Content uploaded by Jonathan Pinckney
Author content
All content in this area was uploaded by Jonathan Pinckney on Nov 08, 2023
Content may be subject to copyright.
1
How Civil Resistance Improves Inclusive
Democracy
October 24, 2023
Subindra Bogati Nepal Peacebuilding Initiative
Titik Firawati Northern Illinois University
Jonathan Pinckney The University of Texas at Dallas
Ches Thurber Northern Illinois University
Abstract
When do political transitions lead to greater inclusion for people historically excluded from
power? Scholars and policymakers often assume democratization and political inclusion go
hand in hand. Yet many democracies systematically exclude women, particular ethnic groups,
or lower economic classes. Using data on every political transition from 1945 to 2014, this
study shows that a regime transition’s initiating force critically shapes post-transition political
inclusion. When transitions are initiated through primarily nonviolent civil resistance
campaigns they achieve greater advances in inclusion relative to other types of transition. We
propose three mechanisms to explain this effect: civil resistance leads to greater continued
mobilization and civic activism among the historically excluded, provides greater opportunities
for elites from historically excluded groups to rise to positions of leadership in new regimes,
and forges more pluralistic norms of political behavior.
2
In the late 1970s and early 1980s Peru went through a classic “pacted” transition to
democracy. After a period of military rule, moderates from the regime and opposition
negotiated a return to democracy in a comprehensive national dialogue in 1979 followed by
free and fair elections in 1980 and a peaceful transfer of power in the subsequent presidential
election. By well-accepted metrics of electoral democracy, Peru’s transition was a full
success.1 Yet while the transition established democratic institutions: “vast sectors of the
population were not included in the economic, political, social, and legal systems” (Espinosa,
Janos, and Mac Kay 2023, 218). Economic inequality and ethnic and gender discrimination
soared (Glewwe and Hall 1994). Peru soon found itself facing a violent insurgency whose
rhetoric focused on the continuing exclusion of these populations.
In the same period, Uruguay also went through a democratic transition with many
similar characteristics. In both countries, a military dictatorship negotiated a return to
democracy with well-established political parties. Yet, unlike in Peru, the transition in
Uruguay was pushed forward by a mass civil resistance campaign by labor unions, student
associations, and banned leftist political parties (Gillespie 1985). A series of day-long national
general strikes demonstrated the power of the democratizing forces, eventually drawing in
even the legal political parties. When Uruguay held its first free and fair elections in 1985, the
subsequent regime not only established democratic institutions, but was characterized by
ongoing mobilization from historically excluded groups such as women (Espino 2017) and
the Afro-Uruguayan population (López 2022), which led to both recognition of historical
discrimination and significant improvements in political inclusion.
1 For example, the Varieties of Democracy project (Coppedge et al. 2023) reports a jump from roughly 0.1 to
0.7 in its polyarchy measure, while the Polity IV project (Marshall, Gurr, and Jaggers 2016) reports a jump from
-7 to 7.
3
These contrasting outcomes illustrate a puzzle: why do some transitions lead to
significant advances for groups excluded from political power while others fail to do so?
Countries may democratize, and even adopt de jure political inclusion for all, while still
maintaining significant de facto barriers based on ethnicity, class, or gender. An emphasis on
these de jure institutions in existing studies on democratization makes the current literature ill-
equipped to solve this puzzle.
However, the key difference in these cases – the bottom-up resistance campaign in
Uruguay – suggests one possible answer: the impact of primarily nonviolent civil resistance,
in which ordinary citizens use tactics such as protests, strikes, and boycotts, to push for
political change (Chenoweth and Cunningham 2013). Several scholars have shown that civil
resistance is a powerful tool for political transformation (Chenoweth and Stephan 2011;
Nepstad 2011; Svensson and Lindgren 2011), and an extensive literature already shows that
civil resistance advances de jure measures of democratization (Bayer, Bethke, and Lambach
2016; Celestino and Gleditsch 2013; Chenoweth and Stephan 2011; Fetrati 2023; Pinckney
2020). Furthermore, many of the best-known civil resistance campaigns, such as the anti-
Apartheid struggle in South Africa, have focused on advancing the political inclusion of
historically excluded groups.
Yet other work argues that civil resistance is a “weapon of the resourceful” (White et
al. 2015), and less effective for historically excluded groups (Manekin and Mitts 2021;
Pischedda 2020; Thurber 2018). Additional scholarship has claimed that civil resistance
campaigns are particularly vulnerable to reneging and counterrevolution (Clarke 2022; de
Vogel 2023), suggesting that any gains they achieve may be shallow and ephemeral. Thus,
current studies of civil resistance offer competing perspectives as to whether civil resistance
in political transitions is likely to improve the inclusion of historically excluded groups.
4
This paper addresses both a gap in the democratization literature about when
transitions lead to de facto inclusion as well as a debate in the civil resistance literature about
whether unarmed resistance produces gains for historically excluded groups. We argue that
three mechanisms – the elevation of an array of diverse leaders, the creation of social capital
and repertoires conducive to continued mobilization, and the transformation of political
norms – make transitions initiated through civil resistance like Uruguay’s more likely to
result in greater advances in political inclusion than other political transitions. We expect this
effect to hold across three of the most substantive dimensions of political exclusion: gender,
ethnicity, and economic class.
We test this argument using data on all political transitions from 1946 through 2014
(Pinckney 2020), comparing before and after changes in political inclusion between
transitions initiated primarily through civil resistance and those initiated through all other
means. We use a variety of model specifications and conduct several sensitivity analyses to
mitigate the risk of our results being driven by endogeneity or omitted variable bias. We find
that civil resistance at the beginning of a transition has a strong statistically significant
positive correlation with increases in gender, class, and ethnic inclusion, as well as on an
aggregate index of overall inclusion at the transition’s end. In brief, while most political
transitions on average have had minor improvements in inclusion, civil resistance transitions
(CRTs) are followed by much greater advances in inclusion than non-civil resistance
transitions (non-CRTs).
5
We also find evidence supporting our proposed mechanisms pertaining to leadership
and mobilization. Civil resistance campaigns are associated with increased female legislative
participation and post-transition popular mobilization. Consistent with these mechanisms,
we also find a relationship between visible participation by women and lower economic
classes in civil resistance and subsequent advances in gender and class inclusion, respectively.
We find partial evidence that the participation of excluded ethnic groups generates greater
advances in inclusion.
These results suggest that, while certainly not a panacea, civil resistance is a key force
in bringing about more inclusive politics. Advances in inclusion are always difficult,
incremental, and uncertain, and potentially subject to backlash and reversal. But transitions
precipitated by civil resistance are likely to have advantages in producing gains for previously
excluded groups relative to other types of transitions. While democracy may come about
through a variety of avenues, advances in gender, ethnic, and class inclusion are propelled by
nonviolent regime change from below.
Democracy’s Failure to Promote Inclusion
Political exclusion based on identity is an enduring characteristic of most political systems.
Democratic institutions are an imperfect solution for this problem at best. While such
exclusion tends to be worse in authoritarian regimes, and public discourse in liberal
democracies typically decries such exclusion, even advanced democracies systematically
exclude groups from political power based on their gender, race, or class.2
2 The literature demonstrating this point is extensive. A very small sample of research on various aspects of this
question includes Bateson 2020; Cianetti 2019; Hasan 2011; Htun 2016; Lawoti 2005; Layton and Smith 2017;
Magni and Reynolds 2021; Williams 2000; Wolbrecht and Hero 2005; Young 2002)
6
Despite few remaining official restrictions on women’s suffrage in most countries,
women’s exclusion from meaningful political influence remains widespread. This is
evidenced by a lack of political representation in most parliaments and almost no women in
positions of executive leadership (Paxton, Hughes, and Barnes 2020; Wängnerud 2009). The
position is significantly worse for gender and sexual minority groups (Flores 2021; Magni
and Reynolds 2021; Reynolds 2013).
Economic class is another enduring dimension of political exclusion. Despite the key
role of the working class and organized labor in most democratization struggles (Collier
1999; Pinckney, Butcher, and Braithwaite 2022; Rueschemeyer, Stephens, and Stephens
1992), and the significant influence of redistributivist theories of democratization (Acemoglu
and Robinson 2005; Boix 2003), most democracies suffer from major economic inequality.
This inequality in turn undermines the capacity for the poor and working classes to
meaningfully impact politics. The result is that, across most democracies, the preferences of
the poor and working classes are systematically undervalued. For example, several studies
have shown that political decisions in the United States systematically skew towards the rich
(Bartels 2016; Gilens 2012).
Ethnic or racial identity is also a persistent avenue of political exclusion. The
challenges of democracy in ethnically heterogeneous societies are a matter of long-standing
scholarly debate (Fish and Brooks 2004; Horowitz 1993; Lijphart 1977). Political exclusion
through enduring ethnic or racial hierarchies in turn “enervate democratic citizenship for all”
(Morgan and Kelly 2021, 2023) not just through directly excluding marginalized groups from
access to state resources and political influence but through undermining faith in democracy
across the population.
7
Changing these patterns of political exclusion belies easy solutions. Structures of
exclusion are typically deeply embedded in longstanding political norms, undermining the
will for change. Political leaders from privileged groups stand to lose out from greater
inclusion, and thus have few incentives to advocate for change. And political exclusion often
leads to a lack of resources among excluded groups to mobilize and advocate for themselves.
Regime transitions, periods in which an old political regime has broken down and a
new one has yet to be established (O’Donnell and Schmitter 1986), offer a moment in which
change might seem possible, since they represent a fundamental break in old political
routines. The international community and historically excluded groups both frequently seize
on transitions as times to push for greater inclusion (Acemoglu and Robinson 2001; Haggard
and Kaufman 2016). Yet their efforts are often disappointed. Transitions to democracy
frequently result in new polities dominated by a majority or historically privileged ethnic
group (Bertrand and Haklai 2014; Snyder 2000). Similarly, political transitions have typically
had little impact on patterns of gender inclusion. As Georgina Waylen (2007, 522) writes:
“the majority of transitions have been disappointing in gender terms, bringing few positive
gender outcomes in their immediate aftermath.”3 And even if a political transition results in
democratization, income inequality often endures or even worsens (Acemoglu et al. 2015;
Dorsch and Maarek 2019). How might this problem of enduring political exclusion be
resolved?
3 See also Walsh 2012.
8
The Inclusionary Advantage of Civil Resistance
As we briefly mention above, our answer to this question centers on a new theory of
the long-term political impacts of civil resistance. Civil resistance is a tactical repertoire that
has been employed from the local to the international level, and for many diverse political
objectives. We follow much of the literature in limiting the scope of our inquiry to
“maximalist” campaigns that take place at the national level, have sustained mobilization of
over a thousand participants for at least a week, and have goals that would fundamentally
reshape their country’s political system.
There is abundant evidence that maximalist civil resistance campaigns have
significant political impacts long after the campaign ends, particularly if the campaign is
successful in achieving its short-term goals (Chenoweth and Stephan 2011). In particular, a
growing number of studies have identified a strong relationship between civil resistance and
democratization. Ackerman and Karatnycky (2005) were the first to specify this relationship
quantitatively, finding that a set of transitions preceded by peaceful civic action from 1972 to
2004 were much more likely to democratize. Chenoweth and Stephan (2011) find that even
failed nonviolent resistance campaigns have a positive effect on levels of democracy five
years after the end of a nonviolent resistance campaign, and Pinckney (2020) finds that “civil
resistance transitions” are roughly three times as likely to result in democratization as
transitions initiated through any other means.4
4 See also Bayer, Bethke, and Lambach 2016; Bethke and Pinckney 2019; Celestino and Gleditsch 2013;
Johnstad 2010; Lambach et al. 2020; Rød, Knutsen, and Hegre 2020; and Teorell 2010.
9
The literature to date, following the broader democratization literature, has focused
mostly on the de jure institutional aspects of democracy.5 We expand this focus to theorize on
how civil resistance might impact political inclusion.
First, civil resistance tends to foster a greater culture of pluralism and
accommodation for heterogeneous preferences (Ives 2021). Civil resistance campaigns tend
to be most effective when their movement structures and culture value and incorporate a
wide range of perspectives (Ganz 2009; Polletta 2002). Marshall Ganz’s work in particular
highlights how diverse movements tend to have heightened “strategic capacity” that enables
them to put their resources to more efficacious use. Thus, while valuing pluralism may not
be inherent in the practice of civil resistance, the advantages in a pluralistic approach are
likely to create selection pressures. Those campaigns that can successfully initiate a political
transition will be more likely to be those that have such values.
If they succeed in initiating a political transition, then their adoption of those values
is likely in turn to shape the character of transitional institutions. For example, many
nonviolent resistance campaigns have been followed by comprehensive national dialogues
that established more inclusive political institutions (Dudouet and Pinckney 2021). Such
institutions in turn can lock in more inclusive political arrangements when the transition
comes to an end. As Robert Fishman (2017, 392) writes, “the historically decisive
interactions between civil society actors and political forces during the founding of a
democratic regime forge cultural parameters of political life, which in turn have important
consequences for the ability of social forces to influence agenda-setting, policy-making, and
5 For important recent exceptions, see Fetrati 2023; Ives 2021; and Marks and Chenoweth 2019.
10
other political outcomes under democracy.” The values and practices of pluralism emerging
from civil resistance may reshape the long-term political environment to be more inclusive.
Second, initiating a political transition through civil resistance is likely to diffuse
organizing skills among historically excluded groups that will enable them to continue to
mobilize for their interests in a new political regime. The act of participating in nonviolent
resistance creates both interpersonal ties between individuals as well as new organizational
forms that bring people together over common political goals (Madestam et al. 2013). Once
the campaign is “over” and the transition has begun, these forms of social capital remain.
Organizations formed during civil resistance campaigns can form the basis of a strong civil
society while new social ties facilitate political participation and mobilization (Fernandes
2015). It is almost a truism among social movement scholars that, as Sidney Tarrow (1998)
puts it, “activism begets future activism” (see also Giugni 2004; McAdam 1989). Participants
from historically excluded groups who have activated as part of a civil resistance campaign to
initiate a transition will be loath to return to political inaction and more likely to take the
organizing skills they have learned to advocate for their interests during the transition and
beyond. Civil resistance may thus reshape not just political norms but patterns of political
behavior among the historically excluded, equipping them to mobilize against elite attempts
to maintain old patterns of exclusion.
Examples of this dynamic abound. After participating in the successful 2006
nonviolent revolution against the Nepalese monarchy, the long-disenfranchised Madheshi
community was better equipped to organize their own subsequent series of movements
demanding greater political inclusion in Nepal’s new constitutional arrangements (Jha 2017).
Women’s organizations in Brazil, after having participated in the movement for direct
presidential elections in 1984, later repurposed that movement’s slogan of Diretas Ja (direct
11
elections now!) to successfully call for “family planning now!” (Alvarez 1989, 221). In
Tunisia, the 2011 “Jasmine Revolution” invigorated a “politics from below” regarding
women’s rights that bore fruit in successful campaigns requiring gender parity on electoral
lists, and to prevent women from being described as “complementary” to men in Tunisia’s
post-revolution constitution (Charrad and Zarrugh 2014).
This mechanism parallels and expands on mechanisms in the literature on
democratization and inclusion after civil wars (Berry 2018; Tripp 2015; Wood 2000), in
particular the work of Reyko Huang (2016). Huang shows that rebel governance in civil war
leads to greater postwar democratization because it gives civilian populations the awareness,
engagement, and expectation of political change. This in turn leads to greater mobilization in
the post-civil war context that raises the costs of repression and holds new elites
accountable. These mechanisms are likely to be supercharged in the context of civil
resistance for three reasons. First, the broader tactical repertoire provides a larger set of
potential avenues for popular engagement (Sharp 1973). Second, nonviolent campaigns on
average tend to involve the political activation of a much larger proportion of the population
than civil wars (Chenoweth and Stephan 2011). Third, the primarily nonviolent character of
civil resistance means that transitions after civil resistance campaigns should have fewer of
the social and political traumas associated with civil war.
Finally, civil resistance transitions may advance inclusion by bringing to power a
more diverse political leadership. As described above, civil resistance campaigns have a
strong strategic incentive to build diverse membership (Chenoweth and Stephan 2011).
Participation in the campaign may then result in members of historically excluded
communities taking positions of leadership in the transition more frequently than in top-
down or violent transitions. Such positions are an advance in inclusion in and of themselves,
12
yet their impact goes beyond this, as the individuals in these positions may then be able to
use their offices to advance policies that promote greater inclusion for their constituency.
For instance, in Chile many indigenous leaders participated in the campaign against dictator
Augusto Pinochet, while maintaining their demands for land justice and political rights. In
the subsequent political transition many of these leaders participated in the Special
Commission for Indigenous Peoples (Spanish acronym CEPI), which later resulted in the
passage of a new indigenous law that institutionalized indigenous political participation
(Rodriguez and Carruthers 2008).6
We expect these mechanisms to apply across all three of our dimensions of interest:
gender, class, and ethnicity. Nonviolent resistance campaigns are often led by women, and
provide significant opportunities for women’s inclusion (Chenoweth 2019). The “People
Power” revolution in the Philippines in 1986 took place in response to the attempt by the
Filipino government to deny the presidency to Corazon Aquino. When “People Power” was
victorious, Aquino became the Philippines’ first female president (Reid and Guerrero 1995).
Unions and the working class are also key participants in many nonviolent resistance
campaigns (Butcher, Gray, and Mitchell 2018; Butcher and Svensson 2016), and their
participation tends to lead to greater subsequent democracy (Pinckney, Butcher, and
Braithwaite 2022). The UGTT labor union in Tunisia played an important role in the 2011
Tunisian revolution, and later became a central power player during the political transition,
winning significant concessions for its members (Hartshorn 2019; Niazi 2021). And while
ethnic marginalization may make civil resistance more challenging, many civil resistance
campaigns have been organized by inter-ethnic coalitions. The 1989 “Velvet Revolution” in
6 Though the future development of this institution has been fraught with challenges (Carter 2010).
13
Czechoslovakia gave special positions of influence to the country’s marginalized Slovak
minority, which they were later able to leverage to gain Slovakia’s independence.
For these reasons, we make the following hypotheses about civil resistance and
inclusion:
H1: Transitions initiated through civil resistance will lead to greater advances in
inclusion than transitions initiated through other means.
H1a: Transitions initiated through civil resistance will lead to regimes with
greater gender inclusion than transitions initiated through other means.
H1b: Transitions initiated through civil resistance will lead to regimes with
greater economic inclusion than transitions initiated through other means.
H1c: Transitions initiated through civil resistance will lead to regimes with
greater ethnic inclusion than transitions initiated through other means.
Consistent with our theoretical logic, our hypotheses make predictions about relative
changes or advances in inclusion, not absolute levels. We expect CRTs to produce increases
in inclusion across all three dimensions that are greater than any changes associated with
non-CRTs.
Key Measures and Descriptive Comparisons
We test our hypotheses with an analysis of global cross-national data on political
transitions. Our population of transitions is Pinckney's (2020) list of 315 political transitions
from 1945 to 2014. The transitions in this dataset are identified based on changes in regime
type in the Geddes et al. (2014) Authoritarian Regimes dataset with some additional cases to
move the Geddes et al. original end-date of 2010 forward to 2014. Transitions in the dataset
14
begin the year the old regime type ended according to GWF and end when annual changes
in the V-Dem Polyarchy score fall to below 0.05 for at least two years (Pinckney 2020, 160).
From these 315 cases, 78 are coded as “civil resistance” transitions (CRTs), based on
whether “civil resistance play[ed] a crucial role in breaking down the prior regime” (Pinckney
2020, 16). This determination is based on a combination of examining overlap with
nonviolent resistance campaigns in the NAVCO 2.1 dataset (Chenoweth and Shay 2019) and
original research into each of the transitions. All other transitions are coded as “non-civil
resistance” transitions (non-CRTs). While this includes any type of transition that does not
involve a civil resistance campaign, the vast majority are “top-down” or “pacted”
transitions.7 Since our hypotheses concern whether a transition was initiated through civil
resistance, we directly employ this indicator of a civil resistance transition as our primary
independent variable.
To assess the impacts of political transitions on inclusion, we draw on measures
from the Varieties of Democracy (V-Dem) dataset (Coppedge et al. 2023). Specifically, we
use V-Dem's “Egalitarian Component Index,” which averages measures of equal protection,
access to political power, and distribution of resources from 0 (least egalitarian) to 1 (most
egalitarian).8 We also analyze specific measures pertaining to exclusion based on gender, class
(socioeconomic position), and ethnicity (social group). We subtract their levels from 1 so
that they are scaled in the same direction as the egalitarian index, with 1 representing higher
levels of inclusion.
7 A smaller subset of 33 transitions were directly the result of armed insurgency, either complete rebel victory
or a negotiated peace settlement. We disaggregate these cases in supplementary analyses.
8 For more on this index, see Sigman and Lindberg 2019.
15
It is important to note that these variables attempt to capture concepts related to
political inclusion and exclusion that are not captured in traditional democracy scoring
systems like the Polyarchy index. While democracy metrics might capture the extension of
suffrage to women, they do not attempt to capture (either conceptually or operationally) the
degree to which women have access to primary education, public services, or government
jobs. The same can be said regarding ethnicity and class. The VDEM inclusion measures we
use in this study capture exactly these dimensions of exclusion that are overlooked in studies
of de jure democratization.9
Figure 1 shows trends in the annual averages of each of these four inclusion
measures. They largely move together, increasing over the course of the late 20th and early
21st centuries. Gender inclusion starts lower than the others, but also sees the greatest gains
over the period.
Figure 1: Trends in Annual Averages of VDem Inclusion Measures, 1945-2014
9 An alternative measure considered for this study was the Ethnic Power Relations data which counts the
percentage of the population belonging to ethnic groups excluded from political power. We opted not to use
this measure because of 1) its applicability only to the dimension of ethnicity; 2) its limited focus on
representation in executive political positions as the basis for “inclusion;” and 3) the large fluctuations that
occur in this data when the entire demographic percentage of an ethnic group changes category as a result of
potentially small changes in the composition of the executive.
16
To assess the transition’s impact on inclusion, we compare the V-Dem inclusion measures in
the year before the transition began, to those in the year after the end of the transition (Y1)
as well as those five years after the transition (Y5). In other words, our dependent variable is
the before/after change in the inclusion measure.
Table 1 presents the average changes in these inclusion measures across CRTs and
non-CRTs. Both types of transition produce positive changes. However, the average changes
are consistently higher for CRTs. For the overall equality index, civil resistance transitions
resulted in an average increase of 0.099 while non-civil resistance transitions had an increase
of 0.074. As a benchmark, the average annual change in the equality index across the
timespan 1946-2014 was 0.004. The five-year averages show identical patterns, though the
pace of advancement tapers off to levels closer to historical averages.
Table 1:
Average Changes in Inclusion Measures One Year and Five Years after Transition
Type Total Equality
∆Y1
∆Y5
Gender
∆Y1
∆Y5
Class
∆Y1
∆Y5
Ethnicity
∆Y1
∆Y5
Non-CRT 238 0.076 0.079 0.073 0.086 0.067 0.070 0.073 0.072
CRT 77 0.010 0.106 0.105 0.120 0.094 0.102 0.124 0.133
While this simple cross-tabulation provides initial support for our hypotheses,
numerous factors may correlate both with civil resistance transitions and with greater
advances in inclusion, necessitating multivariate regression analysis to attempt. We now turn
to this analysis
17
Regression Analysis
Since our dependent variables are continuous, our primary testing strategy is linear
OLS regression. Our main models include standard state-level control variables, taken in the
year before the transition. We include V-Dem’s Polyarchy score to account for whether civil
resistance transitions occur in states that are already more democratic. We also include V-
Dem’s measure of civil society participation. High levels of civil society participation could
make civil resistance more likely and also create pressure for inclusion in ways unrelated to
the use of civil resistance. We include per capita GDP from the Maddison Project to control
for possible effects of greater wealth. We also control for the duration of the transition in
years. Because transitions range in length from 1 to 10 years, controlling for duration ensures
that our results are not merely the artifact of some transitions being longer than others.
Longer transitions create more years over which the inclusion measures have a chance to
increase. To control for the impacts of violence that may be occurring alongside any type of
transition, we create a binary indicator for the presence of such violence either during a
transition or in the five years prior. This takes a positive value if there is any armed conflict
in the UCDP/PRIO Armed Conflict Dataset (Gleditsch et al. 2002, Shawn, Petterson &
Öberg 2023) in the country during that time period, or if there is a campaign in the NAVCO
2.1 (Chenoweth and Shay 2021) dataset that is coded as primarily violent, with simultaneous
violence, or with a violent. We also created a binary measure for any prior CRT in the
country.
Finally, we include two measures to control for both absolute levels as well as trends
in inclusion prior to the transition. The first measure is the country’s inclusion index (overall,
gender, ethnicity, or class) in the year prior to the transition. The second is the annualized
18
average change in the inclusion index in the country over the five years prior to the
transition. This allows us to control for the possibility that civil resistance transitions emerge
in polities that had already begun to experience increases in inclusion. In the Online
Appendix, we present descriptive statistics of each of these control variables, disaggregated
by transition type.
Table 2 presents the results of our primary regression models. Figure 2 shows the
marginal effects of civil resistance transitions, while Figure 3 shows the expected changes in
inclusion for both civil resistance and non-civil resistance transitions. The results provide
strong support for the positive relationship between civil resistance transitions and greater
advances in inclusion. The coefficient for civil resistance transitions (compared to the
reference category of non-CRTs) is consistently positive and statistically significant across all
dimensions of inclusion. Figure 2 shows that across the four dependent variables, CRTs are
correlated with increases in inclusion that are between 0.039 and 0.052 higher than the
increases that accompany non-CRTs. Figure 3 provides greater overall context. It shows that
across these dimensions, non-CRTs produce an expected increase of between 0.07 and 0.08.
This is a substantively large increase given the overall 0-1 scale of the indices. But the
increase after CRTs is even larger: between 0.11 and 0.13.
In the Appendix, we report models based on inclusion levels 5 years after the end of
the transitions that show the inclusionary advantage of CRTs is maintained over five years.
We also further explore the role of violence, disaggregating 33 non-CRTs in which the
transition was directly precipitated by an armed insurgency (either through outright rebel
victory or a negotiated peace settlement). We find those “primarily violent” transitions
experience higher average increases in inclusion than the other non-CRTs. However, the
small number of cases results in high levels of uncertainty. This suggests that these violent
19
transitions possibly activate some of the same mechanisms that CRTs do, such as bringing
more diverse leaders into power or creating social capital for continued collective action.10
Figure 2:
Coefficient estimates (marginal effects) from full models of civil resistance transition on change in
inclusion measures from before to one year after transition. Coefficients represent how much greater an increase
in inclusion is expected from a CRT as compared to a non-CRT.
Figure 3:
Expected changes in inclusion measures for CRTs versus non-CRTs, controls held at their
means.
10 The dataset of democratic transitions contains too few of such cases for further quantitative exploration.
However, work on post-civil war democratization provided insights into the dynamics at play (Huang 2016)
20
Table 3:
OLS Regressions of Effect of Transition Type on Changes in Inclusion, Before vs. One Year
After Transition
Model 1: Equality Model 2: Gender Model 3: Class Model 4: Ethnicity
CR Transition 0.042** 0.048** 0.039* 0.053**
(0.016) (0.017) (0.016) (0.017)
Prior Inclusion -0.131*** -0.082** -0.082** -0.073*
(0.038) (0.029) (0.027) (0.029)
Prior Trend 0.828*** 0.816*** 0.825*** 0.820***
(0.074) (0.088) (0.063) (0.061)
Duration 0.009* 0.010* 0.010+ 0.005
(0.004) (0.004) (0.005) (0.004)
VDem Polyarchy 0.010 0.040 0.029 0.000
(0.047) (0.050) (0.044) (0.053)
Civil Society -0.068* -0.076* -0.044 -0.077*
(0.028) (0.032) (0.029) (0.034)
GDP pc (log) 0.000 0.001 0.001 0.001
(0.001) (0.001) (0.001) (0.001)
Violence 0.008 0.007 0.010 0.007
(0.011) (0.012) (0.012) (0.011)
Prior CRT 0.002 0.013 0.009 0.020
(0.013) (0.013) (0.011) (0.014)
Cold War 0.009 0.021 0.040** 0.016
(0.016) (0.016) (0.015) (0.016)
Num.Obs. 276 273 273 273
R2 0.609 0.529 0.583 0.624
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
HC1 robust standard errors used across models. Prior Inclusion and Prior Trend covariates are specific to the
inclusion measure DV and vary between models. CR Transition is a dichotomous variable indicating a Civil
Resistance Transition (CRT) as compared to a non-Civil Resistance Transition (non-CRT).
21
Mechanisms and Additional Observable Implications
Our theory proposed three mechanisms through which civil resistance transitions might
produce greater gains in political inclusion: the elevation of leaders from historically
excluded groups, increased mobilization of historically excluded groups, and more inclusive
values and norms. In this section, we test for evidence consistent with the first two of these
mechanisms.
If the leadership mechanism is true, we should expect CRTs to result in more diverse
governments than non-CRTs. We can measure this outcome on the gender dimension using
V-Dem’s “Women’s political participation index,” which measures the degree to which
women are descriptively represented in the legislature and overall distribution of power
(Coppedge 2023).11 If the social capital and repertoires mechanism is true, we should expect
more frequent protest in the aftermath of CRTs than non-CRTs. To capture this, we turn
again to V-Dem, using their “Mass mobilization” scale that measures protest events such as
demonstrations, strikes, and sit-ins. Unfortunately, we were not able to find an adequate way
of measuring changes in norms for this study, or political participation beyond the gender
dimension.12
These variables employ the same 0 to 1 interval scale used in the earlier inclusion
analyses, thus we use OLS regressions in a similar structure to our previous models. For
women’s representation, we once again use as our dependent variable the difference in thes
measure one year after as compared to the year before the transition. For mobilization, we
compare to five years before the transition, so as not to count mobilization that was part of
11 Admittedly, representation in a legislature is a facet of inclusion itself. But we also view it as a pathway
through which other facets of inclusion, such as equal protections and equal access to resources, is achieved.
12 For inclusive values, we explored using World Values Survey measures, but could not find questions that
were asked consistently across countries and waves that aligned with our concept. For ethnic participation, we
again considered the use of EPR, but decided against it due to its limited focus on executive representation.
22
the transition itself. In both cases, we use the same binary indicator of CRT and same state-
level control variables. For the women’s political representation model, as before, we control
for the absolute level of women’s political representation in the year prior to transition, as
well as the average change in this measure over the five years leading into the transition. In
the mobilization model, we use the absolute level of mobilization five years prior to the start
of the transition, for the same reasons described earlier, hoping to capture baseline levels of
mobilization prior to the civil resistance campaign. We do not include any rolling average
trend.
Table 3 presents the results of these two models. In both cases, we see correlations
consistent with the proposed mechanisms. CRTs are associated with an increase in women’s
political representation that is 0.046 greater than that associated with non-CRTs. Meanwhile,
for countries that experience a CRT the model expects levels of mobilization 0.548 greater
than for countries that experience a non-CRT.
23
Table 3:
OLS Regressions of Effect of Transition Type on Changes in Women’s Political Representation
and Mass Mobilization, Before vs. One Year After Transition
Model 5:
Women’s Political
Participation
Model 6:
Post-Transition
Mass Mobilization
CR Transition 0.046* 0.548**
(0.020) (0.169)
Prior Pol. Rep. -0.311***
(0.052)
Prior Mobilization
(5-years prior) -0.555***
(0.055)
Prior Trend 0.555***
(0.073)
Duration 0.013* 0.037
(0.006) (0.045)
VDem Polyarchy -0.166+ -0.634
(0.090) (0.598)
Civil Society 0.055 0.016
(0.062) (0.416)
GDP pc (log) 0.006* 0.005
(0.002) (0.016)
Violence -0.013 0.357**
(0.015) (0.136)
Prior CRT 0.035 -0.364*
(0.025) (0.178)
Cold War -0.070** -0.400*
(0.025) (0.196)
Num.Obs. 255 270
R2 0.460 0.374
24
The Importance of Minority Participation
An additional observable implication of our theory is that we should expect the
greatest advances in inclusion to come when members of marginalized groups participate in
the civil resistance campaign that drives a CRT. In terms of leadership, members of
historically excluded groups are more likely to rise to power in the aftermath of a transition
when they were leaders of the movement that brought about regime change. They are also
only likely to be able to build social capital through civil resistance if they participated in
direct actions. And finally, by participating in the campaign, they can better bring their issues
to the center of national debate, thus propelling changes in values that favor inclusion.
While theories of civil resistance posit that nonviolent campaigns are likely to attract
participation from minority groups, the historical record shows that this is not always the
case. For example, only forty-seven percent of civil resistance campaigns feature
participation from more than one ethnic group (Thurber 2018, 260). There is also significant
variation in the participation of workers and the economically disadvantaged in civil
resistance (Dahlum, Knutsen, and Wig 2019). And while some participation by women is
prevalent in most civil resistance campaigns, this participation is most commonly occasional,
rather than a regular part of the campaign (Chenoweth 2019).
To test the impacts of participation by excluded groups in civil resistance on future
levels of inclusion, we disaggregate civil resistance transitions by whether they included
participation by women, lower economic classes, and excluded ethnic groups. Data on
female participation comes from the NAVCO 2.1 dataset’s “gender diversity” measure
which takes the value of 1 when a campaign included significant observable participation by
women (Chenoweth and Shay 2022). For class, we again use NAVCO 2.1’s parallel measure
25
of “class diversity.” This takes the value of 1 when there is observed diversity in terms of
class. Finally, for ethnicity, we use the Ethnic Groups in Contention dataset (Thurber 2018)
to determine whether or not at least one politically excluded ethnic group participated in the
campaign.
Table 4 breaks down the sample of CRTs by participation in terms of gender, class,
and ethnicity, along with the average changes for the related measure of inclusion. We do
not conduct any analysis using the overall equality index as there is no single specific type of
participation to link it to. Most CRTs included participation by women and by diverse social
classes. However, the CRTs that did not feature such participation experienced smaller
average changes in inclusion. In the case of ethnicity, the breakdown of participation was
more balanced. But, again, CRTs with participation by excluded ethnic groups experienced
larger average gains in ethnic inclusion.
Table 4:
Average Changes in Inclusion Measures by Group Participation
Type
Gender
N
∆Y1 ∆Y5
Class
N
∆Y1 ∆Y5
Ethnicity
N
∆Y1
∆Y5
CRT No
Participation
11
0.076
0.089
8
0.023
0.026
35
0.101
0.112
CRT w/
Participation
57
0.113
0.126
60
0.108
0.114
29
0.138
0.146
As before, we turn to regression analysis both to assess statistical significance as well as to
control for confounding factors. In these models, our independent variable takes three
levels: a non-CRT, a CRT with no participation from the excluded group, and a CRT with
participation. The results are reported in Table 5. Figure 4 presents a plot of pairwise
26
marginal effects contrasting all three levels of the transition participation variable, while
Figure 5 presents expected values.
Figure 5 shows that when historically excluded groups participated in civil resistance
campaigns, the subsequent transition produced advances in inclusion between 0.12 and 0.14
on the relevant V-Dem scale. By contrast, when CRTs did not include women or members
of lower social classes, the increases were only about half that, 0.07 and 0.05, respectively,
roughly similar to the gains achieved by non-civil resistance transitions. The one exception to
the trend is ethnic participation, where campaigns that did not include members of excluded
ethnic groups nevertheless produced advances in inclusion of 0.12, significantly
outperforming non-civil resistance transitions. But campaigns that included participation
from ethnic minorities still yielded greater advances of 0.14.
Figure 4 shows the pairwise marginal effects between non-CRTs, CRTs with no
participation from historically excluded groups, and CRTs with participation from
historically excluded groups. In particular, the bottom-most set shows that CRTs with
participation by the excluded group achieved higher rates of inclusion across all three
dimensions. However, this difference was only statistically significant in the case of gender.
Overall, the results are consistent with our expectations about the importance of minority
participation in a civil resistance movement to activate the mechanisms that yield post-
transition advances in inclusion.
27
Table 5:
Regressions of Campaign Participation and Transition Type
Model 9: Gender Model 10: Class Model 11: Ethnicity
CRT No
Participation -0.006 -0.017 0.051*
(0.020) (0.019) (0.021)
CRT w/
Participation 0.065** 0.051** 0.071*
(0.022) (0.019) (0.028)
Prior Inclusion -0.096** -0.091** -0.075*
(0.031) (0.029) (0.029)
Prior Trend 0.812*** 0.816*** 0.806***
(0.088) (0.064) (0.066)
VDem Polyarchy 0.031 0.029 0.006
(0.051) (0.044) (0.055)
Civil Society -0.068* -0.043 -0.078*
(0.033) (0.030) (0.035)
GDP pc (log) 0.000 0.001 0.001
(0.001) (0.001) (0.001)
Violence 0.005 0.007 0.009
(0.012) (0.012) (0.012)
Prior CRT 0.017 0.007 0.017
(0.013) (0.012) (0.014)
Cold War 0.025 0.041** 0.021
(0.016) (0.015) (0.016)
Num.Obs. 266 266 264
R2 0.535 0.590 0.596
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
HC1 robust standard errors used across models. Prior Inclusion and Prior Trend
covariates are specific to the inclusion measure DV and vary between models.
28
Figure 4:
Pairwise marginal effects of contrasts between CRTs that included participation from members of
each excluded group, CRTs that did not, and non-civil resistance transitions. Participation is specific to the
dimension being evaluated: i.e. the “gender” model codes for the participation of women, the “ethnicity” model
for the participation of excluded ethnic groups, and the “class” model for class diversity.
Figure 5:
Expected changes in inclusion measures for CRTs with excluded group participation, versus
without, versus non-CRTs. Participation is specific to the dimension being evaluated: i.e. the “gender” model
codes for the participation of women, the “ethnicity” model for the participation of excluded ethnic groups, and
the “class” model for class diversity.
29
Robustness and Sensitivity
Our reliance on cross-national observational data makes our findings vulnerable to
omitted variable bias. In particular, it is possible that civil resistance transitions are more
likely in countries and at times that are otherwise ripe for advances in gender, class, and
ethnic inclusion. Our basic regression approach tried to limit this risk by comparing only
cases where a political transition took place, by controlling for several structural variables
such as regime type, and including both lagged measures of the dependent variable as well as
the preceding 5-year trend. To further test for possible temporal and geographic
dependencies, we run additional models in the appendix that include regional fixed effects as
well as a continuous variable of the year of transition.
Finally, we conduct sensitivity analyses to assess how strong of an omitted
confounding variable it would take to overturn our finding. Using the “sensemakr” package
in R (Cinelli and Hazlett 2020), we find that in Model 1 (Table 3), an unobserved confounder
would have to explain 17.5% of the residual variance of both our independent and
dependent variable in order to produce a coefficient estimate for civil resistance transitions
of zero. An unobserved confounder that explained 6.5% of the residual variance in each
would make our results drop below statistical significance at the p < 0.05 level. Such an
omitted variable would also have to be orthogonal to our existing control variables, including
the lagged dependent variable as well as the 5-year trend.
As a benchmark, we can compare a hypothetical unmeasured confounder to one of
the covariates in our analysis, such as the lagged dependent variable for the equality index.
We show in Figure 6 that it would take an unmeasured confounder roughly three times as
strong (in terms of its ability to explain residual variance in both X and Y) as the lagged
30
equality term to produce a result that was no longer statistically significant (t < 1.97). In the
Appendix, we present parallel analyses for Models 2-4 in Table 3, yielding similar results.
Figure 6:
Changes to T-value for Civil Resistance Transition coefficient estimate from a hypothetical
omitted confounder 1x/2x/3x as strong as the lagged dependent variable.
Discussion and Conclusion
Dissidents from communities excluded based on their gender, ethnicity, or class have
long recognized that achieving the visible markers of democracy is by no means a sufficient
condition for ensuring a just and equal society. Even when leaders are chosen through free
and fair elections, the leaders elevated through those elections to positions of power are
most typically members of dominant identity groups and use their power to pursue the
interests of those groups.
Our results indicate that one key way of changing this picture comes via the
mechanism through which old regimes break down and new ones come into being.
Transitions achieved through the mass popular mobilization of a maximalist civil resistance
campaign result in systematically greater increases in inclusion across the dimensions of
gender, class, and ethnicity, and these effects endure for years after the transition ends. This
contradicts anecdotal examples and recent research that has suggested that civil resistance is
31
particularly “ineffective” for marginalized groups, especially ethnic minorities. In fact, along
the dimension of ethnicity, where prior literature would suggest that civil resistance
transitions might struggle to produce gains, we find the greatest observed difference between
transitions.
We find evidence as well that the impact of civil resistance campaigns on political
transitions can at least in part be explained by civil resistance’s impact elevating leaders from
historically excluded groups and diffusing civic mobilization more broadly throughout
society. We theorize, based on anecdotal evidence and previous research, that civil resistance
also significantly leads to more inclusive norms of political behavior, but leave the direct
testing of this mechanism for future research as we lack systematic data with which to test it.
The evidence further suggests that participation in civil resistance by excluded groups
has important effects on inclusion, though there is less certainty in these results due to small
sample sizes. Much of the general effect of civil resistance transitions on inclusion can be
explained by that subset of transitions that featured participation by women, lower social
classes, and (to a lesser extent) excluded ethnic groups. This supports evidence from the
examples cited earlier in our paper showing that participation in civil resistance by excluded
groups can elevate leaders from those groups and diffuse social capital and organizing
capacity in ways that promote greater inclusion.
Civil resistance is also by no means the only factor affecting patterns in political
inclusion. Indeed, the strongest factor consistently predicting higher inclusion is the time
trend. Improving global norms, including the general trend of better human rights
protections (Fariss 2014), likely play a central role here. Yet despite the impact of these other
factors, the effect of civil resistance remains substantive. For example, the marginal effect of
32
a civil resistance transition on the gender inclusion index (0.04) is equivalent to Nepal’s
improvement on gender inclusion following the legalization of abortion in 1999.
It is also important to consider the scope of these results in terms of the comparative
baseline. We are examining political transitions, and the factors that shape the character of
the political regimes that follow them. In this context, it appears that civil resistance has a
significant positive effect. While we consider it reasonable to look at this as a piece of
evidence for a broader effect of civil resistance on inclusion of excluded groups, these results
do not directly show that such an effect obtains outside of the conditions of political
transitions. Could civil resistance campaigns produce advances in inclusion even when they
fail to precipitate regime change? How do the outcomes of civil resistance campaigns in
terms of inclusion compare more generally to the outcomes of armed insurgencies? We leave
such questions for further research. Future studies might also attempt to parse which of the
specific explanatory mechanisms have the greatest effect.
In terms of practical and policy recommendations, we echo longstanding calls to take
seriously the promise of civil resistance to achieve major political transformation.
Policymakers, practitioners, and scholars have long been interested in post-civil war regime
trajectories. This is understandable given the threat of reversion to violence. But post-war
contexts represent a small percentage of cases of regime transition or democratization. Civil
resistance transitions are both far more common and have yielded substantial advances in
political inclusion.
Finally, the concept of inclusion is not merely academic. It represents meaningful
improvements in the lives of members of historically excluded social groups. It also aligns
with types of structural inequalities that are associated with an increased risk of political
violence (Cederman, Gleditsch, and Buhaug 2013). Civil resistance is thus not simply a
33
means of achieving regime change without violence; it has the potential to address the
underlying grievances that fuel conflict.
34
References
Acemoglu, Daron, Suresh Naidu, Pascual Restrepo, and James A. Robinson. 2015.
“Democracy, Redistribution, and Inequality.” In Handbook of Income Distribution, eds.
Anthony B. Atkinson and Francois Bourguignon. Amsterdam, Netherlands, 1885–
1966.
Acemoglu, Daron, and James A. Robinson. 2001. “A Theory of Political Transitions.”
American Economic Review 91(4): 938–63.
———. 2005. Economic Origins of Dictatorship and Democracy. New York, NY: Cambridge
University Press.
Ackerman, Peter, and Adrian Karatnycky. 2005. “How Freedom Is Won: From Civic
Resistance to Durable Democracy.” Washington, DC: Freedom House.
Alvarez, Sonia E. 1989. “Politicizing Gender and Engendering Democracy.” In Democratizing
Brazil: Problems of Transition and Consolidation, ed. Alfred C. Stepan. New York, NY:
Oxford University Press, 205–51.
Bartels, Larry M. 2016. Unequal Democracy: The Political Economy of the New Gilded Age. 2nd ed.
Princeton, NJ: Princeton University Press.
Bateson, Regina. 2020. “Strategic Discrimination.” Perspectives on Politics 18(4): 1068–87.
Bayer, Markus, Felix S Bethke, and Daniel Lambach. 2016. “The Democratic Dividend of
Nonviolent Resistance.” Journal of Peace Research 53(6): 758–71.
Berry, Marie E. 2018. War, Women, and Power: From Violence to Mobilization in Rwanda and
Bosnia-Herzegovina. New York, NY: Cambridge University Press.
Bertrand, Jacques, and Oded Haklai. 2014. “Democratization and Ethnic Minorities.” In
Democratization and Ethnic Minorities: Conflict or Compromise?, eds. Jacques Bertrand and
Oded Haklai. New York, NY: Routledge, 1–17.
Bethke, Felix S., and Jonathan Pinckney. 2021. “Nonviolent Resistance and the Quality of
Democracy.” Conflict Management and Peace Science 38(5): 503–23.
Boix, Carles. 2003. Democracy and Redistribution. New York, NY: Cambridge University Press.
Butcher, Charles R., John L. Gray, and Liesel Mitchell. 2018. “Striking It Free? Organized
Labor and the Outcomes of Civil Resistance.” Journal of Global Security Studies 3(3).
Butcher, Charles, and Isak Svensson. 2016. “Manufacturing Dissent: Modernization and the
Onset of Major Nonviolent Resistance Campaigns.” Journal of Conflict Resolution 60(2):
311–39.
35
Carter, Daniel. 2010. “Chile’s Other History: Allende, Pinochet, and Redemocratisation in
Mapuche Perspective.” Studies in Ethnicity and Nationalism 10(1): 59–75.
Cederman, Lars-Erik., Gleditsch, Kristian Skrede., Buhaug, Halvard. 2013. Inequality,
Grievances, and Civil War. New York: Cambridge University Press.
Celestino, Mauricio Rivera, and Kristian Skrede Gleditsch. 2013. “Fresh Carnations or All
Thorn, No Rose? Nonviolent Campaigns and Transitions in Autocracies.” Journal of
Peace Research 50(3): 385–400.
Charrad, Mounira M., and Amina Zarrugh. 2014. “Equal or Complementary? Women in the
New Tunisian Constitution after the Arab Spring.” The Journal of North African Studies
19(2): 230–43.
Chenoweth, Erica. 2019. Women’s Participation and The Fate Of Nonviolent Campaigns. One Earth
Future. https://oefresearch.org/publications/women%E2%80%99s-participation-
and-fate-nonviolent-campaigns-report (February 23, 2021).
Chenoweth, Erica, and Kathleen Gallagher Cunningham. 2013. “Understanding Nonviolent
Resistance: An Introduction.” Journal of Peace Research 50(3): 271–76.
Chenoweth, Erica, and Christopher Wiley Shay. 2022. “Updating Nonviolent Campaigns:
Introducing NAVCO 2.1.” Journal of Peace Research 59(6): 876–89.
Chenoweth, Erica, and Maria J. Stephan. 2011. Why Civil Resistance Works: The Strategic Logic of
Nonviolent Conflict. New York: Columbia University Press.
Cianetti, Licia. 2019. The Quality of Divided Democracies: Minority Inclusion, Exclusion, and
Representation in the New Europe. Ann Arbor, MI: University of Michigan Press.
Cinelli, Carlos, and Chad Hazlett. 2020. “Making Sense of Sensitivity: Extending Omitted
Variable Bias.” Journal of the Royal Statistical Society: Series B (Statistical Methodology) 82(1):
39–67.
Clarke, Killian. 2022. “Revolutionary Violence and Counterrevolution.” American Political
Science Review: 1–17.
Collier, Ruth Berins. 1999. Paths Toward Democracy: The Working Class and Elites in Western
Europe and South America. New York, NY: Cambridge University Press.
Coppedge, Michael et al. 2023. “V-Dem Country-Year Dataset V13.” V-Dem Country-Year
Dataset v13. https://www.v-dem.net/data/the-v-dem-dataset/.
Dahlum, Sirianne, Carl Henrik Knutsen, and Tore Wig. 2019. “Who Revolts? Empirically
Revisiting the Social Origins of Democracy.” The Journal of Politics 81(4): 1494–99.
Davies, Shawn, Therese Pettersson & Magnus Öberg. 2023. "Organized violence 1989-2022
and the return of conflicts between states?" Journal of Peace Research 60(4): 691-708.
36
Dorsch, Michael T., and Paul Maarek. 2019. “Democratization and the Conditional
Dynamics of Income Distribution.” American Political Science Review 113(2): 385–404.
Dudouet, Veronique, and Jonathan Pinckney. 2021. Nonviolent Action and Transitions to
Democracy: The Role of Inclusive Dialogue and Negotiation. United States Institute of Peace.
Peaceworks. https://www.usip.org/publications/2021/07/nonviolent-action-and-
transitions-democracy-impact-inclusive-dialogue-and (July 7, 2021).
Espino, Alma. 2017. “Gender Equality in Uruguay.” In Gender Equality in a Global Perspective,
eds. Anders Ortenblad, Raili Marling, and Snjezana Vasiljevic. New York, NY:
Routledge, 126–51.
Espinosa, Agustín, Erika Janos, and Martín Mac Kay. 2023. “Indigenous Political
Participation in Peru: A History of Racism, Exclusion, and Violence.” In Indigenous
Political Representation in Latin America, eds. Adrian Albala and Alejandro Natal. Cham,
Switzerland: Springer Nature, 201–32.
Fariss, Christopher J. 2014. “Respect for Human Rights Has Improved over Time: Modeling
the Changing Standard of Accountability.” American Political Science Review 108(2):
297–318.
Fernandes, Tiago. 2015. “Rethinking Pathways to Democracy: Civil Society in Portugal and
Spain, 1960s–2000s.” Democratization 22(6): 1074–1104.
Fetrati, Jalal. 2023. “Non-Violent Resistance Movements and Substantive Democracy.”
Democratization 30(3): 378–97.
Fish, M. Steven, and Robin S. Brooks. 2004. “Does Diversity Hurt Democracy?” Journal of
democracy 15(1): 154–66.
Fishman, Robert M. 2017. “How Civil Society Matters in Democratization: Setting the
Boundaries of Post-Transition Political Inclusion.” Comparative Politics 49(3): 391–409.
Flores, Andrew R. 2021. Social Acceptance of LGBTI People in 175 Gountries and Locations: 1981
to 2020. Los Angeles, California: UCLA Williams Institute.
https://williamsinstitute.law.ucla.edu/wp-content/uploads/Global-Acceptance-
Index-LGBTI-Nov-2021.pdf.
Ganz, Marshall. 2009. Why David Sometimes Wins: Strategic Capacity in Social Movements. New
York, NY: Oxford University Press.
Geddes, Barbara, Joseph Wright, and Erica Frantz. 2014. “Autocratic Breakdown and
Regime Transitions: A New Data Set.” Perspectives on Politics 12(2): 313–31.
Gilens, Martin. 2012. Affluence and Influence: Economic Inequality and Political Power in America.
Princeton, NJ: Princeton University Press.
Gillespie, Charles G. 1985. “Uruguay’s Return to Democracy.” Bulletin of Latin American
Research 4(2): 99–107.
37
Giugni, Marco. 2004. “Personal and Biographical Consequences.” In The Blackwell Companion
to Social Movements, eds. David A. Snow, Sarah A. Soule, and Hanspeter Kriesi.
Malden, MA: Blackwell Publishing, 489–507.
Gleditsch, Nils Petter, Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg, and
Håvard Strand. 2001. "Armed Conflict 1946-2001: A New Dataset." Journal of Peace Research
39(5): 615-637.
Glewwe, Paul, and Gilette Hall. 1994. “Poverty, Inequality, and Living Standards during
Unorthodox Adjustment: The Case of Peru, 1985-1990.” Economic Development and
Cultural Change 42(4): 689–717.
Haggard, Stephan, and Robert R. Kaufman. 2016. Dictators and Democrats: Masses, Elites, and
Regime Change. Princeton, NJ: Princeton University Press.
Hartshorn, Ian M. 2019. Labor Politics in North Africa: After the Uprisings in Egypt and Tunisia.
Cambridge University Press.
Hasan, Zoya. 2011. Politics of Inclusion: Castes, Minorities, and Affirmative Action. Oxford
University Press.
Horowitz, Donald L. 1993. “Democracy in Divided Societies.” Journal of Democracy 4(4): 18–
38.
Htun, Mala. 2016. Inclusion without Representation in Latin America: Gender Quotas and Ethnic
Reservations. New York, NY: Cambridge University Press.
Huang, Reyko. 2016. The Wartime Origins of Democratization: Civil War, Rebel Governance, and
Political Regimes. New York, NY: Cambridge University Press.
Ives, Brandon. 2021. “Trickledown Politics: Do Excluded Ethnic Groups Benefit from
Non-Violent National Resistance Campaigns?” Conflict Management and Peace Science:
07388942211045936.
Jha, Kalpana. 2017. The Madhesi Upsurge and the Contested Idea of Nepal. Singapore: Springer.
Johnstad, Petter Grahl. 2010. “Nonviolent Democratization: A Sensitivity Analysis of How
Transition Mode and Violence Impact the Durability of Democracy.” Peace & Change
35(3): 464–82.
Lambach, Daniel et al. 2020. Nonviolent Resistance and Democratic Consolidation. Springer.
Lawoti, Mahendra. 2005. Towards A Democratic Nepal: Inclusive Political Institutions for a
Multicultural Society. SAGE Publications India.
Layton, Matthew L., and Amy Erica Smith. 2017. “Is It Race, Class, or Gender? The Sources
of Perceived Discrimination in Brazil.” Latin American Politics and Society 59(1): 52–73.
38
Lijphart, Arend. 1977. Democracy in Plural Societies: A Comparative Exploration. Yale University
Press.
López, Mónica Olaza. 2022. “Afro-Descendants and the Restoration of Democracy in
Uruguay: A New Vision of Citizenship?” Latin American and Caribbean Ethnic Studies
17(3): 281–95.
Madestam, Andreas, Daniel Shoag, Stan Veuger, and David Yanagizawa-Drott. 2013. “Do
Political Protests Matter? Evidence from the Tea Party Movement.” Quarterly Journal
of Economics 128(4): 1633–85.
Magni, Gabriele, and Andrew Reynolds. 2021. “Voter Preferences and the Political
Underrepresentationof Minority Groups: Lesbian, Gay, and TransgenderCandidates
in Advanced Democracies.” Journal of Politics 83(4): 1199–1215.
Manekin, Devorah, and Tamar Mitts. 2021. “Effective for Whom? Ethnic Identity and
Nonviolent Resistance.” American Political Science Review: 1–20.
Marks, Zoe, and Erica Chenoweth. 2019. “Empowerment or Backlash? How Women’s
Participation in Mass Uprisings Provides a Rising Tide.” Paper Presented at the
International Studies Association International Conference, Accra, Ghana.
Marshall, Monty G., Ted Robert Gurr, and Keith Jaggers. 2016. Polity IV Project: Political
Regime Characteristics and Transitions, 1800-2015. Center for Systemic Peace.
McAdam, Doug. 1989. “The Biographical Consequences of Activism.” American sociological
review: 744–60.
Morgan, Jana, and Nathan J. Kelly. 2021. “Inequality, Exclusion, and Tolerance for Political
Dissent in Latin America.” Comparative Political Studies 54(11): 2019–51.
Nepstad, Sharon Erickson. 2011. Nonviolent Revolutions: Civil Resistance in the Late 20th Century.
New York, NY: Oxford University Press.
Niazi, Golrokh. 2021. “Episodes of Contention, Bonds of Trust: The UGTT and ‘New’
Trade Union Activism in Post-Ben Ali Tunisia.” Globalizations 0(0): 1–15.
O’Donnell, Guillermo, and Philippe C. Schmitter. 1986. Transitions from Authoritarian Rule:
Tentative Conclusions about Uncertain Democracies. Baltimore, MD: JHU Press.
Paxton, Pamela Marie, Melanie M. Hughes, and Tiffany Barnes. 2020. Women, Politics, and
Power: A Global Perspective. Rowman & Littlefield Publishers.
Pinckney, Jonathan. 2020. From Dissent to Democracy: The Promise and Perils of Civil Resistance
Transitions. New York: Oxford University Press.
Pinckney, Jonathan, Charles Butcher, and Jessica Maves Braithwaite. 2022. “Organizations,
Resistance, and Democracy: How Civil Society Organizations Impact
Democratization.” International Studies Quarterly 66(1): 1–14.
39
Pinckney, Jonathan, and Miranda Rivers. 2021. Precarity and Power: Reflections on Women and
Youth in Nonviolent Action. Washington, DC: United States Institute of Peace. USIP
Peaceworks. https://www.usip.org/sites/default/files/2021-09/pw_178-
precarity_and_power_reflections_on_women_and_youth_in_nonviolent_action.pdf
(October 12, 2021).
Pischedda, Costantino. 2020. “Ethnic Conflict and the Limits of Nonviolent Resistance.”
Security Studies 29(2): 362–91.
Polletta, Francesca. 2002. Freedom Is an Endless Meeting: Democracy in American Social Movements.
Chicago, IL: The University of Chicago Press.
Reid, Robert H., and Eileen Guerrero. 1995. Corazon Aquino and the Brushfire Revolution. Baton
Rouge, LA: Louisiana State University Press.
Reynolds, Andrew. 2013. “Representation and Rights: The Impact of LGBT Legislators in
Comparative Perspective.” American Political Science Review 107(2): 259–74.
Rød, Espen Geelmuyden, Carl Henrik Knutsen, and Håvard Hegre. 2020. “The
Determinants of Democracy: A Sensitivity Analysis.” Public Choice 185(1): 87–111.
Rodriguez, Patricia, and David Carruthers. 2008. “Testing Democracy’s Promise: Indigenous
Mobilization and the Chilean State.” European Review of Latin American and Caribbean
Studies 85: 3–21.
Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John D. Stephens. 1992. Capitalist
Development and Democracy. New York, NY: Cambridge Polity.
Sharp, Gene. 1973. The Politics of Nonviolent Action. Boston, MA: Porter Sargent.
Sigman, Rachel, and Staffan I. Lindberg. 2019. “Democracy for All: Conceptualizing and
Measuring Egalitarian Democracy.” Political Science Research and Methods 7(3): 595–612.
Snyder, Jack. 2000. From Voting to Violence: Democratization and Nationalist Conflict. W.W.
Norton and Company.
Svensson, Isak, and Mathilda Lindgren. 2011. “Community and Consent: Unarmed
Insurrections in Non-Democracies.” European Journal of International Relations 17(1):
97–120.
Tarrow, Sidney. 1998. Power in Movement: Social Movements and Contentious Politics. Cambridge:
Cambridge University Press.
Teorell, Jan. 2010. Determinants of Democratization: Explaining Regime Change in the World, 1972–
2006. New York, NY: Cambridge University Press.
Thurber, Ches. 2018. “Ethnic Barriers to Civil Resistance.” Journal of Global Security Studies
3(3): 255–70.
40
Tripp, Aili Mari. 2015. Women and Power in Post-Conflict Africa. New York, NY: Cambridge
University Press.
de Vogel, Sasha. 2023. “Reneging and the Subversion of Protest-Driven Policy Change in
Autocracies.” Democratization 0(0): 1–25.
Walsh, Denise M. 2012. “Does the Quality of Democracy Matter for Women’s Rights? Just
Debate and Democratic Transition in Chile and South Africa.” Comparative Political
Studies 45(11): 1323–50.
Wängnerud, Lena. 2009. “Women in Parliaments: Descriptive and Substantive
Representation.” Annual Review of Political Science 12(1): 51–69.
Waylen, Georgina. 2007. “Women’s Mobilization and Gender Outcomes in Transitions to
Democracy: The Case of South Africa.” Comparative Political Studies 40(5): 521–46.
White, Peter B. et al. 2015. “Nonviolence as a Weapon of the Resourceful: From Claims to
Tactics in Mobilization.” Mobilization: An International Quarterly 20(4): 471–91.
Williams, Melissa S. 2000. Voice, Trust, and Memory: Marginalized Groups and the Failings of
Liberal Representation. Princeton University Press.
Wolbrecht, Christina, and Rodney E. Hero, eds. 2005. The Politics of Democratic Inclusion.
Philadelphia, PA: Temple University Press.
Wood, Elisabeth Jean. 2000. Forging Democracy from Below: Insurgent Transitions in South Africa
and El Salvador. New York, NY: Cambridge University Press.
Young, Iris Marion. 2002. Inclusion and Democracy. New York, NY: Oxford University Press.
