Ghosts of the Black Decade: How legacies of violence shaped Algeria’s Hirak protests
| Author | M Tahir Kilavuz,Sharan Grewal,Robert Kubinec |
| DOI | http://doi.org/10.1177/00223433221137613 |
| Published date | 01 January 2023 |
| Date | 01 January 2023 |
| Subject Matter | Regular Articles |
Regular Articles
Ghosts of the Black Decade: How legacies
of violence shaped Algeria’s Hirak protests
M Tahir Kilavuz
Department of Political Science and International Relations, Marmara University
Sharan Grewal
Department of Government, College of William & Mary
Robert Kubinec
Social Science Division, New York University Abu Dhabi
Abstract
Episodes of mass political violence, such as genocide and civil war, have been thought to both encourage and
discourage future political mobilization. We square these competing hypotheses by disaggregating between protest
onset and resilience. We argue that exposure to mass violence decades ago should on average decrease protest onset,
by heightening fears of repression and retribution. However, conditional on protesting, prior exposure to violence
should increase protest longevity, by generating greater political grievances that fuel commitment to the cause. We
find evidence of both effects in Algeria during the 2019–20 Hirak protests that toppled President Abdelaziz
Bouteflika. Pairing an original dataset on massacres during the 1990s civil war with a rolling online survey of
18,000 Algerians in 2019–20, we find that areas exposed to greater violence in the 1990s had on average fewer,
but more committed, protesters in 2019–20.
Keywords
Algeria, legacies of violence, protest, revolution, survey
Introduction
Episodes of mass political violence, such as genocide and
civil war, can have legacies that span decades (Bellows &
Miguel, 2009; Blattman, 2009; Finkel, 2015; Walden &
Zhukov, 2020). Citizens exposed to such violence often
develop distinct political attitudes, which in turn may be
passed down to their descendants (Balcells, 2012;
Lawrence, 2017; Lupu & Peisakhin, 2017). As a result,
violence from decades past may continue to shape polit-
ical developments well into the future.
However, scholars are divided over whether exposure
to such violence encourages or discourages future polit-
ical mobilization. On the one hand, widespread violence
may signal the regime’s resolve to repress protests in the
future (Zhukov & Talibova, 2018; Wang, 2021). By
heightening fears of repression, indiscriminate violence
may induce obedience and deter dissent (Young, 2019).
On the other hand, exposure to violence may also gen-
erate political grievances against the perpetrators of such
violence. In turn, these grievances may fuel moral out-
rage and thus mobilization (Lawrence, 2017; Rozenas,
Schutte & Zhukov, 2017).
Initial attempts to square these two competing
hypotheses emphasize the political opportunity struc-
ture. Where the regime remains strong, both willing and
able to repress, the ‘coercive effect’ should be dominant,
producing ‘silent dissidents’ (Wang, 2021): no mobiliza-
tion despite strong grievances. By contrast, when the
regime collapses or is defeated in war, the threat of
Corresponding author:
ssgrewal@wm.edu
Journal of Peace Research
2023, Vol. 60(1) 9–25
ªThe Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/00223433221137613
journals.sagepub.com/home/jpr
repression should fade, and the ‘grievance effect’ should
become dominant, producing mass dissent (Rozenas &
Zhukov, 2019).
In between, however, is a wide range of cases in which
the threat of repression might weaken but still remain
salient. In these middle cases, we argue, both the coercive
and grievance effects of prior violence should structure
the pattern of mobilization. In particular, both effects
should be evident in shaping two distinct protest out-
comes: onset and resilience. We hypothesize that prior
exposure to mass political violence should on average
decrease protest onset, by heightening fears of repression
and retribution. However, conditional on protesting,
prior exposure to violence should increase protest resili-
ence, by generating greater political grievances that fuel
commitment to the cause. In other words, areas exposed
to mass violence decades ago should on average see fewer
but more committed protesters.
We explore these hypotheses in the case of Algeria,
which was rocked by a brutal civil war in the 1990s
claiming at least 100,000 lives. The trauma of that ‘Black
Decade’ deterred future mobilization, with the 2011
Arab Spring uprisings largely bypassing Algeria (Pearl-
man, 2013; Khalil, 2014). However, a spark in February
2019 – President Abdelaziz Bouteflika’s pursuit of a fifth
term in office – finally broke the barrier of fear. Massive
and long-lasting peaceful protests known as the Hirak
toppled Bouteflika by April 2019 and then continued on
for nearly a year demanding deeper political reforms.
To measure the legacies of violence in Algeria, we
exploit spatial variation in exposure to violence during
the Black Decade, leveraging an original dataset on mas-
sacres in the 1990s. We then trace the legacies of these
killings onto protest participation in the 2019–20 Hirak
protests.Wefieldamassive,onlinesurveyofover
18,000 Algerians on a rolling basis between April 2019
and February 2020, allowing us to measure self-reported
protest participation, grievances, and expectations of
repression over time.
We find that areas exposed to more violence in the
1990s saw significantly lower rates of protest participa-
tion during the 2019–20 protests. In line with our first
hypothesis, this negative correlation between violence
and protest participation was mediated through heigh-
tened expectations of repression. Two decades later, areas
that had experienced more killings showed greater fear of
repression and therefore fewer protests.
However, we also find that the effe ct of massacres
flipped over time. Six months into the protests, the
regime began to shift gears, no longer granting conces-
sions and instead beginning a campaign of targeted
repression. Protest participation accordingly shrank. At
this stage, we find that the committed protesters who
continued on in the face of targeted repression were
significantly more likely to come from areas exposed to
violence in the 1990s. In line with our second hypoth-
esis, this positive correlation between violence and pro-
test participation in the later stages of the Hirak was
mediated through greater grievances towards the regime,
including by having personally lost a family member
during the war.
While the data are correlational, they suggest that
exposure to violence might have competing, long-term
legacies, on the one hand decreasing protest onset, but
conditional on protesting, increasing protest resilience.
The areas exposed to violence in Algeria in the 1990s saw
fewer, but more committed, protesters in 2019, who
continued on for months, stopping only in March
2020 with the outbreak of the Coronavirus pandemic.
This study thus provides both substantive and metho-
dological contributions to the literature on the legacies of
political violence. Theoretically, we show how the com-
peting legacies of violence fit together, differentially shap-
ing protest onset and longevity. Methodologically, we
leverage a unique, large-scale survey that allows us to tease
out the precise mechanisms through which these compet-
ing effects occur. Overall, the results suggest that a
regime’s use of indiscriminate repression and violencemay
be a double-edged sword, fueling intergenerational grie-
vances that come back to haunt the regime decades later.
Political violence and its legacies
Episodes of mass political violence, whether civil war,
genocide, or indiscriminate repression, often produce
traumatic long-term legacies that shape political attitudes
and behavior for decades. One strain of thought empha-
sizes that political violence induces fear of future vio-
lence, deterring mobilization. Even generations later,
when the leader that perpetrated the violence is no longer
in power, these communities may still carry heightened
fear of repression. Zhukov & Talibova (2018: 268), for
instance, find that ‘communities more heavily repressed
under Stalin are less likely to vote today’ under Putin,
due to Stalin’s terror having ‘raised the expected costs of
even seemingly benign political participation’. Wang
(2021: 469) likewise finds that areas more exposed to
terror under China’s Cultural Revolution are less likely
to protest today, ‘fearing that the same tragedy might
happen again’.
On the other hand, other scholars contend that polit-
ical violence instead encourages future mobilization, in
10 journal of PEACE RESEARCH 60(1)
To continue reading
Request your trial