Swords into ploughshares? Why human rights abuses persist after resistance campaigns
| Author | Christopher Wiley Shay |
| DOI | http://doi.org/10.1177/00223433221140432 |
| Published date | 01 January 2023 |
| Date | 01 January 2023 |
| Subject Matter | Regular Articles |
Swords into ploughshares? Why human rights
abuses persist after resistance campaigns
Christopher Wiley Shay
Human Rights Institute, University of Connecticut & Belfer Center for Science and International Affairs,
Harvard Kennedy School
Abstract
Human rights abuse tends to increase during national crises, such as civil wars and mass nonviolent uprisings. Under
what conditions does this abuse abate or persist? I argue that violent challenges provoke much more coercive state
responses, exposing more personnel within the security forces to extreme forms of repression and priming them (both
leaders and followers) to reproduce these behaviors after the conflict has terminated. This effect is mitigated or
avoided when challengers rely on nonviolent tactics instead of violence, leading to less post-conflict abuse. I test this
argument with several quantitative methods, which establish a positive association between nonviolent resistance
campaigns and subsequent freedom from political killings. Results also suggest that democratization can partially –
but not fully – counteract the repressive legacies of violent conflict. By choosing to specialize in nonviolent tactics,
therefore, resistance leaders avoid a repression trap that not even democratization can fully disarm.
Keywords
democracy, human rights, insurgencies, nonviolent resistance, state repression
Introduction
Existing research shows that regimes respond to internal
threats – such as violent insurgencies and populist
uprisings – by increasing repression, leading to more
human rights abuse and greater suffering (Davenport,
2007a; Conrad & Moore, 2010; Hill & Jones, 2014).
While scholars have explored why repression varies
across conflicts (DeMeritt, 2015; Cohen, 2016; Stanton,
2016; Green, 2018), the long-term effects of conflict on
state repression remain unexplored. In this article,
I argue that variation in post-conflict repression levels
is rooted in whether resistance groups choose to specia-
lize in violent or nonviolent methods.
My argument builds on existing work showing, first,
that violent conflicts are transformative for the security
forces
1
who participate in them, altering both leaders’
and followers’ ideas about the utility and appropriateness
of violence and coercion (Gurr, 1988; Valentino, 2013;
Cesur & Sabia, 2016; Green, 2018). Second, dissidents
do not always resort to violent methods to challenge the
state. By specializing in nonviolent tactics, challengers
can make violent repression costly for state leaders,
disincentivizing the bloodier forms of repression
(Chenoweth & Stephan, 2011; Perkoski & Chenoweth,
2018). From these starting points, I argue that a challen-
gers’ choice of resistance method – violent vs. nonviolent
tactics – can create legacy effects which influence security
forces’ proclivities for violence. These proclivities are
inculcated when, while suppressing violent resistance,
leaders and security forces add lethal tactics to their
repertoires of coercion, then learn to see these tactics as
effective, legitimate, and normal. Consequently, state
repression and human rights abuse will be more prevalent
in the future, after the conflict has ended. By contrast, a
country is unlikely to fall into this ‘repression trap’ when
resistance movements specialize in nonviolent methods.
Corresponding author:
christopher.wiley.shay@gmail.com
1
This logic also applies to rebel militants who sometimes – though
rarely – go on to replace the state’s security forces.
Journal of Peace Research
2023, Vol. 60(1) 141–156
ªThe Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/00223433221140432
journals.sagepub.com/home/jpr
If insurgencies teach security forces to rely on vio-
lence, they are not always free to act on this proclivity.
Primarily, this occurs when violent movements result in
democratization (Toft, 2010; Huang, 2016). However
established, electoral institutions incentivize leaders to
restrain their security forces from perpetrating abuse
(Davenport & Armstrong, 2004; Davenport, 2007b).
Democratization, in other words, likely nullifies the dif-
ference in post-campaign repression outcomes between
violent and nonviolent campaigns. But, because demo-
cratization is empirically rare, resistance methods play an
important role across most post-campaign contexts.
My argument implies that violent resistance move-
ments should lead to worse post-conflict human rights
outcomes than nonviolent movements, except when
resistance leads to democratization. To test this impli-
cation, I draw primarily fromtwodatasources. First,
the Nonviolent and Violent Campaigns and Outcomes
(NAVCO) 2.1 dataset (Chenoweth & Shay, forthcom-
ing) allows me to track all cases of mature and maxim-
alist conflict – campaigns – between 1945 and 2013,
and to sort them by resistance method.
2
Second, the
Varieties of Democracy (V-Dem) project (Coppedge
et al., 2021) allows me to track both the prevalence
of political killings perpetrated by state agents in the
aftermath of NAVCO campaigns and whether demo-
cratization follows.
I use several statistical techniques to test my argu-
ment. Results from linear hierarchical models show that
nonviolent campaigns are associated with greater respect
for freedom from political killings in the short term – five
years after campaign termination. Survival models pro-
vide consistent results, showing that nonviolence reduces
the time it takes to significantly improve respect for
human rights in the first decade after campaign termina-
tion. Surprisingly, these results hold even when control-
ling for democratization. Finally, results from mediation
models suggest that resistance method partially affects
future state repression through its intermediate effect
on democratization. On balance, these results support
the argument that violent resistance methods teach lead-
ers and security forces to be more repressive in the future.
At the same time, they also suggest that electoral incen-
tives do not fully counteract the repressive legacies of
violent insurgencies. Potentially, this is because elected
leaders never have complete control over their followers
in the security services (Svolik, 2012; DeMeritt, 2015;
Hendrix & Salehyan, 2017; Green, 2018), who will
retain affinities for violence regardless of political change.
Though an imperfect solution, democratization does
clearly lead to fewer post-conf lict political killings. In
fact, the results from mediation analyses suggest this is
one pathway by which nonviolent methods reduce post-
conflict repression – nonviolence leads to democratiza-
tion (consistent with Chenoweth & Stephan (2011) and
Bayer, Bethke & Lambach (2016)), which thereby con-
strains repression. However, most of the effect of non-
violence is independent of democratization. These
findings yield practical insights. For policymakers look-
ing to promote democratization abroad, this goal can be
advanced by supporting nonviolent resistance move-
ments.
3
Even when these investments fail to achieve
democratization, they are likely to protect affected popu-
lations from state repression.
Background and theory
The legacies of violent vs. nonviolent resistance
Everyday repression entails surveillance, harassment, and
myriad other forms of social control (Earl, 2011). To the
extent severe repression like killings and disappearances
takes place, the regime can rely on a relatively small pool
of specialists.
4
However, should an opposition group
mobilize and mount a serious challenge to the regime
– a campaign – security forces will be called on to sup-
press the challenger, as predicted by the Law of Coercive
Response (Davenport, 2007a). The severity and form of
the response will depend largely on the challenger’s
choice of resistance method.
For their part, challengers typically choose to
specialize in either violent or nonviolent resistance.
When challengers choose nonviolent tactics, regimes
rarely respond with extreme repression. This is not
because regimes lack the capacity to do so. Nor do lead-
ers necessarily have qualms about killing unarmed pro-
testers. Rather, leaders fear a potential backfire effect, or
the resulting loss of support (and even active retaliation)
from domestic and international audiences (Schock,
2004; Chenoweth & Stephan, 2011; Svensson &
Lindgren, 2011). Potentially even more costly, the
regime’s security forces may refuse to carry out orders
2
The challengers who launch these campaigns are all domestic actors,
meaning that cases of interstate conflict are excluded.
3
Scholars have recently begun to explore the ways in which foreign
support benefits (and potentially undermines) nonviolent movements
(Chenoweth & Stephan, 2021).
4
Per the authoritarian dilemma logic (Svolik, 2012; Greitens, 2016),
the smaller the pool of specialists, the less danger they pose to the
regime itself.
142 journal of PEACE RESEARCH 60(1)
To continue reading
Request your trial