Content uploaded by Madhav Joshi
Author content
All content in this area was uploaded by Madhav Joshi on Dec 27, 2024
Content may be subject to copyright.
Political Geography 109 (2024) 103044
Available online 23 December 2023
0962-6298/© 2023 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Full Length Article
Civil war violence and competing legitimacy claims: Evidence from district
level courts cases in Nepal
Madhav Joshi
Peace Accords Matrix (PAM), Kroc Institute for Int’l Peace Studies, Keough School of Global Affairs, University of Notre Dame, USA
ARTICLE INFO
Keywords:
Civil war violence
Contested territories
State legitimacy
Civilian compliance
Nepal
ABSTRACT
Internal armed conict is the ultimate instantiation of two actors – the state and a non-state group – challenging
the legitimacy of one another. The state primarily responds violently to those who challenge, either violently or
non-violently, its legitimacy. To what extent does the violent suppression of rebel activity erode state legitimacy?
How does the potential for the use of violence by non-state armed groups and the offer of parallel governance
erode state legitimacy? This research develops an argument specic to contested territories where both sides
cannot offer model governance but use violence for opposing reasons, namely the rebel group to delegitimize the
state and the state to regain its monopoly on the legitimate use of force. Because the use of violence from either
side creates a demand for judicial remedies, I conceptualize the degree of legitimacy in terms of the utilization of
the state court system in the face of increased violence by the state or non-state armed groups. In line with
theoretical expectations, empirical analyses of criminal and civil cases led in Nepal’s district courts during the
Maoist insurgency suggest a statistically signicant negative impact of the rebel use of violence and their offer of
alternative governance on state legitimacy. Similarly, the state’s use of violence had a negative and signicantly
detrimental effect on state legitimacy.
1. Introduction
Civil war or rebellion is a condition of “multiple sovereignty,” in
which some people in a country regard multiple “power blocks” as
legitimate (Tilly, 2018, 1977, p. 191). It is a situation in which both state
and non-state actors or rebel groups use violence to substantiate their
claim of political authority and coerce civilian support. Extant literature
on the legitimacy of political authority points to the role of political
obligations (Buchanan, 2002; Linz, 2019), suggesting the signicance of
good governance for civilian acceptance of a political authority and their
claim of legitimacy (Gilley, 2006b; Rothstein, 2011). Therefore, rebel
groups that aspire to control the state act like the state by establishing
alternate institutions and offering governance during rebellion (Albert,
2022; Schlichte & Schneckener, 2015; Arjona, 2016; Huang, 2016b;
Stewart, 2018).
Rebel governance has recently gained signicant scholarly attention
for its relevance in understanding civil war processes and post-war po-
litical development (Cunningham & Loyle, 2021; Loyle, Cunningham,
Huang, & Jung, 2021, 2022a). As a new and emerging research area,
researchers focus on the institutions built and the governance offered by
rebel groups, with a particular focus on the way they organize their
governance and offer various services to civilians to gain their support
(Albert, 2022; Arjona, 2016; Huang, 2016b; Mampilly, 2011; Stewart,
2021a). For rebel groups’ ability to govern and generate civilian loyalty,
territorial control is argued to be a fundamental prerequisite (Arjona,
2016; Kasr, 2015; Stewart, 2021b, p. 72; Mampilly, 2011, p. 63;
Worrall, 2017; Loyle et al., 2022a, pp. 16–20; Huang, 2016b, p. 60).
Existing research on rebel governance suggests contested territories,
which are the zones of indiscriminate violence against civilians (Hult-
man, 2007; Kalyvas, 2006; Metelits, 2009), are not suitable for rebel
governance. Further, the use of violence is recognized not to have a
persistent and long-term effect on civilian compliance (Kalyvas, 2006;
Mampilly, 2011; Podder, 2017; Waterman, 2023; Worrall, 2017). This
raises the question: what is the objective of the state and rebel groups’
use of violence, and how does violence help achieve or degrade legiti-
macy in contested territories?
In this research, I conceptualize legitimacy in terms of civilian
acceptance of political authority and theorize the distinctive purpose of
the state and rebel use of violence in contested territories in relation to
legitimacy. The most positive way to frame the state’s use of violence is
that it is employed to protect civilians from rebel violence and to
maintain and expand its acceptance among civilians. For the rebels,
E-mail address: mjoshi2@nd.edu.
Contents lists available at ScienceDirect
Political Geography
journal homepage: www.elsevier.com/locate/polgeo
https://doi.org/10.1016/j.polgeo.2023.103044
Received 24 May 2023; Received in revised form 5 November 2023; Accepted 8 December 2023
Political Geography 109 (2024) 103044
2
violence can either be used to protect civilians from state violence or, on
the other hand, to target civilians in order to weaken state institutions
and provide alternative governance to prevent the state from gaining
civilian compliance. Therefore, I expect the state and rebel use of
violence and rebel alternative governance to affect civilian compliance
and legitimacy.
I test the implications of the theoretical argument specic to the
impact of state and rebel violence and rebel alternative governance on
civilian compliance by utilizing both district-level annual data on court
cases and district-level data on the establishment of the Maoist gov-
ernment during the Maoist civil war in Nepal. The Maoists established
the people’s government and people’s court and governed civilians by
adjudicating cases, offering administrative services, and providing se-
curity (Onesto, 2005; Pathak, 2005; Yami, 2021). The underlying
assumption for the empirical strategy is that civilians would bring cases
to the state court system for judicial remedies only if the state is
considered legitimate. While conict and rebel governance can create
different social, economic, and political order in an active conict zone,
which would deter civilians from ling claims against both the rebel
group or the state, civilians would still be able to bring other cases
outside of those directly associated with civil war related damages.
Further, it is understood as the responsibility of the legitimate state to
establish law and order by prosecuting those who would oppose the
state, harm civilians, or destroy civilian property. The data on criminal
and civil cases during the Maoist insurgency (1996–2006) suggest that
the court system at the district level continued to function, which sug-
gests a degree of civilian support and legitimacy for the state. Never-
theless, the empirical analysis suggests a signicant decline in the
number of court cases in response to an increase in violence by both the
state and insurgents, as well as in response to the establishment of a
district-level rebel authority (commonly referred to as the Maoist gov-
ernment). Thus, while the state court continued to operate, it was being
utilized at a lesser degree than before the conict. While the use of
violence by rebels could preclude the state from bringing criminal cases,
state violence, especially against civilians, would erode state legitimacy,
which makes citizens less inclined to use the state’s court system. The
overall conclusion is that the use of violence is counterproductive for
state legitimacy.
This article primarily contributes to the literature at the intersection
of civil war violence and rebel governance (Albert, 2022; Cunningham &
Loyle, 2021; Kasr, 2015; Loyle et al., 2022a; Mampilly, 2011; Stewart,
2021a). By theorizing how civil war violence in contested territories can
be linked to the state and the rebel group’s efforts to gain civilian
compliance and legitimacy, this paper unpacks the complexities
involving the issue of governance in contested territories. In so doing,
this paper contributes to and advances the broader discussion on civil
war violence and governance in contested territories. By focusing on
contested territories and courts at the subnational level, this research
places emphasis on the signicance of local institutions and geography
in the study of war and peace (Megoran, 2011; Wig & Tollefsen, 2016).
Equally important, this research advances rebel governance research at
the subnational level. Furthermore, this research suggests the signi-
cance of court cases in civil war violence and state legitimacy research.
The article is divided into four sections. The rst section discusses the
puzzle of legitimacy in both a general sense and specically in relation
to contested territories, explains the state and rebel use of violence for
opposing objectives, and ties the use of violence to demands for judicial
remedies. The argument is then used to derive empirically testable hy-
potheses. The second section, or the research design section, discusses
the case of the Maoist civil war in Nepal and presents the data used in
this research. The third section offers the results and analysis from the
empirical analyses. The nal section summarizes the research, presents
possible implications for policy, and offers new avenues for future
research.
2. The concept of legitimacy in civil war
Civil war can be understood as a phenomenon in which the state and
non-state armed actors violently challenge one another to maintain or
gain legitimacy for governance over territory and civilians. In political
science and sociology, legitimacy is understood and dened in terms of a
shared belief that a rule, institution, or individual has the right to govern
(Gilley, 2006a; Weber, 1946). Therefore, governance, in terms of the
provision of political goods and services that the political authority
provides, has a direct effect on legitimacy (Gilley, 2006a; Mcloughlin,
2015). Civil wars or armed conicts are explained as having either
territorial, governmental, or both types of incompatibilities (Wallens-
teen & Axell, 1993). Therefore, it is logical to understand rebel groups as
having the objective of taking over the state and changing a regime,
reforming it by providing better governance, or seceding from it (Arjona,
2016; Huang, 2016b; Kasr, 2015; Mampilly, 2011; Rubin, 2020;
Stewart, 2021b). For groups seeking legitimacy, they adopt governance
structures and provide law and order, security, and goods and services
that enhance their legitimacy among civilians (Huang, 2016a; Led-
widge, 2017; Loyle & Binningsbø, 2018; Loyle et al., 2021). Yet, this
exchange of governance for civilian acceptance can be challenging for
rebel groups in the initial conict phase, as rebels lack evidence of their
governance or have yet to completely deliver their governance.
The extant research points to coercion, or the use of violence, as
necessary for generating civilian compliance towards the rebel author-
ity. According to Kalyvas (2006, p. 114), the use of violence “operates as
a rst order condition that makes the production of loyalty possible.” In
the long run, however, overreliance on the use of violence to maintain
civilian compliance may not work (Kalyvas, 2006, pp. 114–17; Wick-
ham-Crowley 1987; Podder, 2017; Worrall, 2017). Therefore, studies on
rebel governance emphasize the need for rebels to establish and main-
tain their relations with civilians, as well as establish institutions and
offer governance (Arjona, 2017; Cunningham & Loyle, 2021; Huang,
2016b; Loyle et al., 2022a; Wickham-Crowley 1987).
Similar to the state, rebel groups need territorial control and resi-
dential population for their governance. The degree of territorial control
is instrumental to understanding the degree of civilian compliance and
the use of violence (Kalyvas, 2006, pp. 114–17; Wickham-Crowley
1990). Therefore, the rebel governance literature emphasizes the sig-
nicance of rebel territorial control on their ability to govern and
generate civilian loyalty (Arjona, 2016; Kasr, 2015; Mampilly, 2011;
Stewart, 2021b; Worrall, 2017).
Territorial control is foundational for a political authority to emerge
and to ensure its legitimacy (Weber, 1946). According to Sack, the ter-
ritory is “a means of making authority visible” (1986, 38). The role of
territory in civil war literature is discussed in reference to the role of
state capacity, with the argument that a weak state that is not able to
govern its territory or is simply not present within its peripheries creates
conditions for civil wars to begin (Braithwaite, 2010; Collier & Hoefer,
2002, 2004; Gurses & Mason, 2010; Hendrix, 2010; Mason, & Michael
Greig, 2017; Thies, 2010).
1
The cause of armed conict has been a topic
of debate, with some attributing it to the absence or weakness of the
state. However, recent research indicates that the state’s efforts to
govern previously ungoverned areas could also be a contributing factor
(Dowd & Raleigh, 2013; Ying, 2021). Nevertheless, the main topic of
this discussion pertains to territorial control and authority, with
differing viewpoints regarding the ability of the state to govern.
In civil wars, while the state primarily defends its territory or ghts
to reclaim territorial control, rebels aim to gain territory or hold existing
territorial gain with the objective of establishing political authority
(Kalyvas, 2006; Kasr, 2015; Loyle et al., 2022b; Mampilly, 2011;
1
Except for urban insurgency (i.e. the Irish Republican Army (IRA) in
Northern Ireland and the Movement 19 (M-19) in Colombia, most rebel groups
organize and initiate the conict in rural areas that are mostly ungoverned.
M. Joshi
Political Geography 109 (2024) 103044
3
Stewart, 2018). A civil war state often has areas under state control,
rebel control, and areas contested by the state and rebel groups, which
translates into the state or rebels’ claim of civilian acceptance and,
therefore, legitimacy. Both the rebels and the state claim their legiti-
macy in the areas they respectively control, with civilians, in most in-
stances, complying with the political order voluntarily and collaborating
with that authority to enforce rules and orders. Civilians might still use
state or rebel institutions despite rejecting their legitimacy in order to
navigate the condition of active armed conict or a perceived threat of
violence. The wartime social, political, and economic cleavages that
were salient during the war might distort civilian preference (Balcells,
2012). In other cases, warring parties might forcefully displace civilians
from the areas of their control depending on civil society organizations
mediating between the civilians and armed actors (Steele, 2009; Kaplan,
2013; Lichtenheld & Schon, 2021). The cost of governance for the rebels
could also deter them from controlling the territory, which often is the
case in the presence of state service provisions and civilian collective
action (Rubin, 2020). Nevertheless, it is in these respectively controlled
areas where the state or rebels can claim to run their model governance.
2
While civilians may be concerned with who will end up controlling
territory after the active contestation declines, the establishment of rebel
governance in contested territories where civilians do not necessarily
comply, however, would be challenging to sustain (Arjona, 2016; Kasr,
2015; Mampilly, 2011; Stewart, 2021b; Worrall, 2017). Therefore, our
current understanding of rebel and state legitimacy in contested areas is
limited.
To conclude, the issue of legitimacy in contested areas in civil war is
consequential for three reasons. First, similar to controlled zones or
areas, in contested territories, both armed actors primarily focus on
civilian support and compliance with their authority. Second, contested
areas in civil war are the ones where civilians see both the state and the
rebel groups’ reciprocal violent actions. And third, both the rebels and
the state use discriminate and indiscriminate violence, as well as coer-
cion or repression, to generate compliance in contested territories (Joshi
& Quinn, 2017; Kalyvas, 2006; Lichbach, 1998; Mason & Krane, 1989).
3. A theory of contested legitimacy during civil war
The theoretical argument developed in this paper suggests that the
purpose of the use of violence for the state and rebels are different in
contested territories and have signicant implications for civilian
compliance. There are at least three primary purposes for the state’s use
of violence in a contested territory. First, contested territory is a space
where the state was formerly absent and is now attempting to expand its
authority and governance. Second, the state aims to protect civilians
from rebel violence and their ideological inuences. Third, these efforts
to expand authority and governance and protect civilians from rebel
violence can help the state gain civilian compliance. There are also at
least three primary objectives for the use of violence by the rebel group,
which serve different purposes than those of the state. The rst objective
of the rebels is to convince civilians that the state is weak, which man-
ifests itself through specic targeting of individuals who work closely
with the state and the security infrastructures responsible for main-
taining order. The second objective is to offer an alternative governance
structure, which at times is met with civilian resistance, compelling
rebels to use violence to establish order. And third, both the civilians and
the state need to see the functioning, however symbolic, of the rebel
offer of alternative governance. The use of violence is the most effective
alternative and psychological technique available for the rebels in the
initial phase of dominating territory (Kalyvas, 2006; Lichbach, 1998;
Worrall, 2017).
3
As such, the use of violence by the state and rebels in
contested territories can divide civilian compliance between the rebel
and state authorities.
This research is focused on changes in legitimacy, which is the
civilian acceptance and use of state institutions. The legitimacy of state
institutions is based on the perception of civilians who live within a
specic territory or administrative unit. Therefore, civilian mass support
is absolutely necessary for establishing legitimacy. When territorial
control is secured, the institutions and governance offered by the po-
litical authority become non-competitive or zero-sum, meaning multiple
“power blocks” no longer exist (Tilly, 1977). Yet territorial contestation
and rival governance in civil war suggests legitimacy to be non-zero sum
(Podder, 2017). This paper understands legitimacy in contested terri-
tories as non-binary, as Podder argues (2017), and as similar to what
Huang (2016b, p. 136) suggests as “some degree of popular support.” It
is also reasonable to expect civilians to prefer rebel authority and
governance if existing state institutions are deemed illegitimate, corrupt,
or inadequate in meeting their needs. Therefore, the theoretical argu-
ment focuses on the extraordinary nature of the loss and damage civil-
ians incur when the state and rebels use violence for territorial
contestation and whether civilians trust and utilize the existing state
institutions to demand justice for those losses.
Civil war is destructive in many ways. There is considerable physical
violence, with civilians displaced, targeted, and killed in territories
where the rebel and state are ghting (Adhikari, 2013; Czaika &
Kis-Katos, 2009; Hultman, 2007; Kalyvas, 2006; Metelits, 2009). There
is also considerable structural violence, with high loss of economic op-
portunity and infrastructure, which indirectly harms civilians
throughout the territory (Collier & Hoefer, 1998; del Castillo, 2008). In
addition, civil war violence often also relates to local and private con-
icts or “blood revenge”, which may not relate to the main cause of the
war (Kalyvas, 2003; Souleimanov, Siroky, & Colombo, 2022). For
example, the formation of local alliances, allegiance, and violence on the
ground often relate to family feuds, individual vendettas, and rivalry.
This is evident in cases like Sri Lanka’s Tamil Civil War, Colombia’s La
Violencia, the Spanish Civil War, the French Revolution, the Huk
Rebellion, and even the Afghan War (See Kalyvas, 2003). While civil war
violence is considered political violence, individuals often score their
personal differences under this pretext of political violence.
As has become evident, contested territories in civil wars produce
high levels of violence involving the state, rebels, and civilians. And
while civil war violence and atrocities are often put aside to be dealt
with under post-conict transitional justice mechanisms due to their
political nature (Sooka, 2006), it is reasonable to expect that most of the
harms that are not political are criminal and civil in nature. Therefore,
civil war violence in contested territories is expected to generate higher
demands for judicial remedies, which should lead to higher utilization of
the state court system. One could argue that the fear of further prose-
cutions would deter civilians and rebel supporters from bringing cases to
the court system. Additionally, both the state and rebel often fail to
accept culpability for damages incurred by civilians, as happened in Sri
Lanka and Colombia where civilians were targeted (Colombia Reports,
2019; Cronin-Furman, 2020), which could also deter the ling of cases
during the conict period. Nevertheless, if the state is deemed legitimate
and civilians trust the court to organize a free trial, the utilization of the
2
According to Breslawski (2021), rebel governance falls between the exclu-
sive and inclusive institutional continuum by the rebels. Rebel-held territory
can have different institutions depending on whether the rebel constituency and
community are cohesive or non-cohesive. Rebels can still use coercion in those
held territories that are inhibited by civilians who are not a part of the rebel
constituency.
3
Rebels often work with and through state institutions in areas of their
control (Mampilly, 2011; Waterman, 2023). Waterman (2022) offers insights
from India’s United Liberation Front of Assam (ULFA) that the insurgents did
not have territorial control but had a parallel government in place because of
their penetration into the state machinery, their preexisting social ties, and the
local media that spread fear by drawing public attention to the violent acts of
ULFA. The ULFA’s use of violence was the key factor in this process.
M. Joshi
Political Geography 109 (2024) 103044
4
court should increase.
Unlike prior research that seeks to explain rebel governance in terms
of rebels providing schools, health services, security, and a judiciary in
conict zones (Huang, 2016a; Loyle, 2021; Mampilly, 2011; Stewart,
2021b), this research focuses on the state court system for several rea-
sons. First, the loss and damage incurred by civilians and the harm done
to them can only be redressed through judicial remedies. If state courts
are deemed fair and acceptable, civilians will utilize them, which can be
regarded as a signicant boost in state legitimacy. Second, and to the
previous point, it is therefore in the interest of rebel groups to discourage
civilians from utilizing the court system for the purpose of limiting the
state from gaining legitimacy in contested territories. Third, although it
may be inefcient and corrupt, the state court is uniformly available in
all administrative units which can be utilized for the supremacy of the
law, whereas the extant rebel governance research does not suggest
uniformity and consistency in rebel provision of public goods and ser-
vices. As such, the court is more than an institutional mechanism for
civilians seeking judicial remedies for their damage and loss claims; it is
a reection of civilian acceptance of a legitimate government. There-
fore, the utilization of the state court serves as a good indicator of
legitimacy for the state and a lack of attribution of legitimacy to the
rebels.
Both the state and civilians utilize the court system. The state utilizes
the court system to charge suspected supporters of rebels in criminal
cases, arguably to maintain the rule of law. During a civil war, civilians
may view such cases as an alternative form of state repression and state
violence. When the state turns to overt violence, criminal cases may
decrease. The state’s use of such violence erodes government legitimacy.
On the other hand, civil cases are initiated by citizens. As such,
increasing use of courts for civil matters is an indicator of increasing
state legitimacy in the eyes of the population: they trust state institutions
enough to initiate legal actions in them on civil matters.
To summarize the theoretical argument, the primary objective of the
use of rebel and state violence in contested territories in civil wars re-
lates to establishing legitimacy or gaining civilian compliance with one’s
own rules and authority. Civilians suffer from violence perpetrated by
the state, the rebels, and fellow civilians, for which they may seek
judicial remedies through the available state court system. State
perpetrated violence against civilians is done in an effort to convince
civilians that the state is in control, and the level of this violence in-
dicates their ability to maintain authority and ensure civilians feel safe
to bring cases in the court system the state provides. But, violence
against civilians is counterproductive to state legitimacy. On the other
hand, rebels challenge the legitimacy of the state, and the level of their
use of violence demonstrates the state’s weakness, thereby dissuading
civilians from seeking justice through the state court system. Because the
objective of the use of violence is different for the state and the rebels,
the use of violence affects the civilian utilization of the court system
differently. Based on the theoretical discussion, I formulate two hy-
potheses for the state’s effort to expand its legitimacy amidst rebel
challenge to its authority and the rebel violence attempting to constrain
the state.
H1. State violence in civil war decreases the civilian utilization of the
state court system.
H2. Rebel violence in civil war decreases the utilization of the state
court system.
One of the measures of rebel capacity is their ability to establish a
legal system (Mampilly, 2011, p. 44; Huang, 2016b). However, in con-
tested territories, the scope of the rebel court can be limited and avail-
able only symbolically. Nevertheless, Mampilly (2015) suggests the
signicance of symbolic processes, as they may increase civilian iden-
tication with the rebel government and ensure compliance. For
example, the Maoist rebel group in Nepal instituted a people’s court as a
component of the people’s government. While the rebel court was not
universally accessible, individuals were afforded the opportunity to
bring cases to this alternate system for adjudication. Further, the pre-
sence/absence of a people’s court in a district would indicate some de-
gree of territorial control by the rebels, which should result in a decline
in both types of cases in the government’s court system. Therefore, I
hypothesize about the alternative court system offered by the rebels to
depress the civilian utilization of the state court system.
H2b. Rebel establishment of the court leads to a decrease in the uti-
lization of the state court system.
4. Research design
4.1. The Maoist conict in Nepal, territorial contestation, and legitimacy
The case of the Maoist conict in Nepal is ideal for testing the
theoretical argument and hypothesis for several reasons. First, Nepal is a
case where the Maoist conict broke out in 1996 when the state was still
consolidating the transition to multiparty democracy. Nepal began this
transition in 1990, when the state was not present throughout the
country and most of the governance issues were handled by landed elites
with close ties to the ruling class (Joshi & Mason, 2011; Whelpton,
2005). The democratic transition of 1990 kindled high civilian expec-
tations from the state, but the state performance was not up to the
popular expectations (Joshi & Mason, 2011; UNDP Nepal, 1998;
Whelpton, 2005). Second, while the state was weak and not delivering to
the level of people’s expectations, Nepal was still making steady prog-
ress even during the war years. According to UNDP reports, between
1996 and 2000, life expectancy increased from 55 years to 59.5 years,
adult literacy rate from 36.72 to 50.7 percent, and mean years of
schooling from 2.25 to 2.89 (UNDP Nepal, 1998, 2002). Third, Nepal is a
case where the Maoist rebels questioned the legitimacy of the state for its
negligence of remote districts, women, and marginalized communities,
and in response, took up arms to establish the people’s government
(United People’s Front Nepal, 1996). At the village level, where the
Maoists were able to maintain their hold, they established the people’s
government and people’s court, governing civilians by offering admin-
istrative services and providing security (Onesto, 2005; Pathak, 2005;
Yami, 2021). Yet while the Maoist rebels had a more robust presence in
villages in some western districts, they never maintained absolute ter-
ritorial control or dispelled the state at the district level.
4
Fourth, and
nally, Nepal is a case in which both the state and the Maoist rebels used
discriminate and indiscriminate violence against civilians (Joshi &
Pyakurel, 2015).
4.2. The data
To test the relationship between civil war violence and state legiti-
macy, I construct a district level annualized dataset that includes the
levels of state and rebel violence, the Maoist establishment of parallel
government, and district level court cases for all 75 districts between
1996 and 2006. This covers the period of the Maoist insurgency in
Nepal, which started in February 1996 and ended in November 2006,
after the Nepal government and the Maoist group signed a compre-
hensive peace agreement.
The unit of analysis is a district and the dependent variable is the
utilization of the court system, which is measured using two variables:
civil cases and criminal cases. In criminal cases, the state is directly
4
According to Pathak (2005), the Maoist rebels were more effective in rural
villages and not in district headquarters or cities. In my informal conversations
with political party leaders, including ex-Maoist rebel leaders, it was suggested
that the state would govern during the day while allowing the Maoists to govern
at night.
M. Joshi
Political Geography 109 (2024) 103044
5
involved as the one responsible for prosecuting the defendant.
5
In other
words, this variable captures the extent to which the state is committed
to maintaining the rule of law by prosecuting those that the state nds
opposing its legitimacy. In civil cases, the state is not directly involved
but provides institutional space through the court system, so that civil-
ians can seek judicial remedies for any loss or damage based on existing
laws and legal precedence.
6
The civil and criminal cases are combined to
form a variable, all court cases, which is also used as a dependent vari-
able. Information about the civil and criminal cases for each district
comes from the annual report from the Supreme Court of Nepal (Su-
preme Court of Nepal, 2006).
7
The descriptive statistics and correlation
matrix can be found in Appendix Tables 1 and 2, respectively.
Fig. 1 provides an annualized view of the number of criminal and
civil cases led across all district courts in Nepal. As can be seen in the
gure, the state had a degree of civilian compliance during the in-
surgency years, as the state court system was still used by civilians.
Nevertheless, both the criminal and civil cases declined considerably
during the period of the Maoist insurgency. The number of civil cases
increased from 2001 to 2002, which corresponds to the period when the
constitutional monarch assumed executive power after declaring a state
of emergency. The increase in the number of civil cases might indicate
civilian condence in state institutions, though the criminal cases did
not increase in the same time period.
The main independent variable is civil war violence. H1 suggests a
negative relationship between state violence and the utilization of the
state court system. To test this hypothesis, I use the variable state
violence, which is a count variable providing an annualized cumulative
sum of the number of people killed by the state in a given district. H2
suggests a negative relationship between rebel violence and the utili-
zation of the state court system. To test this expectation, I use the var-
iable Maoist violence. This variable is constructed similarly to the state
violence variables with annualized sums of the number of people killed
by the Maoist rebels in a given district. Both variables come from
individual victims’ data from Nepal (Joshi & Pyakurel, 2015) and are log
normalized. The source of this data is the Informal Sector Service Centre
(INSEC), a human rights organization that comprehensively gathered
and independently veried all records of individual victims of the
Maoist insurgency.
As a way to challenge the state legitimacy, the Maoist rebels grad-
ually established the people’s government, which served as a parallel
government similar to the state administrative mechanism at the village,
district, and central level.
8
The people’s government was responsible for
collecting taxes for administrative services, such as settling land disputes
and land transactions and consolidating the base areas (International
Crisis Group, 2005). It facilitated much of this dispute resolution work
through the people’s court. H2b suggests a negative relationship be-
tween the rebel court and the utilization of the state court system. To test
this hypothesis, a variable People’s government is used, which is a binary
variable coded “1” for the district when and where the Maoist rebel
group announced the people’s government, otherwise “0”. A central
component of the people’s government was the people’s court, where
the civilians could voluntarily bring cases for adjudication.
9
To sys-
tematize the court precedence, the Maoist party had a legal team that
developed codes to be uniformly followed in administrative units.
10
As
such, the Maoist party did not only ght against the state, but also
offered civilians an alternative institutional mechanism in which to seek
justice.
11
While the Maoist rebels never had absolute territorial control
at the district level, they had parallel governments established in 89% of
the districts (67 out of 75) before they signed a comprehensive peace
agreement and ended the conict in November 2006.
12
Subnational-level socio, economic, and political variation can affect
the subnational-level demand for judicial remedies during the armed
conict. Therefore, the empirical analyses control for the literacy rate gap
and life expectancy gap variables with data from the Central Bureau of
Statistics (1991; 2003). The 1991 census data is used for the years be-
tween 1996 and 2000 and the 2001 census data is used until 2006. Filing
a suit in a court or seeking judicial remedies can be costly for civilians.
One of the ways the cost can be incurred is in time spent travelling to
district headquarters to le a case in a district court. Therefore, the
analyses use a variable road density gap with the 1998 data from the
United Nations Development Program – Nepal (UNDP Nepal, 1998).
This variable captures the accessibility of individual districts. Further,
the analysis includes a variable political participation gap as a measure of
opportunity for civilians to express their political preferences, which
Fig. 1. Civil and criminal court cases before, during, and after the
Maoist insurgency.
5
State presence may vary signicantly within a district due to ongoing
conict. Therefore, the analysis includes a district-level xed effect.
6
It can be argued that the civilian’s use of state institutions in prosecuting
civil cases is not a perfect measure of legitimacy as it does not distinguish be-
tween the use of state institutions as an expression of support for state legiti-
macy and the use of such institutions as the outcome of a coercive process.
Nevertheless, this measure is the best available alternative.
7
In compliance with the constitution, the Supreme Court collects and pre-
sents court case information to the head of the state. The data used in this
analysis is manually coded from these reports.
8
The Maoist rebel group established the 35-member central-level people’s
government on November 23, 2003, under the leadership of Baburam Bhat-
tarai. The group announced a 75-point common minimum policy and program
that sought uniformity in the people’s government’s conduct at the village,
district, and national levels. By the time the central-level people’s government
was announced, the people’s government was already in place in many villages
and districts (See, Pathak, 2005).
9
In June 2004, the author witnessed a people’s court proceeding in a local
town in the Chitwan district. The case, held in a mobile court, involved a do-
mestic violence issue and the proceeding was open for all civilians to witness.
10
The author interviewed the lawyer who acted as chief justice in the Maoist
people’s court and a lawyer responsible for drafting these codes and training
those responsible for adjudicating the court proceedings.
11
Data on the cases led in the Maoist people’s court system does not exist.
Conversations with the rank-and-le members of the Maoist party revealed that
the court’s activities were primarily symbolic. As such, they were intended to
generate civilian obedience through intentionally creating fear amongst the
population.
12
The Maoist group declared a people’s government in two districts in 2000,
which increased to 28 districts in 2001. The rebels further announced their
government in 10 districts in 2002, 19 in 2003, 9 in 2004, and one in 2005. This
data is compiled by detailed readings of the available documents that led to the
identication of the chiefs who oversaw the rebel government in each district.
The identied individuals serving in the peoples’ government were interviewed
to conrm the date of the initiation of the people’s government.
M. Joshi
Political Geography 109 (2024) 103044
6
could affect both the state legitimacy and citizen awareness. This vari-
able is created by subtracting the percentage of district-level voter
turnout from the national voter turnout data from the 1991, 1994, and
1999 parliamentary elections with data from the Election Commission of
Nepal (1991; 1994; 1999). A positive value in these variables constitutes
above average from the national value and a negative value suggests
lower than the national average. I use the term “gap” because the Maoist
conict in Nepal is explained in terms of relative inequality rather than
absolute conditions, as the rebellion initiated and gained support in
districts that were relatively impoverished. Additionally, prior studies of
the Maoist conict have utilized a comparable method in modeling
district level violence (Joshi & Mason, 2008, 2010; Murshed & Gates,
2005). The analysis also includes a variable, post-2001 period, coded “1
″
for years between 2002 and 2006, otherwise “0". The post-2001 is the
period when the constitutional monarch declared a state of emergency
and assumed executive power. Further, the number of court cases went
up from 2001 to 2002 (Fig. 1).
For a possible positive relationship between population and court
cases, the analyses control for a district’s population with a population
variable. The population data comes from annual reports of the
Department of Health Services (2006) and the variable is log normal-
ized. One of the key ndings in armed conict and state capacity liter-
ature relates to rough terrain or geography impacting the levels of
violence (Buhaug & Rød, 2006; Fearon & Laitin, 2003). Therefore, the
analysis controls the Hill region and Mountain region variables with the
Tarai (southern and plain) region as a reference category. Each of these
variables takes a value “1” if the districts fall in each of these regions,
otherwise “0”. The ve different regions in Nepal - Eastern, Central,
Western, Midwestern, and Far Western - are used as controls. Variables for
each region take a value of “1” for the district falling in each region,
otherwise “0.” The reference category is the Midwestern region, where the
Maoist conict started.
4.3. Empirical strategies and ndings
In this study, the dependent variable is the district level utilization of
the state court during the Maoist insurgency, which is a count of cases.
Therefore, the empirical analyses utilize a negative binomial model for
panel data.
13
Table 1 presents six different models, with civil cases as the
dependent variable in Models 1 and 2, criminal cases as the dependent
variable in Models 3 and 4, and all court cases, or combined civil and
criminal cases, as the dependent variable in Models 5 and 6. All of these
models also include a lagged dependent variable for cases led in the
immediate previous year, which can be a signicant predictor of cases
led in the current year. Further, statistical models presented in the
table include the Maoist violence and the state violence variables in
different models, as these two variables are highly correlated.
14
All
models include the People’s government variable, which captures the year
the Maoist rebels formally announced their government in a given
district.
Hypothesis 1 suggests a negative relationship between the state use
of violence and the number of court cases. As shown in Table 1, the
coefcient for the variable state violence is negative and signicant
across all models (p <0.001). The state violence variable is negatively
related to both civil cases and criminal cases, as well as a variable that
combines these two categories of cases. Based on models in Table 1, one
standard deviation increase in cumulative state violence leads to a more
than 19% decrease in the log count of civil cases (Model 1) and a more
than 21% decrease in criminal cases (Model 3). The effect is a 20%
decrease for all court cases (Model 5). The signicant negative effect is
presented in Fig. 2, which presents the estimated marginal effect of state
violence on both civil and criminal cases (based on Model 1 and 3 in
Table 1).
These ndings are in line with the hypothesized relationship that the
state cannot maintain its legitimacy by utilizing violence. In other
words, the nding is conclusive that the state that claims the legitimate
right to use violence decreases its legitimacy when it utilizes that right.
One possible explanation for this nding could be related to civilian
trust in the court system. In Nepal, the state directly and indirectly used
the court system to target political opposition and their civilian sup-
porters. As such, the judiciary was under the direct inuence of the state,
which civilians did not trust for fair trials. Another possible explanation
could be related to civil war violence limiting the opportunity for ci-
vilians to interact with state institutions. While plausible, this explana-
tion might have limited merit. This is particularly because data suggests
civilians were accessing public education and public health systems
during conict years. According to World Bank data, over 3.4 million
pupils were in primary education in 1996, which increased to over 4.5
million in 2006 (World Bank, 2019). Similarly, Nepal made progress in
16 out of 19 health indicators specic to the Millennium Development
Goals during the conict years (Devkota & Van Teijlingen, 2010). Yet
civilians were not bringing cases to the court system. In other words,
civilians were selectively accessing state institutions. As the state used
violence to consolidate its authority, civilians interacted less with the
court (fewer civil cases). The state also prosecuted less through criminal
cases. In other words, the state did less to maintain rule and order and
protect civilians by prosecuting more criminal cases, which would have
strengthened the state’s legitimacy. Another possible reason could be
the inuence of the rebel violence as suggested in H2.
The rebel objective of the use of violence, as explained in the theo-
retical argument, is to delegitimize the state. Therefore, rebel violence is
expected to have a negative effect on the utilization of the court system
(H2). As shown in Table 1, the relationship between the Maoist violence
variable and the civil case variable is negative in Model 2. This nding
holds for criminal cases and all court cases in Model 4 and 6 respectively.
As such, a one standard deviation increase in the Maoist violence vari-
able leads to a 13.2% decline in the log count of civil cases, a 17.6%
decline in criminal cases, and an over 12% decline in total court cases.
The signicant negative effect is shown in Fig. 3, which presents the
estimated marginal effect of rebel violence both on civil and criminal
cases (based on Model 2 and 4 in Table 1). The statistically signicant
negative coefcient lends support to the theoretical argument that rebel
violence in contested territories creates a condition of dual legitimacy,
which divides civilian loyalty between the state and the rebel groups.
This unleashes the process of declining state authority and legitimacy,
which is further affected by the rebel establishment of their government.
Therefore, the establishment of the rebel government is hypothesized to
have a negative effect on the utilization of the court system, as the ci-
vilians can approach the rebel government to redress their grievances
(H2a).
In empirical analysis, as expected, the coefcient for the variable
people’s government is negative and statistically signicant in all models
presented in Table 1. The establishment of a rebel government further
undermines state legitimacy: it reduces the number of both types of
cases. The negative and statistically signicant effect of the people’s
government exists even when the same model includes the Maoist
violence variable (Model 2, 4, and 6, Table 1). As such, the rebel offer of
an alternative institutional mechanism signicantly depressed the uti-
lization of the state court system. Further, the offering of a parallel
government has an independent, negative, and statistically signicant
effect on both the civil and criminal cases. In terms of marginal effect,
the decline is 0.28 for civil cases when the people’s government is
present, which is a decline of almost 11% (Model 2, Table 1). For
13
I compared the Poisson xed effect and negative binomial with control for
individual districts models that included all time-varying covariates. Compared
to the Poisson model, the negative binomial model provided a more accurate
estimate of the dispersion parameter and demonstrated lower Akaike infor-
mation criterion (AIC) and Bayesian information criterion (BIC) statistics. Re-
sults from LR Chi-Square statistics are in Appendix Table 3.
14
See Appendix Table 2 for the correlation matrix.
M. Joshi
Political Geography 109 (2024) 103044
7
criminal cases, the difference in margin is 0.18, a drop of nearly 7%
(Model 4, Table 1). While this nding does not suggest civilians would
prefer the people’s court over the state court system, the act of estab-
lishing a people’s government as an alternative to the state coupled with
the declining number of cases in the state court system suggests
competing legitimacy between the state and non-state actors. The
nding indicates a decline in people’s perception of government
legitimacy.
In all models, I nd that the political participation gap variable is
positive but signicant only in Model 3 and 4 (p <0.05). This suggests
districts that had a lower political participation from the national
average tended to have lower criminal cases. In addition, the literacy gap
variable is positive across all models and signicant in Model 3–6 (p <
0.05). As such, the districts with the highest literacy rate gap tend to
have fewer court cases. This could mean that civilians who are illiterate
are not aware of their rights or the court system, which they could use to
address damages and loss claims. It is also possible that rebels may in-
uence this segment of the population more, as they can be easily
indoctrinated. Taken together, these two ndings could also be a func-
tion of urban and rural district dynamics, as the state mostly caters to
urban dwellers and underserves people in rural districts.
Among the control variables, the estimated coefcients for the Post-
2001 period variable are positive across all models, except Model 4,
which aligns with the observed increase in cases, as noted in Fig. 1. The
coefcient for the hill region variable is consistently negative and sig-
nicant (p <0.05), as compared to the Tarai region. The mountain region
variable is also consistently negative, but signicant only in models for
civil and total cases (p <0.05). Similarly, there is a substantial variation
across regions. The far western region variable is positive and signicant
only for civil and total cases models (p <0.05). The western region
variable is negative and signicant only for criminal cases. The esti-
mated coefcients for the central region variable are negative across all
models except in Model 1 and 6 when state violence is not included in
the model (p <0.05), in which case they are positive. The estimated
coefcient for the eastern region variable is insignicant. The reference
region in the analysis is the mid-western region, the epicenter where the
Maoist insurgency started.
In empirical analyses, the coefcient for the variable population is
positively related to civil cases in Model 1 (p <0.01) and negatively
related to criminal cases in Model 4 (p <0.05). The coefcient for the
variable road density gap is insignicant. In empirical studies, it is also
established that civil war violence is one of the key contributors to
population displacement (Adhikari, 2013; Czaika & Kis-Katos, 2009). In
Nepal, civilians left the district fearing targeted violence by both the
state and the Maoist group or in an effort to earn better livelihoods
(Adhikari, 2013; Joshi, 2013). Therefore, the analysis includes a
Table 1
Violence and state legitimacy.
Civil cases Criminal cases All court cases
(1) (2) (3) (4) (5) (6)
Civil cases (lag) 0.000*** 0.000***
(0.000) (0.000)
Criminal cases (lag) 0.000*** 0.000**
(0.000) (0.000)
All court cases (lag) 0.000*** 0.000***
(0.000) (0.000)
State violence (log) −0.099*** −0.108*** −0.101***
(0.012) (0.012) (0.012)
Maoist violence (log) −0.075*** −0.100*** −0.079***
(0.015) (0.014) (0.013)
People’s government (1, 0) −0.272*** −0.270*** −0.180*** −0.184*** −0.214*** −0.196***
(0.032) (0.037) (0.033) (0.035) (0.032) (0.035)
Population (log) 0.613*** 0.306 −0.177 −0.466* 0.265 −0.091
(0.161) (0.178) (0.202) (0.201) (0.213) (0.175)
Political participation gap 0.007 0.003 0.016*** 0.013** 0.004 0.000
(0.005) (0.005) (0.004) (0.005) (0.005) (0.005)
Literacy rate gap 0.001 0.001 0.015*** 0.015*** 0.007* 0.009*
(0.004) (0.004) (0.004) (0.004) (0.004) (0.004)
Road density gap −19.369 −17.179 3.552 3.429 −19.357 −20.143
(13.040) (13.071) (14.423) (14.582) (13.631) (13.878)
Post 2001 period (1,0) 0.146*** 0.093* 0.025 −0.016 0.084* 0.011
(0.036) (0.045) (0.037) (0.039) (0.038) (0.038)
Hill region (1,0) −2.389*** −2.722*** −1.668** −1.856** −2.249*** −2.596***
(0.541) (0.542) (0.582) (0.586) (0.560) (0.563)
Mountain region (1,0) −3.830* −3.972* −1.294 −1.630 −4.070* −4.627*
(1.686) (1.694) (1.865) (1.884) (1.776) (1.818)
Eastern region (1,0) 5.310 5.214 −2.332 −1.851 4.963 5.573
(3.300) (3.284) (3.654) (3.682) (3.438) (3.488)
Central region (1,0) −1.884 −2.289* −1.409 −1.836 −1.980 −2.566*
(1.088) (1.072) (1.243) (1.249) (1.151) (1.160)
Western region (1,0) −0.050 −0.442 −1.832** −2.190** −0.295 −0.730
(0.610) (0.611) (0.710) (0.714) (0.662) (0.654)
Far western region (1,0) 1.717* 1.759* 0.900 1.024 1.908** 2.053**
(0.700) (0.702) (0.732) (0.737) (0.697) (0.696)
Constant −6.149** −2.238 5.737* 9.234** −1.496 2.859
(2.334) (2.518) (2.819) (2.813) (2.885) (2.488)
Observations 825 825 825 825 825 825
Panels 75 75 75 75 75 75
Wald
χ
2
2012.80 1922.34 2513.41 2325.91 2251.35 2113.13
Probability of
χ
2
0.000 0.000 0.000 0.000 0.000 0.000
AIC 9655.3 9688.3 9256.8 9288.0 10543.2 10580.9
BIC 10051.4 10084.4 9652.9 9684.1 10939.3 10976.9
District xed effects Yes Yes Yes Yes Yes Yes
All results are based on negative binomial models. Two tail tests. *p <0.05, **p <0.01, ***p <0.001.
M. Joshi
Political Geography 109 (2024) 103044
8
variable absent population with the 2001 national census data (Central
Bureau of Statistics, 2003). Because this variable is highly correlated
with the population variable, the absent population variable is not
included in the models presented in Table 1. The ndings are presented
in Appendix Table 4b. While the main ndings related to rebel and state
violence and the people’s government hold, the absent population var-
iable is positive and insignicant in Model 1 and 2 and Model 5 and 6.
The estimated coefcient for this variable is negative and insignicant in
Model 3 and 4.
Models presented in Table 1 include relevant variables discussed in
the literature specic to state legitimacy and armed conict and those
frequently used in analyzing the district-level violence during the Maoist
insurgency in Nepal. Nevertheless, it is crucial to understand that these
Fig. 2. Effect of state violence on civil and criminal cases.
Fig. 3. Effect of rebel violence on civil and criminal cases.
Table 2
Joint signicance test of main independent variables.
Chi Square test/
Models in
Table 1
Civil Cases Criminal cases All cases
Model
1
Model
2
Model
3
Model
4
Model
5
Model
6
Test Statistics 201.72 120.57 138.52 108.20 128.33 100.47
P-value 0.000 0.000 0.000 0.000 0.000 0.000
M. Joshi
Political Geography 109 (2024) 103044
9
models may still present bias because of potential omitted variables. All
models presented in Table 1 are tested for multicollinearity and are
unaffected by this issue.
15
Further, these models are tested for joint
statistical signicance. The null hypothesis in this test is that the state
and rebel violence and the rebel offer of parallel government is jointly
zero. The test statistics, or the estimated Chi Square values, are well
above the level of 99% condence intervals, and therefore, the null
hypothesis is rejected. The test statistics are presented in Table 2.
5. Conclusion
This research developed a theoretical argument specic to the use of
the state and rebel violence in contested territories and found violence
by the state and the rebels to be counterproductive for the civilian
acceptance of the state court system. Civilians le civil cases and the
state les criminal cases for fundamentally different reasons. However,
the use of violence by the state and rebels had similar effects on both
civil and criminal cases. While rebel violence and its offer of parallel
government in contested territories delegitimize the state to a degree,
the state’s use of violence is also counterproductive to its very objective
of gaining civilian compliance to its political legitimacy in contested
conict zones.
The capacity to use and manipulate violence remains at the heart of
the legitimacy claims from both the state and the rebels. Therefore,
ndings from this research indicate the availability of effective alter-
natives for both the state and the rebels. While the state is regarded as
having the authority of monopoly on the use of violence, the actual use
of violence in a civil war is counterproductive to its legitimacy. Rather,
the state gains legitimacy when its use of violence is limited, as empir-
ically demonstrated in this research. Therefore, the nding underscores
the impact of providing goods and services on civilian acceptance of
state legitimacy, as the popular political participation and education
variables were positively correlated with the use of the state court sys-
tem for criminal cases. For the rebels, while a degree of violence can still
be necessary for safeguarding their institutions in contested areas, the
establishment of a parallel government can work as the best substitution
to the use of violence.
The political authority’s legitimacy in fragile and post-conict states
mostly relates to strengthening institutional capacity (Lake, 2013; Rot-
berg, 2003, 2010), but institutional capacity alone is insufcient for the
legitimacy of a post-war state (Mcloughlin, 2015). According to
Mcloughlin (2015), civilians’ perception of the impartiality and
distributive justice of state institutions matters, but civilians have
shifting expectations of state institutions, in part related to their expo-
sure to violence. In other words, civilians tend to reassess their expec-
tations of state institutions in zones where the state and rebel groups use
violence to gain legitimacy. Further, the post-war policy interventions
are signicantly affected by territorial contestations during the war
years (McDowell, Braniff, & Murphy, 2017). Civilians’ perceptions can
vary signicantly given the use of violence by the state and rebels in
contested territories and civilian exposure to the parallel rebel govern-
ment at the subnational level. Therefore, the question specic to the
perception of state institutions in the post-war period can be further
explored. The dynamic interplay of violence and rebel establishment of a
parallel government in the contested territories can also inuence
transitional justice demands. Given that post-Cold War civil wars are
irregular, with rebels using guerrilla tactics (Kalyvas & Balcells, 2010),
neither the state nor the rebels hold territorial control in most of these
civil wars. Therefore, research questions like these can be explored in
the context of Nepal and many other recent civil war cases.
Declaration of competing interest
None.
Data availability
The data and replication material for this study can be found on the
Harvard Dataverse at https://doi.org/10.7910/DVN/OW5OGZ.
Acknowledgment
I would like to acknowledge faculty research grant supports from the
Kroc Institute for International Peace Studies and Liu Institute for Asia
and Asian Studies at the University of Notre Dame. Grace Connors and
Clare Barloon provided excellent research assistant support. I
acknowledge the journal’s anonymous reviewers for their excellent
feedback. I am thankful for constructive comments and suggestions from
Sally Sharif, Department of Political Science at the University of British
Columbia; and remain indebted to Binod Paudel (currently a Foreign
Affairs Advisor to the Ofce of the Vice President of Nepal) who
collected and veried people’s government data used in this research.
An earlier draft of the paper was presented at the University of Notre
Dame’s Keough School of Global Affairs South Asia conference in
Washington DC, May 16–17, 2023.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.polgeo.2023.103044.
References
Adhikari, P. (2013). Conict-induced displacement, understanding the causes of ight.
American Journal of Political Science, 57(1), 82–89. https://doi.org/10.1111/j.1540-
5907.2012.00598.x
Albert, K. E. (2022). What is rebel governance? Introducing a new dataset on rebel
institutions, 1945–2012. Journal of Peace Research, 59(4), 622–630. https://doi.org/
10.1177/00223433211051848
Arjona, A. (2016). Rebelocracy: Social order in the Colombian civil war. Cambridge
University Press.
Arjona, A. (2017). Civilian cooperation and non-cooperation with non-state armed
groups: The centrality of obedience and resistance. Small Wars and Insurgencies, 28
(4–5), 755–778. https://doi.org/10.1080/09592318.2017.1322328
Balcells, L. (2012). The consequences of victimization on political identities: Evidence
from Spain. Politics & Society, 40(3). https://doi.org/10.1177/0032329211424721
Braithwaite, A. (2010). Resisting infection: How state capacity conditions conict
contagion. Journal of Peace Research, 47(3), 311–319. https://doi.org/10.1177/
0022343310362164
Breslawski, J. (2021). The social terrain of rebel held territory. Journal of Conict
Resolution, 65(2–3), 453–479. https://doi.org/10.1177/0022002720951857
Buchanan, A. (2002). Political legitimacy and democracy. Ethics, 112(4), 689–719.
https://doi.org/10.1086/340313
Buhaug, H., & Rød, J. K. (2006). Local determinants of African civil wars, 1970-2001.
Political Geography, 25(3), 315–335. https://doi.org/10.1016/j.polgeo.2006.02.005
Castillo, G.del (2008). Rebuilding war-torn states: The challenge of post-conict economic
reconstruction. Oxford University Press. New York: NY.
Central Bureau of Statistics. (1991). National population census 1991. Kathmandu, Nepal.
Central Bureau of Statistics. (2003). Mapping Nepal census indicators 2001 and trends.
Kathmandu, Nepal.
Collier, P., & Hoefer, A. (1998). On economic causes of civil war. Oxford Economic
Papers, 50(4), 563–573. https://doi.org/10.1093/oep/50.4.563
Collier, P., & Hoefer, A. (2002). On the incidence of civil war in Africa. Journal of
Conict Resolution, 46(1), 13–28. https://doi.org/10.1177/0022002702046001002
Collier, P., & Hoefer, A. (2004). Greed and grievance in civil war. Oxford Economic
Papers, 56(4), 563–595. https://doi.org/10.1093/oep/gpf064
Colombia Reports. (2019). Extrajudicial Executions. Colombia Reports. June 14, 2019 http
s://colombiareports.com/amp/false-positives/.
Cronin-Furman, K. (2020). Human rights half measures: Avoiding accountability in
postwar Sri Lanka. World Politics, 72(1). https://doi.org/10.1017/
S0043887119000182
Cunningham, K. G., & Loyle, C. E. (2021). Introduction to the special feature on dynamic
processes of rebel governance. Journal of Conict Resolution, 65(1), 3–14. https://doi.
org/10.1177/0022002720935153
Czaika, M., & Kis-Katos, K. (2009). Civil conict and displacement: Village-level
determinants of forced migration in Aceh. Journal of Peace Research, 46(3), 399–418.
https://doi.org/10.1177/0022343309102659
15
See Appendix Note 1.
M. Joshi
Political Geography 109 (2024) 103044
10
Department of Health Services. (2006). Annual reports (1991-2006). Kathmandu, Nepal.
Devkota, B., & Van Teijlingen, E. (2010). Understanding effects of armed conict on
health outcomes: The case of Nepal. Conict and Health, 4(1), 1–8. https://doi.org/
10.1186/1752-1505-4-20/TABLES/2
Dowd, C., & Raleigh, C. (2013). Sahel state political violence in comparative perspective.
Stability, 2(2). https://doi.org/10.5334/sta.bl
Election Commission of Nepal. (1991). House of representative election - 1991. Kathmandu,
Nepal.
Election Commission of Nepal. (1994). House of representative election - 1994. Kathmandu,
Nepal.
Election Commission of Nepal. (1999). House of representative election - 1999. Kathmandu,
Nepal.
Fearon, J. D., & Laitin, D. D. (2003). Ethnicity, insurgency, and civil war. American
Political Science Review, 97(1), 75–90. https://doi.org/10.1017/
S0003055403000534
Gilley, B. (2006a). The meaning and measure of state legitimacy: Results for 72
countries. European Journal of Political Research, 45(3), 499–525. https://doi.org/
10.1111/j.1475-6765.2006.00307.x
Gilley, B. (2006b). The meaning and measure of state legitimacy: Results for 72
countries. European Journal of Political Research, 45(3), 499–525. https://doi.org/
10.1111/j.1475-6765.2006.00307.x
Gurses, M., & Mason, T. D. (2010). Weak states, regime types, and civil war. Civil Wars,
12(1–2), 140–155. https://doi.org/10.1080/13698249.2010.492952
Hendrix, C. S. (2010). Measuring state capacity: Theoretical and empirical implications
for the study of civil conict. Journal of Peace Research, 47(3), 273–285. https://doi.
org/10.1177/0022343310361838
Huang, R. (2016a). The wartime origins of democratization: Civil war, rebel governance, and
political regimes. Cambridge University Press.
Huang, R. (2016b). The wartime origins of democratization: Civil war, rebel governance, and
political regimes. The wartime origins of democratization: Civil war, rebel governance, and
political regimes. Cambridge University Press. https://doi.org/10.1017/
CBO9781316711323
Hultman, L. (2007). Battle losses and rebel violence: Raising the costs for ghting.
Terrorism and Political Violence, 19(2), 205–222. https://doi.org/10.1080/
09546550701246866
International Crisis Group. (2005). Nepal’s Maoists: Their aims, structure and strategy Asia
report N◦104. Brussels.
Joshi, M. (2013). Livelihood coping mechanisms, local intelligence, and the pattern of
violence during the maoist insurgency in Nepal. Terrorism and Political Violence, 25
(5), 820–839. https://doi.org/10.1080/09546553.2012.700657
Joshi, M., & Mason, T. D. (2008). Between democracy and revolution: Peasant support
for insurgency versus democracy in Nepal. Journal of Peace Research, 45(6),
765–782. https://doi.org/10.1177/0022343308096155
Joshi, M., & Mason, T. D. (2010). Land tenure, democracy, and patterns of violence
during the maoist insurgency in Nepal, 1996-2005. Social Science Quarterly, 91(4),
984–1006. https://doi.org/10.1111/j.1540-6237.2010.00745.x
Joshi, M., & Mason, T. D. (2011). Peasants, patrons, and parties: The tension between
clientelism and democracy in Nepal. International Studies Quarterly, 55(1), 151–175.
https://doi.org/10.1111/j.1468-2478.2010.00639.x
Joshi, M., & Pyakurel, S. R. (2015). Individual-level data on the victims of Nepal’s civil
war, 1996–2006: A new data set. International Interactions, 41(3), 601–619. https://
doi.org/10.1080/03050629.2015.987345
Joshi, M., & Quinn, J. M. (2017). Who kills whom? The micro-dynamics of civilian
targeting in civil war. Social Science Research, 63, 227–241. https://doi.org/
10.1016/j.ssresearch.2016.09.003
Kalyvas, S. N. (2003). The ontology of ‘political violence’: Action and identity in civil
wars. Perspectives on Politics, 1(3), 475–494. https://doi.org/10.1017/
S1537592703000355
Kalyvas, S. N. (2006). The logic of violence in civil war. Oxford University Press.
Kalyvas, S. N., & Balcells, L. (2010). International system and technologies of rebellion:
How the end of the cold war shaped internal conict. American Political Science
Review, 104(3), 415–429. https://doi.org/10.1017/S0003055410000286
Kaplan, O. (2013). Protecting civilians in civil war: The institution of the ATCC in
Colombia. Journal of Peace Research, 50(3). https://doi.org/10.1177/
0022343313477884
Kasr, N. (2015). Rebel governance–constructing a eld of inquiry: Denitions, scope,
patterns, order, causes. In A. Arjona, K. Nelson, & M. Zachariah (Eds.), 46Rebel
governance in civil war (p. 21). Cambridge University Press.
Lake, D. A. (2013). Building legitimate states after civil wars. In Strengthening peace in
post-civil war states. https://doi.org/10.7208/chicago/9780226351261.003.0002
Ledwidge, F. (2017). Rebel law: Insurgents, courts and justice in modern conict. Oxford
University Press.
Lichbach, M. I. (1998). The rebel’s dilemma. University of Michigan Press.
Lichbach, M. I. (1998). The rebel’s dilemma cited by me. https://books.google.
com/books?hl=en&lr=&id=CWJIl5A5NOMC&oi=fnd&pg=PR23&ots=ukVtd3
1LXr&sig=GveSFZ0y_RVsx8Mbpr521c73sIw.
Lichtenheld, A. G., & Schon, J. (2021). The consequences of internal displacement on
civil war violence: Evidence from Syria. Political Geography, 86. https://doi.org/
10.1016/j.polgeo.2021.102346
Linz, J. J. (2019). Legitimacy of democracy and the socioeconomic system. In M. Dogan
(Ed.), Comparing pluralist democracies (pp. 65–113). Routledge. https://doi.org/
10.4324/9780429036620-3.
Loyle, C. E. (2021). Rebel justice during armed conict. Journal of Conict Resolution, 65
(1), 108–134. https://doi.org/10.1177/0022002720939299
Loyle, C. E., & Binningsbø, H. M. (2018). Justice during armed conict: A new dataset on
government and rebel strategies. Journal of Conict Resolution, 62(2), 442–466.
https://doi.org/10.1177/0022002716655441
Loyle, C. E., Braithwaite, J. M., Cunningham, K. G., Huang, R., Joseph Huddleston, R.,
Jung, D. F., et al. (2022a). Revolt and rule: Learning about governance from rebel
groups. International Studies Review, 24(4). https://doi.org/10.1093/isr/viac043
Loyle, C. E., Braithwaite, J. M., Cunningham, K. G., Huang, R., Joseph Huddleston, R.,
Jung, D. F., et al. (2022b). Revolt and rule: Learning about governance from rebel
groups. International Studies Review, 24(4). https://doi.org/10.1093/isr/viac043
Loyle, C. E., Cunningham, K. G., Huang, R., & Jung, D. F. (2021). New directions in rebel
governance research. Perspectives on Politics. https://doi.org/10.1017/
S1537592721001985
Mampilly, Z. (2011). Rebel rulers: Insurgent governance and civilian life during war. Ithaca,
New York: Cornell University Press.
Mampilly, Z. (2015). Performing the nation-state: Rebel governance and symbolic
processes. In A. Arjona, N. Kasr, & Z. Mampilly (Eds.), Rebel governance in civil war
(Vols. 74–97). Cambridge University Press. https://doi.org/10.1017/
CBO9781316182468.004.
Mason, T. D., & Krane, D. A. (1989). The political economy of death squads: Toward a
theory of the impact of state-sanctioned terror. International Studies Quarterly, 33(2),
175. https://doi.org/10.2307/2600536
Mason, T. D., & Michael Greig, J. (2017). State capacity, regime type, and sustaining the
peace after civil war. International Interactions, 43(6), 967–993. https://doi.org/
10.1080/03050629.2017.1260012
McDowell, S., Braniff, M., & Murphy, J. (2017). Zero-sum politics in contested spaces:
The unintended consequences of legislative peacebuilding in Northern Ireland.
Political Geography, 61(November), 193–202. https://doi.org/10.1016/j.
polgeo.2017.09.001
Mcloughlin, C. (2015). When does service delivery improve the legitimacy of a fragile or
conict-affected state? Governance, 28(3), 341–356. https://doi.org/10.1111/
gove.12091
Megoran, N. (2011). War and peace? An agenda for peace research and practice in
geography. Political Geography, 30(4), 178–189. https://doi.org/10.1016/j.
polgeo.2010.12.003
Metelits, C. (2009). Inside insurgency: Violence, civilians, and revolutionary group behavior.
Inside insurgency: Violence, civilians, and revolutionary group behavior. NYU Press.
https://doi.org/10.5860/choice.47-7123
Murshed, S. M., & Gates, S. (2005). Spatial-horizontal inequality and the maoist
insurgency in Nepal. Review of Development Economics, 9(1), 121–134. https://doi.
org/10.1111/j.1467-9361.2005.00267.x
Onesto, L. (2005). Dispatches from the people’s war in Nepal. London: Pluto Press.
Pathak, B. (2005). Politics of people’s war and human rights in Nepal. Kathmandu, Nepal:
Bimipa Publications.
Podder, S. (2017). Understanding the legitimacy of armed groups: A relational
perspective. Small Wars and Insurgencies, 28(4–5), 686–708. https://doi.org/
10.1080/09592318.2017.1322333
Rotberg, R. I. (2003). Failed states, collapsed states, weak states: Causes and indicators.
In State failure and state weakness in a time of terror (Vols. 1–25)Brookings Institution
Press.
Rotberg, R. I. (2010). The failure and collapse of nation-states breakdown, prevention,
and repair. In When states fail: Causes and consequences. https://doi.org/10.1515/
9781400835799-003
Rothstein, B. (2011). The quality of government: Corruption, social trust, and inequality in
international perspective. University of Chicago Press.
Rubin, M. A. (2020). Rebel territorial control and civilian collective action in civil war:
Evidence from the communist insurgency in the Philippines. Journal of Conict
Resolution, 64(2–3), 459–489. https://doi.org/10.1177/0022002719863844
Sack, R. D. (1986). Human territoriality: Its theory and history. Cambridge University Press.
Schlichte, K., & Schneckener, U. (2015). Armed groups and the politics of legitimacy.
Civil Wars, 17(4), 409–424. https://doi.org/10.1080/13698249.2015.1115573
Sooka, Y. (2006). Dealing with the past and transitional justice: Building peace through
accountability. International Review of the Red Cross, 88(862), 311–325. https://doi.
org/10.1017/S1816383106000543
Souleimanov, E. A., Siroky, D. S., & Colombo, R. (2022). Blood revenge in civil war: Proof
of concept. Security Studies. https://doi.org/10.1080/09636412.2023.2153731
Steele, A. (2009). Seeking safety: Avoiding displacement and choosing destinations in
civil wars. Journal of Peace Research, 46(3), 419–430. https://doi.org/10.1177/
0022343309102660
Stewart, M. A. (2018). Civil war as state-making: Strategic governance in civil war.
International Organization. https://doi.org/10.1017/S0020818317000418
Stewart, M. A. (2021a). Governing for revolution: Social transformation in civil war.
Cambridge university Press.
Stewart, M. A. (2021b). Governing for revolution: Social transformation in civil war.
Cambridge University Press.
Supreme Court of Nepal. (2006). Annual Reports (1996-2006). Kathmandu, Nepal htt
ps://supremecourt.gov.np/web/yearlyreps.
Thies, C. G. (2010). Of rulers, rebels, and revenue: State capacity, civil war onset, and
primary commodities. Journal of Peace Research, 47(3), 321–332. https://doi.org/
10.1177/0022343310361841
Tilly, C. (1977). From mobilization to revolution. Menlo Park, California: Addison-Wesley.
Tilly, C. (2018). From mobilization to revolution. In Collective violence, contentious
politics, and social change (Vols. 71–91)Routledge. https://doi.org/10.4324/
9781315205021-5.
UNDP Nepal. (1998). Nepal human development report 1998. Kathmandu, Nepal: UNDP.
UNDP Nepal. (2002). Nepal human development report 2001. Kathmandu, Nepal.
United People’s Front Nepal. (1996). 40 point demand. Kathmandu, Nepal.
M. Joshi
Political Geography 109 (2024) 103044
11
Wallensteen, P., & Axell, K. (1993). Armed conict at the end of the cold war, 1989-92.
Journal of Peace Research, 30(3), 331–346. https://doi.org/10.1177/
0022343393030003007
Waterman, A. (2023). The shadow of ‘the boys:’ rebel governance without territorial
control in Assam’s ULFA insurgency. Small Wars and Insurgencies, 31(4), 279–304.
https://doi.org/10.1080/09592318.2022.2120324
Weber, M. (1946). Politics as a vocation. In F. M. Weber (Ed.), Hans gerth and C. Wright
mills (Vols. 77–128). New York: Oxford University Press.
Whelpton, J. (2005). A history of Nepal. A history of Nepal. Cambridge University Press.
https://doi.org/10.1017/cbo9781107050860
Wickham-Crowley, T. P. (1987). The rise (and sometimes fall) of guerrilla governments
in Latin America. Sociological Forum, 2(3), 473–499. https://doi.org/10.1007/
BF01106622
Wickham-Crowley, T. P. (1990). Terror and guerrilla warfare in Latin America,
1956–1970. Comparative Studies in Society and History, 32(2), 201–237. https://doi.
org/10.1017/S0010417500016467
Wig, T., & Tollefsen, A. F. (2016). Local institutional quality and conict violence in
africa. Political Geography, 53, 30–42. https://doi.org/10.1016/j.
polgeo.2016.01.003
World Bank. (2019). World development indicator. Washington, D.C.: World Bank.
Worrall, J. (2017). (Re-)Emergent orders: Understanding the negotiation(s) of rebel
governance. Small Wars and Insurgencies, 28(4–5), 709–733. https://doi.org/
10.1080/09592318.2017.1322336
Yami, H. (2021). Hisila: From revolutionary to rst lady. Penguin Books.
Ying, L. (2021). How state presence leads to civil conict. Journal of Conict Resolution,
65(2–3), 506–533. https://doi.org/10.1177/0022002720957063
M. Joshi
