heads in the falling breeze. Occasionally she picked
up Thu Daw whose legs were not, as yet, quite to be trusted, and would
carry him away to the left or to the right, down slight inclines into some
cave of dark foliage.

"And is this Heaven?" asked John, following out some private train of
thought connected with Jellybrand and the concertina, the owner of which
could be heard practising in the distance.

"'Over the fields of glory,' you know, Mother, and 'over the jasper sea'!"
 "Not that Heaven, yet," said Ferlie. "This is more like the Garden of
Paradise in your Hans Andersen."

"I sometimes wish," said Cyprian, "that you and John were not so
inordinately well read."

"Anyway," finished John, after a thoughtful pause, "will I be let sleep in

one of them kennels to-night?"

"There! He has said it." Cyprian rounded on Ferlie. "I trust, Ferlie, you
are going to permit us the unromantic shelter of the forest bungalow, but I
have not really any hope. Of course, if John honestly finds it coincides with
his conception of the Simple Life to share the couch of James Snook—but I
noticed several holes in the roof of the bungalow which should give
primitive colour to the place for our satisfaction and more holes in the floor,
which intuition tells me represent the overcrowded tenement lodgings of
considerable families of snakes and rats."

"Firstly," said Ferlie, "you are sleeping out of doors. We will sling our
hammocks between trees, under the mosquito nets. Secondly, John, you are
not so much as to enter a kennel without leave. I don't know whether that
elephantiasis is catching. Thirdly, an educated man's intuition should assure
him that snakes and rats seldom dwell together in sunny amity."

"Scorpions, then," insisted Cyprian gloomily.

"Fourthly, a perforated roof cannot affect us until the Rains, and there
will be ample time before they break, should we still be here, for you to
have constructed us a model mansion with hot and cold water laid on and
clematis and cabbages before and behind respectively.... Do I hear the sweet
chiming of the village bells?"

"You do," replied Cyprian. "And, if I mistake not, shortly they will be
drowned by the pious fervour of the village choir."

When they emerged at the clearing the bell was, indeed, ringing to
Evensong.
 Jellybrand had added a voluminous surplice and a limp stole to his
attire, but had raptly removed his white canvas shoes instead of his topee, in
absent-minded imitation of his parishioners.

A procession was forming outside the church led by Mr. Pell in the
capacity of cross-bearer; in fact, it was Mr. Pell's top-hat that caused the
padre to recollect that he was being reverent the wrong end for the colour of
his skin, before the red-breeched row of boys proceeded to "survey the
wondrous cross" in the highest key of which the concertina was capable.

The Unsaved of the village were cutting bamboo decorations, in the

background, for a moonlight festival about to be held with the object of
securing a mate for a lonely demon, who had been bringing bad luck to the
community and was expected to depart in peace as soon as a she-demon
could be found to share his flitting.

Ferlie and Cyprian stood some way apart from the two groups of
worshippers while the crimson sunlight streamed across the stems of the
coco-nut palms, and stained the uplifted cross as its followers passed, two
by two, into the toy church and the last notes of the hymn thinned into
silence.

The words were dragged out painfully by the child-voices singing in a

strange tongue of a strange story, brought to them by strangers from another
land.

"His dying crimson like a robe,

Spread o'er His Body on the Tree;
Then am I dead to all the globe,
And all the globe is dead to me."

Ferlie left Cyprian among the lengthening shadows and, herself, moved
slowly across the sunlit space, and so up the wooden steps, into the tiny
forlorn mansion of God.

* * * * * *
 By the time they had decided to settle the bungalow as their
headquarters, and making expeditions in the motor-boat to the neighbouring
islands whenever they desired complete isolation, Cyprian, whose instincts
for research had awakened with the cessation of his ordinary work, brought
Ferlie an oddly carved board covered with paintings and punctured
sketches.

"The padre calls these Henta. They are samples of the form in which the
Nicobarese preserve their legends, and I ask you to look at the tangled
thread of Christianity running through the whole. Jellybrand has been
explaining. The first man offended the Chief of the Spirits at the instigation
of the first woman. The Spirit's name, Deuse, savours of the mission
teaching by your Church. Here, you see Deuse, in arresting négligé,
surrounded by his cooking-pots, and below him huts, trees, birds and
dancing figures. Also mermaids and mermen, because dugongs and whales
are indigenous to these waters. The Henta are made to please the iwi-ka, or
good spirits, and to alarm iwi-pot or demons."

"You are an inquisitive soul, Cyprian. Are you hitting at the Missions,
or what?"

"No. Except that, but for them, in this Eden we should have been able to
escape the story of the Fall. Apart from that teaching, the people believe
their origin dates from the time a shipwrecked sailor was cast up on these
shores with a pet dog. By her he had a son whom she concealed in her leaf
petticoat, or ngong, and this son, on attaining years of discretion, slew his
father and married his mother to continue the race.

"Therefore, the young men wear a bow of coco-nut leaves on their

foreheads, representing the ears of the dog-ancestor, and all dogs are well
treated in her honour."

"And poor little Jellybrand is trying to replace their allegories with

ours? The one they will really appreciate will be Jonah and the whale."

"The most ridiculous part of it all to me is the way that each race,
temporarily in power here, has striven to produce a different form of
Christianity. Think of the waste labour since 1756, after the Danish Mission
departed and was followed by the Dutch. When the Dutch attempt at
colonization failed, the Jesuits and Moravians were inspired to try their luck
till the British took over with Danish consent in order to try and stop the
piracy then rife in these waters. But I don't understand Jelly's line, Ferlie.
He is not a Roman."

"In England he would belong to what was once known as the High
Church Party, but now has developed into Anglo-Catholicism. I can easily
imagine that his successor will probably have ultra-Protestant and anti-
Catholic notions and solemnly smash the pink St. Paul and the battered
Madonna which to-day are the pride of the converts."

"I think he is an amazing specimen," said Cyprian.

"I think he is an amazing duck.... And doesn't he just remind me of a

white rat!"

"Mother," and John broke in upon Cyprian's amusement, "I've found a

house and they say the man in it is going to have a baby. Come and see how
he does it."

Ferlie looked at him.

"The daily round and common task will furnish all you need to ask here,
I can foresee, my son," she told him.

"Dare we investigate?" asked Cyprian, when he had recovered his

gravity. "He may have discovered something interesting in the way of a
rite."

He had. The hut he referred to was, to the islanders, tabu, or forbidden,

since an unfortunate husband, about to become a father, was imprisoned
within it beside his wife, sharing her troubles by sympathetic imitation. He
was deprived of the luxury of a bath and betel-chewing and, for some while
before, he had not been allowed to bind any objects together nor to attend
feasts. Moreover, a whole month must elapse, after the child's birth, before
he would be permitted to escape.
 Friend-of-England, whom they chanced upon in the vicinity, described
the custom in much-broken English, and Ferlie managed to keep a straight
face while Cyprian examined the texture of coco-nut leaves and wished, for
the time being, that he had never been born at all, or, at any rate, that
nobody knew how.

"Ought John to discover these things?" he asked Ferlie later; "isn't it

spoiling his innocence rather young?"

She laughed at him. "I should be sorry to think that you had lost yours
in learning of them," she said. "Nature did not fall with the Fall. But so long
as Mankind is content to lie supine, including all Nature in his own
disgrace, so long will the serpent insinuate horror into the man-made Eden."

"I suppose, in due course, you will interpret; but sometimes I do not
know whether to regard you as a totally unnatural product of Super-
civilization or merely a successful example of the triumph of mind over
matter."

He did not know, either, how to read the look she gave him.

"In a little while, my dear," she said, "you will be able to decide."

* * * * * *

The days fled by; long days of lethargic rest. The children had grown
popular with the Mission, and one lady member in particular had attached
herself to them as a kind of voluntary nurse. This left the two Explorers
time for longer expeditions round the islands than they could have taken
with John and Thu Daw.

One day they lost themselves in a more impenetrable part of the forest,
where ten yards of experimenting with paths was sufficient to shut out the
sky and enclose inexperienced travellers in a leafy tomb. On escaping from
it, stained and breathless, just as they were beginning to think they had
better make their peace with the invisible heavens, preparatory to holding it
for ever, they nearly fell over the bank of an unexpected river.