Rise or Recede?

Rise or Recede? Tobias Ide

How Climate Disasters Affect Armed
Conºict Intensity

Both

scholars

E
policymakers have expressed concerns about the security implications of cli-
mate change, with disasters such as storms, ºoods, or droughts playing a key
role in these debates.1 During a high-level United Nations (UN) Sicurezza
Council meeting on climate change and security in late September 2021,
French Foreign Minister Jean-Yves Le Drian stated: “In recent years, droughts,
ºoods, storms, tropical cyclones and extreme temperatures have directly
caused nearly 2 million deaths, to say nothing of the human tragedies
linked to the conºicts sometimes precipitated by these disasters.” Vietnamese
President Nguyên Xuân Phúc highlighted that “successive natural disasters
have led to considerable loss of life and property” and indicated that climate
change has grave implications for livelihood and food security. Allo stesso modo,
identiªed climate
Tunisian Minister for Foreign Affairs Othman Jerandi
change as a key driver of increasing disaster frequency and severity. “We can
no longer overlook the extent to which climate change exacerbates elements of
fragility and instability,” he said.2

Disasters also played a key role in early climate security assessments.3 In
2007, the inºuential German Advisory Council on Global Change report
identiªed a “climate-induced increase in storm and ºood disasters” as
one of the most relevant conºict constellations.4 Subsequently, scholars de-
bated how droughts affected the onset of civil wars in Darfur (Sudan) E

Tobias Ide is Senior Lecturer in Politics and International Relations at Murdoch University, Australia, E
Specially Appointed Professor for Peace and Sustainability at Hiroshima University, Japan.

The author thanks Jon Barnett, Michael Brzoska, Halvard Buhaug, Luke Derrick, Arden Haar, Ilan
Kelman, and the anonymous reviewers for helpful feedback on previous versions of the article.
Research for this article was funded by the Australian Research Council under the Discovery Early
Career Researcher Award (DECRA) scheme (DE190101268). An online appendix is available at
https://doi.org/10.7910/DVN/QOZX0G.

1. For an overview, see Katie Peters, “Disasters, Climate Change, and Securitisation: The United
Nations Security Council and the United Kingdom’s Security Policy,” Disasters 42, NO. S2 (2018):
S196–S214, https://doi.org/10.1111/disa.12307.
2. UN Security Council, 8864th Meeting: Maintenance of International Peace and Security: Climate and
Sicurezza (New York: United Nations, 2021), quotes on 7, 10, 15.
3. Joshua W. Busby et al., “Climate Change and Insecurity: Mapping Vulnerability in Africa,” In-
ternational Security 37, NO. 4 (Primavera 2013): 132–172, https://doi.org/10.1162/ISEC_a_00116.
4. German Advisory Council on Global Change (WBGU), World in Transition: Climate Change as a
Security Risk (Berlin: WBGU, 2007), 3.

International Security, Vol. 47, No. 4 (Primavera 2023), 50–78, https://doi.org/10.1162/isec_a_00459
© 2023 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
Published under a Creative Commons Attribution 4.0 Internazionale (CC BY 4.0) licenza.

50

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 51

Syria.5 These debates tied in with earlier work in disaster studies that analyzed
the impact of disasters—climate-related or not—on the onset and termination
of various forms of conºict.6

Disasters, understood here as extreme natural events with signiªcant and
adverse societal impacts, hence play a crucial role in the burgeoning research
ªeld of climate change and security. This ªeld considers disasters, along with
temperature and precipitation anomalies, as a threat multiplier that can in-
crease the risk of conºict onset or incidence.7 Disaster-conºict interlinkages
also matter for wider debates in security studies, peace and conºict research,
and international relations. Analysts predict an increase in the number and se-
verity of disasters in the next decades because of climate change, urbanization,
socioeconomic inequalities, and the concentration of populations and infra-
structure in vulnerable areas (per esempio., coastlines and river deltas). Disasters and
disaster-related damage have increased since the 1980s, while some studies
detect a similar trend for disaster-related fatalities. These patterns are particu-
larly robust for climate-related events.8 If climate-related disasters affect politi-
cal stability and conºict, then managing disaster impacts should become a
crucial topic for security politics and international cooperation.

While acknowledging the relevance of disasters for human insecurity,9
gender-based violence,10 and political conºicts such as state repression11 or

5. Konstantin Ash and Nick Obradovitch, “Climatic Stress, Internal Migration, and Syrian Civil
War Onset,” Journal of Conºict Resolution 64, NO. 1 (2020): 3–31, https://doi.org/10.1177%2F00220
02719864140; Alexander De Juan, “Long-Term Environmental Change and Geographical Patterns
of Violence in Darfur, 2003–2005,” Political Geography 45 (2015): 22–33, https://doi.org/10.1016/
j.polgeo.2014.09.001.
6. UN. Cooper Drury and Richard Stuart Olson, “Disasters and Political Unrest: An Empirical In-
vestigation,” Journal of Contingencies and Crisis Management 6, NO. 3 (1998): 153–161, https://
doi.org/10.1111/1468-5973.00084; Ilan Kelman, Disaster Diplomacy: How Disasters Affect Peace and
Conºict (London: Routledge, 2012); Enrico L. Quarantelli and Russell R. Dynes, “Community
Conºict: Its Absence and Presence in Natural Disasters,” Mass Emergencies 1, NO. 1 (1976): 139–152.
7. Cesare M. Scartozzi, “Reframing Climate-Induced Socio-Environmental Conºicts: A Systematic
Review,” International Studies Review 23, NO. 3 (2021): 696–725, https://doi.org/10.1093/isr/
viaa064.
8. Nicolas Boccard, “Analysis of Trends in Disaster Risk,” International Journal of Disaster Risk Re-
duction 53, NO. 1 (2021): 101989, https://doi.org/10.1016/j.ijdrr.2020.101989; Guiseppe Formetta
and Luc Feyen, “Empirical Evidence of Declining Global Vulnerability to Climate-Related Haz-
ards,” Global Environmental Change 57 (2019): 101920, https://doi.org/10.1016/j.gloenvcha.2019
.05.004.
9. Formetta and Feyen, “Empirical Evidence of Declining Global Vulnerability to Climate-Related
Hazards.”
10. Alyssa Mari Thurston, Heidi Stöckl, and Meghna Ranganathan, “Natural Hazards, Disasters
and Violence against Women and Girls: A Global Mixed-Methods Systematic Review,” BMJ Global
Health 6, NO. 4 (2021): e004377, https://doi.org/10.1136/bmjgh-2020-004377.
11. Clair Apodaca, State Repression in Post-Disaster Societies (New York: Routledge, 2017).

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 52

protests,12 this article focuses on disasters and intrastate armed conºicts. Fol-
lowing Nils-Petter Gleditsch et al., I deªne the latter as “a contested incom-
patibility that concerns government or territory or both where the use of
armed force between two parties results in at least 25 battle-related deaths,"
with one party being the government and its main challenger being a domestic
armed group.13

There have been tremendous contributions to the study of climate change,
disasters, and conºict in recent years. Nina von Uexkull et al. ªnd that
droughts during the growing season increase armed conºict incidence,14 E
Ramesh Ghimire and Susana Ferreira report a positive relationship between
ºoods and armed conºict incidence.15 Several other scholars argue that
climate-related disasters increase the likelihood of armed conºict onset16 and
increase civil war duration.17 That said, some studies remain skeptical of
whether disasters drive armed conºict onset,18 and other researchers argue
that disasters can even facilitate conºict termination.19

This article makes four key contributions to this burgeoning literature.
Primo, scholars have thus far mostly focused on disasters and armed conºict on-
set or incidence, and to a lesser degree on duration. By contrast, very little is

12. Kristina Petrova, “Natural Hazards, Internal Migration, and Protests in Bangladesh,"Giornale di
Peace Research 58, NO. 1 (2021): 33–49, https://doi.org/10.1177/0022343320973741.
13. Nils Petter Gleditsch et al., “Armed Conºict 1946–2001: A New Dataset,” Journal of Peace Re-
search 39, NO. 5 (2002): 615–637, esp. 618–619, https://doi.org/10.1177/0022343302039005007. If not
stated otherwise, I use the terms “conºict” and “armed conºict” synonymously.
14. Nina von Uexkull et al., “Civil Conºict Sensitivity to Growing-Season Drought,” Proceedings of
the National Academy of Sciences 113, NO. 44 (2016): 12391–12396, https://doi.org/10.1073/pnas
.1607542113.
15. Ramesh Ghimire and Susana Ferreira, “Floods and Armed Conºict,” Environment and Develop-
ment Economics 21, NO. 1 (Febbraio 2016): 23–52, https://doi.org/10.1017/S1355770X15000157.
16. Ash and Obradovitch, “Climatic Stress, Internal Migration”; Tobias Ide et al., “Multi-Method
Evidence for When and How Climate-Related Disasters Contribute to Armed Conºict Risk,"
Global Environmental Change 62 (2020): 102063, https://doi.org/10.1016/j.gloenvcha.2020.102063;
Philip Nel and Marjolein Righarts, “Natural Disasters and the Risk of Violent Civil Conºict,” Inter-
national Studies Quarterly 52, NO. 1 (Marzo 2008): 159–185, https://doi.org/10.1111/j.1468-2478.2007
.00495.X.
17. Joshua Eastin, “Fuel to the Fire: Natural Disasters and the Duration of Civil Conºict,” Interna-
tional Interactions 42, NO. 2 (2016): 322–349, https://doi.org/10.1080/03050629.2016.1115402.
18. Drago Bergholt and Päivi Lujala, “Climate-Related Natural Disasters, Economic Growth, E
Armed Civil Conºict,” Journal of Peace Research 49, NO. 1 (2012): 147–162, https://doi.org/10.1177/
0022343311426167; Christiane J. Fröhlich, “Climate Migrants as Protestors? Dispelling Misconcep-
tions about Global Environmental Change in Pre-Revolutionary Syria,” Contemporary Levant 1,
NO. 1 (2016): 38–50, https://doi.org/10.1080/20581831.2016.1149355; Ole Magnus Theisen, Helge
Holtermann, and Halvard Buhaug, “Climate Wars? Assessing the Claim That Drought Breeds
Conºict,” International Security 36, NO. 3 (Inverno 2011/12): 79–106, https://doi.org/10.1162/ISEC_a
_00065.
19. Yasutaka Tominaga and Chia-yi Lee, “When Disasters Hit Civil Wars: Natural Resource Ex-
ploitation and Rebel Group Resilience,” International Studies Quarterly 65, NO. 2 (2021): 423–434,
https://doi.org/10.1093/isq/sqab014.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 53

known about how disasters affect the dynamics of ongoing armed conºicts.
The few existing studies on the topic mostly focus on either international con-
ºicts20 or individual countries such as Sri Lanka, Syria, or the Philippines.21 To
gain a full picture of how climate-related disasters affect security risks, more
information is required on whether such disasters shape the intensity of ongo-
ing armed conºicts. The resulting knowledge could help decision-makers as-
sess under which circumstances delivering humanitarian aid may be too
dangerous (if ªghting intensity increases), and whether attempting mediation
is feasible (if the conºict de-escalates). With armed conºicts and disaster inten-
sity and frequency all on the rise, disaster and conºict zones are increasingly
likely to overlap, making such knowledge even more valuable.22

Secondo, most discussion on the conºict implications of both climate change
and disasters focuses on whether they increase armed conºict risks.23 By also
focusing on conºict de-escalation, I include the possibility of at least tempo-
rary reductions in conºict risks after disasters.

Third, this article examines the causal pathways through which a change in
conºict intensity happens. It does so by qualitatively tracing disaster-conºict
interlinkages and by including relevant context factors for (de-)escalation into
the cross-case analysis. Given that “the mechanisms of climate–conºict link-
ages remain a key uncertainty”24 and that the pathways connecting disasters
to conºicts remain understudied,25 this article addresses an important knowl-
edge gap.

20. Ilan Kelman, “Connecting Theories of Cascading Disasters and Disaster Diplomacy,” Interna-
tional Journal of Disaster Risk Reduction 30, Part B (2018): 172–179, https://doi.org/10.1016/j.ijdrr
.2018.01.024.
21. Joshua Eastin, “Hell and High Water: Precipitation Shocks and Conºict Violence in the Phil-
ippines,” Political Geography 63 (2018): 116–134, https://doi.org/10.1016/j.polgeo.2016.12.001;
Kyosuke Kikuta, “Postdisaster Reconstruction as a Cause of Intrastate Violence: An Instrumental
Variable Analysis with Application to the 2004 Tsunami in Sri Lanka,” Journal of Conºict Resolution
63, NO. 3 (2019): 760–785, https://doi.org/10.1177/0022002717753919; Andrew M. Linke and Brett
Ruether, “Weather, Wheat and War: Security Implications of Climate Variability for Conºict in
Syria,” Journal of Peace Research 58, NO. 1 (2021): 114–131, https://doi.org/10.1177/00223433209
73070.
22. Shawn Davies, Therése Pettersson, and Magnus Öberg, “Organized Violence 1989–2021 and
Drone Warfare,” Journal of Peace Research 59, NO. 4 (2022): 593–610, https://doi.org/10.1177/002234
33221108428.
23. Jon Barnett, “Global Environmental Change I: Climate Resilient Peace?,” Progress in Human
Geography 43, NO. 5 (2019): 927–936, https://doi.org/10.1177/0309132518798077.
24. Katharine J. Mach et al., “Climate as a Risk Factor for Armed Conºict,” Nature 571 (2019): 193–
197, esp. 193, https://doi.org/10.1038/s41586-019-1300-6.
25. Jiuping Xu et al., “Natural Disasters and Social Conºict: A Systematic Literature Review,” In-
ternational Journal of Disaster Risk Reduction 17 (2016): 38–48, http://dx.doi.org/10.1016/j.ijdrr.2016
.04.001.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 54

Fourth and relatedly, existing work mostly focuses on structural context fac-
tors that make disaster-conºict linkages more likely, such as ethnic exclusion
and low economic development.26 By contrast, I also include dynamic,
situation-speciªc factors, including how the disaster affects the capabilities of
and power relations between government and rebels. I further elaborate on
each of these four points below.

My article addresses these lacunae by analyzing the impact of climate-
related disasters on armed conºict dynamics. Primo, I argue that disasters can
result in armed conºict escalation or de-escalation through various pathways.
In particular, I introduce a power differential mechanism, highlighting how di-
sasters change power relations between groups and how such changes affect
conºict dynamics. Then, I combine quantitative data with rich qualitative in-
formation using qualitative comparative analysis (QCA) to conduct a unique
empirical analysis of twenty-one cases of disasters striking armed conºict
zones. The ªnal two sections present and discuss the study’s core ªndings and
conclude the article.

Pathways Linking Disasters and Conºict Intensity

This article follows the UN General Assembly’s deªnition of a disaster as a
“serious disruption of the functioning of a community or a society at any scale
due to hazardous events interacting with conditions of exposure, vulnerability
and capacity, leading to one or more of the following: human, Materiale, eco-
nomic and environmental losses and impacts.”27 There are no so-called natural
disasters because natural hazards alone (per esempio., a riverine ºood) are insufªcient
to cause signiªcant losses and other damage. Piuttosto, hazards need to strike
populated and vulnerable areas (per esempio., places with weak early warning systems
and insufªcient river embankments) to have socioeconomic impacts. Related-
ly, climate-related disasters result from a climatic hazard (per esempio., extreme temper-
ature, precipitation, or wind) striking a vulnerable society.

Disasters can increase conºict risks by causing or exacerbating strong
grievances. Extreme weather events are usually associated with signiªcant
personal and material losses. In 2021 alone, disasters caused 9,200 fatalities
and economic losses worth $280 billion.28 If survivors blame governance actors 26. Ide et al., “Multi-Method Evidence”; von Uexkull et al., “Civil Conºict Sensitivity.” 27. UN General Assembly, Report of the Open-Ended Intergovernmental Expert Working Group on Indi- cators and Terminology Relating to Disaster Risk Reduction (New York: UN, 2016), 13. 28. Munich Re, NatCatSERVICE, “Factsheet: Natural Catastrophes in 2021,” January 2022, https:// l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u / i s e c / a r t i c e – p d l f / / / / 4 7 4 5 0 2 0 9 0 9 8 6 / i s e c _ a _ 0 0 4 5 9 p d . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Rise or Recede? 55 (usually state authorities) for those losses and insufªcient relief or recovery support, tensions between the governing and the governed will rise. Infatti, various studies show how disasters can cause a steep decline of trust in and support for authorities,29 even though it remains unclear whether such a de- cline affects willingness to support an armed rebellion. Tensions are particularly acute if (perceived) disaster-related inequalities overlap with existing political, socioeconomic, or ethnic cleavages. Govern- ments have historically fueled such sentiments: by providing insufªcient re- lief, by evicting victims from coastal areas to make space for tourism or military projects, or by increasing repression to maintain stability after extreme events.30 Furthermore, migration facilitated by disasters can result in ethnic tensions, resource competition, or frustration about low government support in receiving areas.31 While these insights on disaster-related grievances are mostly derived from the literature on armed conºict onset and incidence, they can also be valid for conºict escalation. People may switch their support to an opposing conºict party if they become frustrated or disappointed by how governance actors (sometimes rebel groups, but usually governments) respond to a disaster. If there is a social contract between the government/rebels and the population, the latter might also urge their “violent representatives” to express grievances by staging further attacks.32 Furthermore, disasters can act as symbolic trigger events that political elites draw on to justify further violence against an oppo- nent.33 Similarly, research tends to associate disasters with negative percep- www.munichre.com/content/dam/munichre/mrwebsiteslaunches/natcat-2022/2021_Figures- of-the-year.pdf/_jcr_content/renditions/original./2021_Figures-of-the-year.pdf. 29. Sung Hoon Kang and Mark Skidmore, “The Effects of Natural Disasters on Social Trust: Evidence from South Korea,” Sustainability 10, NO. 9 (2018): 1–16, https://doi.org/10.3390/su100 92973; Gabriel Katz and Ines Levin, “The Dynamics of Political Support in Emerging Democracies: Evidence from a Natural Disaster in Peru,” International Journal of Public Opinion Research 28, NO. 2 (Estate 2016): 173–195, https://doi.org/10.1093/ijpor/edv010. 30. Apodaca, State Repression in Post-Disaster Societies; Donald Grasse et al., “Opportunistic Re- pression: Civilian Targeting by the State in Response to COVID-19,” International Security 46, NO. 2 (2021): 130–165, https://doi.org/10.1162/isec_a_00419; Naomi Klein, The Shock Doctrine: The Rise of Disaster Capitalism (New York: Picador, 2007). 31. Vally Koubi et al., “The Determinants of Environmental Migrants’ Conºict Perception,” Inter- national Organization 72, NO. 4 (2018): 905–936, https://doi.org/10.1017/S0020818318000231. 32. Ted Robert Gurr, Why Men Rebel (Princeton, NJ: Princeton University Press, 1970). 33. Stuart J. Kaufman, “Symbolic Politics or Rational Choice? Testing Theories of Extreme Ethnic Violence,” International Security 30, NO. 4 (Primavera 2006): 45–86, https://doi.org/10.1162/isec.2006.30 .4.45. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u / i s e c / a r t i c e – p d l f / / / / 4 7 4 5 0 2 0 9 0 9 8 6 / i s e c _ a _ 0 0 4 5 9 p d . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 International Security 47:4 56 tions of outgroups,34 droughts with higher support for political violence,35 and rapid-onset disasters with more frequent civil unrest.36 Another pathway connecting disasters to higher conºict intensity is based on strategic incentives and opportunities. If government forces are unable to access an area, for instance because roads are ºooded or destroyed, rebels can expand their inºuence.37 A government’s capabilities can be weakened when it deploys security forces and reconstruction funds to affected areas or when it incurs disaster-induced tax income losses. After the 1999 Quindío earth- quake in Colombia, for instance, the government sent an additional six thou- sand soldiers and police to assist with the disaster response, while the country’s gross domestic product (GDP) declined by 6 percent in the three months after the event.38 Rebels can also be weakened by disasters—for in- stance, if they receive or extort lower contributions from the disaster-affected population. Inoltre, the state can use disaster relief to gain sympathies and collect information in rebel strongholds, thereby enabling pro-government forces to scale up their attacks.39 Disasters can also facilitate pro-government militias’ and rebel groups’ re- cruitment efforts. Engaging in violent conºict poses a high risk of injury and death, but people who have lost their livelihoods because of a disaster and who have few legal options to earn an income tend to be more willing to accept such risks.40 This is particularly the case because disasters can result in sustained GDP reductions.41 There is also evidence that armed groups may ex- tract or raid some of the humanitarian aid ºowing into a disaster-affected area 34. Eunbin Chung and Inbok Rhee, “Disasters and Intergroup Peace in Sub-Saharan Africa,” Jour- nal of Peace Research 59, NO. 1 (2022): 58–72, https://doi.org/10.1177/00223433211065249. 35. Nina von Uexkull, Marco d’Errico, and Julius Jackson, “Drought, Resilience, and Support for Violence: Household Survey Evidence from DR Congo,” Journal of Conºict Resolution 64, NO. 10 (2020): 1994–2021, https://doi.org/10.1177/0022002720923400. 36. Peter F. Nardulli, Buddy Peyton, and Joseph Bajjalieh, “Climate Change and Civil Unrest: The Impact of Rapid-Onset Disasters,” Journal of Conºict Resolution 59, NO. 2 (2015): 310–335, https:// doi.org/10.1177/0022002713503809. 37. Michael Brzoska, “Weather Extremes, Disasters, and Collective Violence: Conditions, Mecha- nisms, and Disaster-Related Policies in Recent Research,” Current Climate Change Reports 4 (2018): 320–329, https://doi.org/10.1007/s40641-018-0117-y. 38. Dawn Brancati, “Political Aftershocks: The Impact of Earthquakes on Intrastate Conºict,” Jour- nal of Conºict Resolution 51, NO. 5 (2007): 715–743, https://doi.org/10.1177/0022002707305234. 39. Eastin, “Hell and High Water.” 40. Joshua Eastin and Steven T. Zech, “Environmental Pressures and Pro-Government Militias: Evidence from the Philippines,” Conºict Management and Peace Science (2022, online ªrst): 1–22, https://doi.org/10.1177/07388942221110128. 41. Nourin Shabnam, “Natural Disasters and Economic Growth: A Review,” International Journal of Disaster Risk Science 5, NO. 2 (2014): 157–163, https://doi.org/10.1007/s13753-014-0022-5. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u / i s e c / a r t i c e – p d l f / / / / 4 7 4 5 0 2 0 9 0 9 8 6 / i s e c _ a _ 0 0 4 5 9 p d . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Rise or Recede? 57 and use the revenues to boost their war efforts.42 After the 2010 ºoods in Pakistan, for instance, the UN Ofªce for the Coordination of Humanitarian Affairs channeled more than $1.3 billion into the worst-affected regions,
raising concerns that Islamist groups might beneªt from the aid.43

Based on these considerations, I formulate the following hypothesis:

H1: Climate-related disasters facilitate the escalation of armed conºicts.

there are also several

Yet,
theoretical and empirical reasons for why
disasters could facilitate armed conºict de-escalation. Per esempio, Alexander
De Juan and Niklas Hänze ªnd that social trust tends to increase in post-
disaster settings.44 Contrary to the grievances approach, disaster sociology has
long emphasized that at the local level, “the net result of most disasters is a
dramatic increase in social solidarity among the affected populace during the
emergency and immediate post emergency periods.”45 Disasters can be a
shared threat that increases incentives for cooperation and facilitates a com-
mon identity among a “community of sufferers.”46 The post-disaster uptick
in solidarity, trust, and cooperation may not only engender ceaseªres but
also cause conºict parties to scale down armed conºict
intensity, at
least momentarily.47

Inoltre, there are several ways that disasters can reduce conºict inten-
sity by restraining conºict parties’ violent actions. Perhaps most obviously, UN
storm that causes ºooding or destroys bridges can inhibit the mobility of
troops and restrict their ability to engage in battles. The same is true if ªghters
are injured or killed during a disaster, or if bases and equipment are damaged.

42. Kikuta, “Postdisaster Reconstruction.”
43. United Nations Ofªce for the Coordination of Humanitarian Affairs, “Humanitarian Appeals,"
accessed March 8, 2023, https://www.unocha.org/media-centre/news-updates/humanitarian-
appeals.
44. Alexander De Juan and Niklas Hänze, “Climate and Cohesion: The Effects of Droughts on
Intra-Ethnic and Inter-Ethnic Trust,” Journal of Peace Research 58, NO. 1 (2021): 151–67, https://
doi.org/10.1177/0022343320974096.
45. Charles E. Fritz and Harry B. Williams, “The Human Being in Disasters: A Research Perspec-
tive,” ANNALS of the American Academy of Political and Social Science 309, NO. 1 (1957): 42–51,
esp. 48, https://doi.org/10.1177/000271625730900107.
46. Quarantelli and Dynes, “Community Conºict.”
47. Joakim Kreutz, “From Tremors to Talks: Do Natural Disasters Produce Ripe Moments for Re-
solving Separatist Conºicts?,” International Interactions 38, NO. 4 (2012): 482–502, http://dx.doi.org/
10.1080/03050629.2012.697404; Rune T. Slettebak, “Don’t Blame the Weather! Climate-Related Nat-
ural Disasters and Civil Conºict,” Journal of Peace Research 49, NO. 1 (2012): 163–176, https://
doi.org/10.1177/0022343311425693.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 58

IL 2004 tsunami in Aceh, for instance, killed 2,698 Indonesian security
forces and an unknown number of Free Aceh Movement rebels. It also caused
considerable destruction to military bases and insurgent supply lines along
the coast.48

Disaster-induced agricultural shocks and food price spikes can compromise
the ability of armed groups (including state forces) to feed their ªghters, hence
forcing them to scale down their operations. Even if they can extract food
by force, doing so risks lowered public support.49 Given a disaster’s economic
impacts, revenues tend to decline for both the state (from taxes) and the rebels
(from decreased levels of voluntary contributions and extortion). At the same
time, rebels (just like governments) often channel additional resources to those
disaster-ridden areas where they have some territorial control or a social con-
tract with the population, thus further straining their capacities.50 Islamist in-
surgents, for instance, were among the ªrst to provide disaster assistance after
IL 2010 ºoods in Pakistan, and their ability to generate funds in these regions
was reduced for several months.51 The simultaneous increase in demand for
and decrease of armed groups’ capabilities could result in a (temporary) Rif-
duction of violence.52

My second hypothesis, running contrary to H1, is thus:

H2: Climate-related disasters facilitate the de-escalation of armed conºicts.

Armed conºict intensity is usually driven by a range of political, economic,
and social factors. Therefore, disasters are likely to affect conºict dynamics
only under certain circumstances and through speciªc pathways. The existing
literature on climate-related disasters and armed conºict onset or incidence

48. Philippe Le Billon and Arno Waizenegger, “Peace in the Wake of Disaster? Secessionist
Conºicts and the 2004 Indian Ocean Tsunami,” Transactions of the Institute of British Geographers 32,
NO. 3 (Luglio 2007): 411–427, https://doi.org/10.1111/j.1475-5661.2007.00257.x.
49. Benamin E. Bagozzi, Ore Koren, and Bumba Mukherjee, “Droughts, Land Appropriation, E
Rebel Violence in the Developing World,” Journal of Politics 79, NO. 3 (Luglio 2017): 1057–1072, http://
dx.doi.org/10.1086/691057.
50. Colin Walch, “Weakened by the Storm: Rebel Group Recruitment in the Wake of Natural Di-
sasters in the Philippines,” Journal of Peace Research 55, NO. 3 (2018): 336–350, https://doi.org/
10.1177/0022343317741535.
51. Ayesha Siddiqi, “Climatic Disasters and Radical Politics in Southern Pakistan: The Non-Linear
Connection,” Geopolitics 19, NO. 4 (2014): 885–910, http://dx.doi.org/10.1080/14650045.2014
.920328.
52. For a similar argument on conºict incidence, see Idean Salehyan and Cullen S. Hendrix, “Cli-
mate Shocks and Political Violence,” Global Environmental Change 28 (settembre 2014): 239–250,
http://dx.doi.org/10.1016/j.gloenvcha.2014.07.007.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 59

identiªes three factors that facilitate disaster-conºict linkages: a high economic
dependence on agriculture, low levels of socioeconomic development, and the
political exclusion of ethnic groups.53 Yet, as with other variables on climate,
disasters, and conºict (per esempio., population density, presence of transport infra-
structure), these factors are mostly structural and change slowly over time. By
contrasto, one key feature of disasters is that they can catalyze large-scale, short-
term social changes, such as rapid economic declines or extraordinary govern-
ment and civil society responses.54 As a consequence, research needs to pay
greater attention to dynamic, situation-speciªc factors that shape disaster-
conºict linkages.

I argue that disasters can trigger short-term changes in power relations be-
tween governments and rebel groups, which has follow-on effects on armed
conºict intensity. International relations research has long discussed how ten-
sions in the international system can rise if weaker actors, dissatisªed with the
status quo, gain power relative to stronger actors.55 Likewise, Kyosuke Kikuta
ªnds that sudden perceived changes in the balance of power during civil wars
cause more violence because the beneªting party seeks to capitalize on its ad-
vantage, while the opponent tries to prevent such a move.56

Such arguments are applicable to the impact of disasters on armed conºict
dynamics. Disasters can strengthen one armed conºict actor relative to an-
other, either because a conºict party beneªts from the disaster (per esempio., Quando
it can recruit deprived survivors or loot international aid), or because a conºict
party suffers from the disaster (per esempio., when its income is reduced or when it
needs to divert resources for the disaster response). In such a situation, IL
conºict party that beneªts from the disaster (relative to the other) could use
this advantage militarily and hence escalate the ªghting, which results in
higher armed conºict intensity. After the 2004 tsunami in Sri Lanka, for
instance, both the government and the Liberation Tigers of Tamil Eelam be-
lieved that they could scale up their respective military activities to capitalize
on their opponent’s disaster-related weakening.57

53. Joshua Busby, States and Nature: The Effects of Climate Change on Security (Cambridge: Camera-
bridge University Press, 2021); Ide et al., “Multi-Method Evidence”; von Uexkull et al., “Civil
Conºict Sensitivity.”
54. Jörg Birkmann et al., “Extreme Events and Disasters: A Window of Opportunity for Change?
Analysis of Organizational, Institutional and Political Changes, Formal and Informal Responses
after Mega-Disasters,” Natural Hazards 55, NO. 3 (2010): 637–655, https://doi.org/10.1007/s11069-
008-9319-2.
55. Jacek Kugler and Douglas Lemke, eds., Parity and War: Evaluations and Extensions of the War
Ledger (Ann Arbor: University of Michigan Press, 1996).
56. Kikuta, “Postdisaster Reconstruction.”
57. Idil Tunçer-Kilavuz, “Success or Failure in the Peace Processes of Aceh and Sri Lanka: A Com-

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 60

It is also possible, Tuttavia, that both the government and the rebels
are weakened by the disaster and thus scale down their military activi-
ties, leading to a decline in conºict intensity. Per esempio, IL 1997 ºoods
in Somalia heavily constrained the income sources (cioè., cattle and ba-
nana exports) and mobility of the two competing United Somali Congress
leaving them unable to pay their respective ªghters and stage
factions,
new offensives.58

The third hypothesis of this article tests this power differential mechanism:

H3: If disasters beneªt one armed conºict party relative to the other, IL
conºict will escalate. D'altra parte, a conºict will de-escalate if both
conºict parties suffer signiªcant adverse impacts from the disaster.

Studying the Disaster-Conºict Nexus with QCA

sample of cases

The sample of this study consists of twenty-one cases where major climate-
related disasters struck armed conºict zones. Such a medium-N approach has
two major advantages. Primo, it allowed me to collect qualitative information
on key factors for which cross-case quantitative data were not readily avail-
able, such as the disasters’ impacts on the armed conºict parties and on overall
conºict dynamics. Secondo, evidence on various disaster and conºict types
from fourteen different countries makes this analysis more comprehensive
than single or small-N comparative case studies.

I chose the country as the unit of analysis in this article for all government-
rebel conºict dyads because several pathways connecting disasters to conºict
intensity (per esempio., migration, loss of tax revenues, enhanced national solidarity)
operate beyond the local level.59 Using the International Disasters Database
(EM-DAT)60 and the Uppsala Conºict Data Program (UCDP),61 I compiled a

parative Study,” Terrorism and Political Violence 31, NO. 4 (2019): 712–732, https://doi.org/10.1080/
09546553.2017.1282860.
58. Peter D. Little, Hussein Mahmoud, and D. Layne Coppock, “When Deserts Flood: Risk Man-
agement and Climatic Processes among East African Pastoralists,” Climatic Research 19 (2001): 149–
59, https://doi.org/10.3354/cr019149.
59. The exception to this country level of analysis is India, an extremely large country with multi-
ple ongoing armed conºicts and major disasters in any given year. For Indian cases, the state (ªrst-
level administrative unit) is the unit of analysis.
60. Centre for Research on the Epidemiology of Disasters, EM-DAT: The International Disasters
Database, 2021, https://www.emdat.be/.
61. Gleditsch et al., “Armed Conºict 1946–2001.”

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 61

list of all country years with an ongoing armed conºict that were struck by
a disaster from 1990 A 2020 that caused more than 1,000 deaths. EM-DAT in-
cludes a disaster event in its database if it meets at least one of the following
criteria: ten or more deaths, one hundred or more people affected, the govern-
ment declares a state of emergency, or the government requests international
assistance. I limited the sample to a maximum of three cases per country be-
cause India and the Philippines would otherwise dominate the sample. To in-
crease sample heterogeneity regarding both countries and disaster types, IL
sample also includes ªve countries where highly destructive disasters struck
but caused fewer than 1,000 deaths (Burundi, Egypt, Indonesia, Nepal, E
Uganda).62 The fourteen countries in my sample are depicted in ªgure 1.

assessing disaster-related conºict dynamics

The outcomes of interest of this article are disaster-related armed conºict esca-
lation and de-escalation as indicated by the number and intensity of battles be-
tween the government and a rebel group. I used a two-staged procedure to
assess where the outcome is present. Primo, I used information from UCDP’s
Georeferenced Event Dataset to calculate the difference (in standard devia-
zioni) between the monthly number of battle-related deaths twelve months be-
fore and after the disaster month(S).63 Secondo, I assessed a wide range of
academic studies, gray literature, and media reports to determine whether the
quantitative trends are supported by qualitative evidence, and whether the re-
spective disasters facilitated these trends.

The results highlight the relevance of this procedure. In several cases,
large standard deviations in battle-related deaths were not disaster-related
(per esempio., Pakistan 2015 con (cid:2)16.22, or Nepal 1996 con (cid:3)7.10). In other contexts,
disaster-related changes in armed conºict intensity were masked by rather
small standard deviations (per esempio., Indonesia 1997 con (cid:3)1.77).64

To test H3 and to disentangle the causal pathways and context factors con-
necting disasters to armed conºict (de-)escalation, I ran the QCA with eight
conditions.65 The ªrst two conditions are structural factors that researchers

62. All results reported below are robust to dropping those ªve additional cases. The online ap-
pendix provides details and reports the results of all robustness tests.
63. Ralph Sundberg and Erik Melander, “Introducing the UCDP Georeferenced Event Dataset,"
Journal of Peace Research 50, NO. 4 (2013): 523–532, https://doi.org/10.1177/0022343313484347.
64. See table S1 in the online appendix.
65. Table S4 in the online appendix provides extensive information on how each of these eight
conditions was operationalized using which data. Running a QCA with twenty-one cases and
eight conditions poses a problem because too many variables are used in an analysis with too few

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 62

Figura 1. Map of Sample Countries Where Major Climate-Related Disasters Struck

Armed Conºict Zones

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

NOTE: The shaded areas represent the sample of this study, which consists of twenty-one
cases where major climate-related disasters struck armed conºict zones. In Bangladesh,
India, Pakistan, the Philippines, and Somalia, several cases were analyzed. Table S1 in the
online appendix provides short summaries of how disasters shaped conºict dynamics in
each case.

have identiªed as relevant scope conditions for disaster-related conºict onset
and incidence (but not yet for conºict intensity).66

The ªrst structural factor is agricultural dependence (agridep). Agricultural
economies and livelihoods are particularly vulnerable to climate-related disas-
ters because crops are highly sensitive to extreme weather events, and ªelds

cases. To mitigate this problem, I subsequently ran core models with the ªve conditions that are
analytically relevant for each outcome, and the results did not change. Axel Marx and Adrian
Dusa, “Crisp-Set Qualitative Comparative Analysis (csQCA), Contradictions and Consistency
Benchmarks for Model Speciªcation,” Methodological Innovations Online 6, NO. 2 (2011): 103–148,
https://doi.org/10.4256/mio.2010.0037.
66. Ide et al., “Multi-Method Evidence”; von Uexkull et al., “Civil Conºict Sensitivity.”

Rise or Recede? 63

often take several months to recover after a disaster. The resulting economic
losses can lead to intense grievances among the affected population and facili-
tate recruitment by armed groups.67 But these losses also reduce the conºict
parties’ revenue base and necessitate increased cooperation across socioeco-
nomic cleavages.68

Poverty (poverty) is the second structural factor. Poor, marginalized commu-
nities tend to disproportionately suffer from a disaster because they tend to
live in substandard housing and in at-risk areas. They usually also have fewer
means to recover in the aftermath of a disaster. Inoltre, state emergency
services more frequently ignore areas with a low socioeconomic proªle.69 Poor
people consequently face lower opportunity costs to join an armed group
after a disaster, particularly if they also develop disaster-related grievances
(per esempio., against state institutions). Allo stesso tempo, high levels of post-disaster
cooperation are reported in regions with high poverty rates.70 Under-ªve mor-
tality rates are a well-established indicator for poverty, with national-level data
available from the World Bank and (only for India) state-level data provided
by the National Family Health Survey.71 To account for endogeneity, I used
data from the pre-disaster year.

I also ran a robustness test for a third widely discussed structural factor for
disaster-conºict linkages, namely the exclusion of ethnic groups from political
power. This robustness test yielded no signiªcant results.72

I then added two conditions measuring how the disaster affected the conºict
parties. Disasters can affect rebels in various adverse ways, for instance by kill-
ing their ªghters, destroying their infrastructure, undermining their revenue
fonti, forcing them to invest resources into the disaster response, or inhibit-
ing their logistics. An adverse impact on rebels (impactrebel) was present if the
disaster affected the rebels’ ªghting capabilities through one or several of these

67. Eastin and Zech, “Environmental Pressures and Pro-Government Militias”; von Uexkull et al.,
“Civil Conºict Sensitivity.”
68. Quarantelli and Dynes, “Community Conºict”; Salehyan and Hendrix, “Climate Shocks and
Political Violence.”
69. Ben Wisner et al., At Risk: Natural Hazards, People’s Vulnerability, and Disasters (London:
Routledge, 2004).
70. Rajesh Venugopal and Sameer Yasir, “The Politics of Natural Disasters in Protracted Conºict:
IL 2014 Flood in Kashmir,” Oxford Development Studies 45, NO. 4 (2017): 424–442, http://
dx.doi.org/10.1080/13600818.2016.1276160.
71. International Institute for Population Sciences, National Family Health Survey (NFHS-5), 2019–
21: India Report (Mumbai: Ministry of Health and Family Welfare, 2022), http://rchiips.org/nfhs/
index.shtml; World Bank, World Bank Open Data, Mortality Rate Under-5 (per 1,000 Live Births),
2022, https://data.worldbank.org/indicator/SH.DYN.MORT?view(cid:4)chart.
72. See the online appendix for the robustness test results.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 64

pathways. Allo stesso modo, impactgov measures a negative disaster impact on the
government. According to the qualitative literature, this condition indicates
whether the disaster signiªcantly and negatively affected the ªghting capabili-
ties of government troops or pro-government militias through the loss of
ªghters, resources, infrastructure, troop mobility, or legitimacy.

The impact of disasters on power relations between a government and a
rebel group is likely shaped by pre-disaster conditions, and speciªcally by
the strength of the respective conºict parties. Two conditions account for such
strength: capable state (capablestate) and weak rebels (weakrebel). A disaster is
unlikely to weaken a state that has stable revenue sources, a strong bureau-
cracy, and a large military, all of which make it harder for rebels to exploit any
opportunity arising in the post-disaster period. Likewise, a capable state can
intensify its attacks to exploit the rebels’ disaster-induced weakness. Non-
capable states, by contrast, could face more rebel attacks after the disaster, for
instance when states have to deploy security forces for the disaster response
and are less likely to capitalize on rebel weakness. I utilize a composite indica-
tor to measure state capability because the latter is a multidimensional con-
cept.73 This indicator combines information on government effectiveness,
military personnel, and the government’s ability to tax economic output.

Strong rebels can intensify their struggle if they perceive the govern-
ment to be weakened by the disaster, or if additional opportunities for re-
cruitment or ªnancing arise. Weak rebel groups, by contrast, are more likely to
be either preoccupied with the disaster response or incapable of upscaling
their activities.

The two ªnal conditions considered in the main analysis relate to the spatial
patterns of the disaster impacts: sovrapposizione (sovrapposizione) and rebel dependence
(rebeldepend). An overlap of the disaster-affected area and armed conºict zone
tends to facilitate several of the mechanisms discussed previously. For in-
stance, a disaster’s adverse effect on the rebel group or government forces is
more likely, and more relevant for conºict dynamics, if it occurs where bases
are established, troops operate, and ªghting takes place. This adverse impact
is most obvious for logistical challenges and for supply chain interruptions
(per esempio., destroyed infrastructure). Likewise, enhanced post-disaster solidarity or
grievances are more likely to affect conºict dynamics if they occur in areas
where the conºict takes place.

The rebeldepend condition measures whether rebels depend on the local pop-

73. Cullen S. Hendrix, “Measuring State Capacity: Theoretical and Empirical Implications for the
Study of Civil Conºict,” Journal of Peace Research 47, NO. 3 (2010): 273–285, https://doi.org/10.1177/
0022343310361838.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 65

ulation in disaster-affected areas for hideouts, supplies, informazione, or recruit-
ment. If the population is severely deprived by a disaster, it is less likely to
have spare food, funds, or workforce that rebels can acquire via donation or
appropriation. Rebels can respond to this situation either by temporarily
downscaling their activities74 or by escalating violence against civilians to en-
force compliance.75 Alternatively, rebels could be pushed by their constituen-
cies, whose grievances have intensiªed due to an insufªcient disaster response
by the government, to scale up their military activities.

Apart from the political exclusion of ethnic groups, I include three further
conditions in various robustness tests: the inºow of large amounts of interna-
tional aid, perceptions of an unfair distribution of disaster-related relief and
aid delivery, and the democratic nature of the political system. None of these
conditions holds explanatory power in my analysis.76

qualitative comparative analysis

This article uses QCA to analyze when and why disasters contribute to the
(de-)escalation of armed conºicts. QCA is increasingly used in security stud-
ies and international relations and has several key advantages.77 It can com-
bine quantitative data (per esempio., on poverty) and qualitative data (per esempio., on disaster
impacts on rebels) in a single analysis. Inoltre, QCA has been designed
to deal with medium-N samples like the one used in this article. My analysis
therefore responds to Lars-Erik Cederman and Manuel Vogt’s call for “con-
ºict scholars . . . to steer a middle course between overgeneralized macro-
models and myopic microinvestigations.”78 Finally, QCA is well-suited to
identify complex causal patterns,79 three of which are particularly relevant for
my analysis.

Primo, QCA can identify “conjunctural causation,” or when various condi-
tions interact to produce an outcome.80 Previous research indicates that several
conditions need to be present simultaneously for climate-related disasters to

74. Salehyan and Hendrix, “Climate Shocks and Political Violence.”
75. Bagozzi, Koren, and Mukherjee, “Droughts, Land Appropriation.”
76. See the online appendix for the respective tests and information on all coding decisions.
77. Tobias Ide and Patrick A. Mello, “QCA in International Relations: A Review of Strengths, Pit-
falls, and Empirical Applications,” International Studies Review 24, NO. 1 (Marzo 2022): viac008,
https://doi.org/10.1093/isr/viac008.
78. Lars-Erik Cederman and Manuel Vogt, “Dynamics and Logics of Civil War,” Journal of Conºict
Resolution 61, NO. 9 (2017): 1992–2016, esp. 2008, https://doi.org/10.1177/0022002717721385.
79. Carsten Q. Schneider and Claudius Wagemann, Set-Theoretic Methods for the Social Sciences:
A Guide to Qualitative Comparative Analysis (Cambridge: Cambridge University Press, 2012).
80. The term “conjunctural causation” is commonly used by set-theoretic approaches and QCA
research.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 66

increase the risk of armed conºict onset or incidence.81 Such conjunctural cau-
sation is also plausible for conºict intensity. For instance, conºicts might
escalate only if the disaster weakens the government, and the rebels are simul-
taneously strong and can recruit from a poor population. Likewise, several fac-
tors discussed previously could facilitate conºict escalation or de-escalation.82
QCA can recognize such conjunctural pathways where the same factors can fa-
cilitate different outcomes, depending on the context.

Secondo, QCA can uncover the existence of several parallel pathways leading
to an outcome (per esempio., conºict escalation), each of which may involve different
causal conditions (equiªnality). For instance, an armed conºict may escalate
after a disaster either because government forces try to exploit rebel weakness,
or because rebels intensify their military activities in response to disaster-
induced grievances.

The third complex pattern that QCA can identify is causal asymmetry, O
when different (combinations of) conditions explain the presence and absence
of an outcome. The literature on armed conºicts points out that conºict escala-
tion and de-escalation are distinct processes with different driving factors.83
Instead of assuming that the absence of the factors that drive disaster-related
conºict escalation will facilitate de-escalation, QCA identiªes separate path-
ways to each outcome.

I used the crisp-set (cioè., binary) version of QCA to disentangle necessary
and sufªcient conditions for disaster-related armed conºict (de-)escalation.
The analysis conceived all twenty-one cases to be members ((cid:4) 1) or non-
members ((cid:4) 0) of different sets84 during the so-called calibration process
(which transforms raw data into binary values of either 0 O 1). Afterward,
I built a truth table (which lists all possible combinations of conditions),85 pop-

81. Busby, States and Nature; von Uexkull et al., “Civil Conºict Sensitivity.”
82. Per esempio, a high poverty level could escalate an armed conºict if it allows conºict parties
to recruit more disaster-deprived survivors. But it can also facilitate conºict de-escalation if preva-
lent poverty impedes the rebels’ efforts to raise resources from the affected population after the
disaster.
83. Christopher Blattman and Edward Miguel, “Civil War,” Journal of Economic Literature 48, NO. 1
(Marzo 2010): 3–57, https://www.doi.org/10.1257/jel.48.1.3; Cederman and Vogt, “Dynamics and
Logics of Civil War.”
84. An example of such a crisp-set is the set of cases where government forces were adversely
Haffected by the disaster (in contrast to the set of cases where such adverse impacts were not
present).
85. For instance, cases with high poverty rates, capable states, and weak rebels (row 1), cases with
high poverty rates, capable states, and strong rebels (row 2), and cases with low poverty rates, NO
capable states, and weak rebels (row 3).

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 67

ulated it by the empirical evidence, and then minimized the truth table using
the Quine-McCluskey algorithm.86 Given that QCA is still an emerging
method in the ªelds of security studies and international relations, I explain
this procedure and how I employed it for my dataset in greater detail in the
online appendix.

How Disasters Affect Armed Conºict Escalation and De-Escalation

Tavolo 1 summarizes the results of my qualitative analysis for all twenty-one
cases of disaster–armed conºict intersections discussed in this article.87 After
disasters, armed conºicts escalate in six of the twenty-one cases (29 per cento),
de-escalate in seven cases (33 per cento), and in eight cases (38 per cento) clima-
related disasters have no relevant
impact on armed conºict dynamics.
Typically, a disaster-conºict intensity nexus was absent for two reasons. Primo,
the disaster had no signiªcant impact on the conºict parties (per esempio., IL 2008 cold
spell in northwestern Afghanistan affected a region of very low strategic rele-
vance for the government and the Taliban).88 Secondo, other, strong drivers of
conºict (de-)escalation were present (per esempio., Maoist rebels decided to act on
widespread grievances and intensiªed the conºict before the 1996 ºoods
in Nepal).89

Overall, these ªndings provide some conditional support for both H1 and
H2: disasters can facilitate both armed conºict escalation and de-escalation.
These relations are by no means deterministic. In most cases, disasters did not
lead to higher (71 per cento) or lower (67 per cento) armed conºict intensity, E
disasters were never the only (and often not the most important) driver
of conºict dynamics. Yet, major climate-related disasters also affected armed
conºict dynamics in 62 percent of all cases analyzed.

In the next step, I conducted a QCA to study when and why disasters facili-
tate armed conºict escalation or de-escalation. Following established QCA
practice, I started by analyzing necessary conditions.90 Of all eight conditions

86. Patrick A. Mello, Qualitative Comparative Analysis: An Introduction to Research Design and Appli-
catione (Washington, DC: Georgetown University Press, 2021); Schneider and Wagemann, Set-
Theoretic Methods for the Social Sciences.
87. Table S3 in the online appendix includes short qualitative summaries of each case.
88. Antonio Giustozzi, The Taliban at War: 2001–2018 (New York: Oxford University Press, 2019).
89. Sonali Deraniyagala, “The Political Economy of Civil Conºict in Nepal,” Oxford Development
Studi 33, NO. 1 (2005): 47–62, https://doi.org/10.1080/13600810500099659.
90. Schneider and Wagemann, Set-Theoretic Methods for the Social Sciences.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 68

Tavolo 1. Disaster Impacts on Conºict Intensity

Case

Bangladesh 2007

Disaster

cyclone

Burundi 2005–2006

drought

Indonesia 1997

drought

Myanmar 2008

cyclone

Pakistan 2010

Somalia 1997

ºood

ºood

Somalia 2010–2011

drought

Bangladesh 1991

cyclone

Egypt 1994

India (Andhra
Pradesh, Orissa) 1999

ºood

cyclone

India (Assam) 1998

ºood

Tajikistan 1992

ºood

Uganda 1999–2001

drought

Conºict

government vs. Purbo
Banglar Communist
Party–Janajuddha faction

government vs. Party for
the Liberation of the
Hutu People–National
Liberation Forces

government vs. National
Council of the Maubere
Resistance

government vs. Karen
National Union

government vs. Tehrik-i-
Taliban Pakistan

United Somali Congress
(Mahdi faction) vs.
United Somali Congress
(Aideed faction)

government vs. Al-
Shabaab

government vs. People’s
Solidarity Association/
Peace Forces

government vs. al-
Gama’a al-Islamiyya

government vs. People’s
War Group

government vs. United
Liberation Front of
Assam

government vs. United
Tajik Opposition

government vs. Lord’s
Resistance Army

Afghanistan 2008

cold spell

government vs. Taliban

India (Assam) 2007

ºood

Nepal 1996

ºood

government vs. United
Liberation Front of
Assam

government vs.
Communist Party of
Nepal (Maoist)

Disaster-
Conºict Link

de-escalation

de-escalation

de-escalation

de-escalation

de-escalation

de-escalation

de-escalation

escalation

escalation

escalation

escalation

escalation

escalation

none

none

none

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 69

Disaster

Conºict

Disaster-
Conºict Link

Tavolo 1. Continued

Case

Pakistan 2015

Philippines 1991

heat
wave

cyclone,
ºood

Philippines 2012

cyclone

Philippines 2013

cyclone

Russia 2010

heat
wave

government vs. Tehrik-i-
Taliban Pakistan

government vs.
Communist Party of the
Philippines–New People’s
Army

government vs. Abu
Sayyaf Group

government vs.
Communist Party of the
Philippines–New People’s
Army

government vs. Forces of
the Caucasus Emirate

none

none

none

none

none

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

NOTE: See the online appendix for more detailed descriptions of and quantitative data for

each case.

in the main model and the four conditions explored in robustness checks, two
pass the consistency threshold of 0.9 recommended for necessary conditions:
agricultural dependence (1.00 for escalation and de-escalation) and poverty
(also 1.00 for escalation and de-escalation). In other words, all cases where di-
sasters affected armed conºict intensity (for good or for bad) featured econo-
mies that were strongly dependent on the agricultural sector and that had low
levels of economic development.

This ªnding is in line with earlier studies that show how these two struc-
tural conditions are important predictors of both climate- and disaster-related
conºict risks.91 Consequently, climate-related disasters affect armed conºict
dynamics only in countries that are already vulnerable to disaster impacts.
This ªnding supports the notion that climate change is a threat multiplier,92
and demonstrates again that disasters are by no means natural; they are con-
tingent on socioeconomic vulnerabilities (and the political decisions that create

91. Busby, States and Nature; Ide et al., “Multi-Method Evidence”; von Uexkull et al., “Civil
Conºict Sensitivity.”
92. Daniel Abrahams, “Conºict in Abundance and Peacebuilding in Scarcity: Challenges and Op-
portunities in Addressing Climate Change and Conºict,” World Development 132 (agosto 2020):
104998, https://doi.org/10.1016/j.worlddev.2020.104998.

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 70

such vulnerabilities). Inoltre, my analysis demonstrates for the ªrst time
that these conditions can also facilitate an (at least temporary) reduction of
armed conºict intensity, meaning that climate-related disasters can also have a
threat-reducing effect.

Before explaining the results that appear in tables 2 E 3, I brieºy introduce
the key measures of ªt for QCA results. The main results come in the form of a
solution that identiªes one or several paths (or combinations of conditions) Quello
are sufªcient for the outcome. Each solution (and each of the paths that it con-
sists of) has a certain coverage indicating how much of the outcome it explains.
For instance, in a sample with ten cases of conºict de-escalation, a coverage of
0.9 would mean that nine out of the ten cases are explained by the respective
path or solution (for crisp-set QCA). Paths come with a raw coverage (the total
percentage of cases explained by a path) and a unique coverage (the percentage
of cases explained by only a certain path and no other path forming the solu-
zione). Each path and solution also have a consistency value indicating the de-
gree to which empirical data are in line with the proposed set relations. If a
path covers ten cases in total, Per esempio, of which nine experienced disaster-
related conºict escalation and one experienced de-escalation, the consistency
would be 0.9 for escalation and 0.1 for de-escalation in a crisp-set QCA.

Tavolo 2 summarizes the QCA solution for disaster-related armed conºict es-
calation. The solution has perfect consistency and coverage scores, indicating
high explanatory power and a strong ªt between the model and the empirical
dati. Infatti, the solution explains all six cases with the conºict escalation out-
come and is present in none of the ªfteen cases where the escalation outcome
is absent.

The solution consists of three separate paths. The ªrst path is characterized
by the rebels depending on the disaster-affected population, yet not being
impactrebel * rebeldepend). In such a situation,
weakened by the disaster itself (
the rebels could recruit more soldiers from populations that were either
aggrieved because of
insufªcient government responses to the disaster
(cioè., Egypt 1994, India [Assam] 1998) or affected by disaster-related impover-
ishment (cioè., India [Assam] 1998, India [Andhra Pradesh and Orissa] 1999,
Tajikistan 1992). This recruitment resulted in a temporary change in the rela-
tive power distribution between the conºict parties. After the 1998 ºoods in
Assam, Per esempio, there was a “growing frustration and unrest in the youth
from ºood-affected families. There are instances of unemployment, impover-
ishment, and grievances against bad governance leading some youths to join

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

P
UN
l
R
e
v
o

*

e
T
UN
T
S
e
l
B
UN
P
UN
C

*

D
N
e
P
e
D
l
e
B
e
R

P
UN
l
R
e
v
o

*

l
e
B
e
R
k
UN
e
w
(cid:5)
*

l
e
B
e
R
T
C
UN
P
M
(cid:5)

io

D
N
e
P
e
D
l
e
B
e
R

*

l
e
B
e
R
T
C
UN
P
M
(cid:5)

io

H
T
UN
P

S
R
e
T
S
UN
S
io
D

R
e
T
F
UN

N
o
io
T
UN
l
UN
C
S
E

T
C
io
º
N
o
C

D
e
M
R
UN

R
o
F

N
o
io
T
tu
o
S

l

)

UN
C
Q

(

S
io
S
sì

l
UN
N
UN

e
v
io
T
UN
R
UN
P
M
o
C

e
v
io
T
UN
T
io
l
UN
tu
Q

.
2

l

e
B
UN
T

8
9
9
1

)

M
UN
S
S
UN

(

UN
io
D
N

IO

1
0
0
2
–
9
9
9
1

UN
D
N
UN
G
U

4
9
9
1

T
P
sì
G
E

0
0
.
1

7
1
.
0

5
.
0

2
9
9
1

N
UN
T
S
io
k
io
j
UN
T

)
8
9
9
1
M
UN
S
S
UN

(

4
9
9
1

T
P
sì
G
E

)
9
9
9
1

UN
S
S
io
R

O
D
N
UN

H
S
e
D
UN
R
P

UN
R
H
D
N
UN

(

UN
io
D
N

IO

UN
io
D
N

IO

2
9
9
1

N
UN
T
S
io
k
io
j
UN
T

1
9
9
1

H
S
e
D
UN
l
G
N
UN
B

4
9
9
1

T
P
sì
G
E

S
e
S
UN
C

0
0
.
1

7
1
.
0

5
.
0

0
0
.
1

7
1
.
0

7
6
.
0

e
G
UN
R
e
v
o
C

e
tu
q
N
U

io

e
G
UN
R
e
v
o
C
w
UN
R

sì
C
N
e
T
S
io
S
N
o
C

N
o
io
T
UN
l
UN
C
S
e
(cid:6)

–

P
UN
l
R
e
v
o

*

e
T
UN
T
S
e
l
B
UN
P
UN
C

*

D
N
e
P
e
D
l
e
B
e
R

(cid:3)

)
P
UN
l
R
e
v
o

*

l
e
B
e
R
k
UN
e
w
(cid:5)
(cid:3)
D
N
e
P
e
D
l
e
B
e
R
(

*

l
e
B
e
R
T
C
UN
P
M
(cid:5)

io

0
0
.
1

0
0
.
1

sì
C
N
e
T
S
io
S
N
o
C

e
G
UN
R
e
v
o
C

N
o
io
T
tu
o
S

l

–

M
o
C

S
l
e
B
e
R

e
H
T
N
o
S
T
C
UN
P
M

io

R
e
T
S
UN
S
io
D
e
S
R
e
v
D
UN

F
o
e
C
N
e
S
B
UN

e
H
T

)
1
(

:
S
R
e
T
S
UN
S
io
D
R
e
T
F
UN
N
o
io
T
UN
l
UN
C
S
e

T
C
io
º
N
o
C
D
e
M
R
UN
o
T

S
H
T
UN
P
e
e
R
H
T

S
T
H
G

io
l

io

H
G
H
N
o
io
T
tu
o
S

l

e
H
T

:
E
T
O
N

–
io
D
e
S
R
e
v
D
UN
F
o
e
C
N
e
S
B
UN
e
H
T

)
2
(

;
)
N
M
tu
o
C

l

T
F
e
l
(

T
R
o
P
P
tu
S
D
N
UN
T
N
e
M

T
io
tu
R
C
e
R

R
o
F
N
o
io
T
UN
l
tu
P
o
P
D
e
T
C
e
F
F
UN
–
R
e
T
S
UN
S
io
D
e
H
T
N
o
S
l
e
B
e
R
e
H
T

F
o
e
C
N
e
D
N
e
P
e
D
UN
H
T
io

w
D
e
N
B

io

e
H
T

)
3
(
D
N
UN

;
)
N
M
tu
o
C

l

l
UN
R
T
N
e
C
(

S
UN
e
R
UN

T
C
io
º
N
o
C
D
N
UN

R
e
T
S
UN
S
io
D
e
H
T
N
e
e
w
T
e
B
P
UN
l
R
e
v
o
N
UN
D
N
UN

S
l
e
B
e
R
G
N
o
R
T
S
H
T
io

w
N
o
io
T
C
N
tu
N
o
C
N

j

io

S
l
e
B
e
R

e
H
T
N
o
S
T
C
UN
P
M

io

R
e
T
S
UN
S

S
UN
e
R
UN

T
C
io
º
N
o
C

D
N
UN

R
e
T
S
UN
S
io
D

e
H
T

N
e
e
w
T
e
B

P
UN
l
R
e
v
o

N
UN

D
N
UN

e
T
UN
T
S

e
l
B
UN
P
UN
C

UN

H
T
io

w
N
o
io
T
UN
N
B
M
o
C

io

N

io

N
o
io
T
UN
l
tu
P
o
P

D
e
T
C
e
F
F
UN
–
R
e
T
S
UN
S
io
D

e
H
T

N
o

io

G
N
D
N
e
P
e
D

S
l
e
B
e
R

G
N
o
R
T
S

UN

D
N
UN

R
e
w
o
P

sì
R
o
T
UN
N
UN
l
P
X
e

H
G
H

io

e
T
UN
C
io
D
N

io

)
S
w
o
R

e
e
R
H
T

T
S
UN
l
(

N
o
io
T
tu
o
S

l

e
H
T

F
o

)
0
0
.
1
(

S
e
R
o
C
S

e
G
UN
R
e
v
o
C

D
N
UN

sì
C
N
e
T
S
io
S
N
o
C

T
C
e
F
R
e
P

e
H
T

.
)
N
M
tu
o
C

l

T
H
G
io
R
(

H
T
UN
P

S
io
H
T

sì
B

D
e
R
e
v
o
C

sì
l
N
o

e
S
UN
C
(cid:4)
D
e
N

io
l
R
e
D
N
tu

R
o
F

T
N
e
io
C
ª
F
tu
S
(cid:4)
(cid:6)

–

R
o
(cid:4)
(cid:3)

D
N
UN
(cid:4)
*

.
UN
T
UN
D

l
UN
C
io
R
io
P
M
e

e
H
T

D
N
UN

l
e
D
o
M
e
H
T

N
e
e
w
T
e
B

T
ª

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

sì
T
R
e
v
o
P

*

P
UN
l
R
e
v
o
(cid:5)
*

v
o
G
T
C
UN
P
M

io

P
e
D
io
R
G
UN

*

P
UN
l
R
e
v
o
(cid:5)
*

v
o
G
T
C
UN
P
M

io

l
e
B
e
R
T
C
UN
P
M

io

*

e
T
UN
T
S
e
l
B
UN
P
UN
C
(cid:5)

N
o
io
T
UN
l
UN
C
S
e
–
e
D
(cid:6)

–

)
sì
T
R
e
v
o
P
(cid:3)
P
e
D
io
R
G
UN
(

*

P
UN
l
R
e
v
o
(cid:5)
*

v
o
G
T
C
UN
P
M

io

(cid:3)

l
e
B
e
R
T
C
UN
P
M

io

*

e
T
UN
T
S
e
l
B
UN
P
UN
C
(cid:5)

0
0
.
1

0
0
.
1

6
0
0
2
–
5
0
0
2

io

D
N
tu
R
tu
B

7
0
0
2

H
S
e
D
UN
l
G
N
UN
B

6
0
0
2
–
5
0
0
2

io

D
N
tu
R
tu
B

7
0
0
2

H
S
e
D
UN
l
G
N
UN
B

0
0
.
1

0
0
.
0

9
2
.
0

0
0
.
1

0
0
.
0

9
2
.
0

7
0
0
2

H
S
e
D
UN
l
G
N
UN
B

7
9
9
1

UN
io
S
e
N
o
D
N

IO

8
0
0
2

R
UN
M
N
UN
sì
M

0
1
0
2

N
UN
T
S
io
k
UN
P

7
9
9
1

0
1
0
2

UN
io
l
UN
M
o
S

UN
io
l
UN
M
o
S

0
0
.
1

1
7
.
0

1
7
.
0

e
G
UN
R
e
v
o
C

e
tu
q
N
U

io

e
G
UN
R
e
v
o
C
w
UN
R

sì
C
N
e
T
S
io
S
N
o
C

S
e
S
UN
C

H
T
UN
P

sì
C
N
e
T
S
io
S
N
o
C

e
G
UN
R
e
v
o
C

N
o
io
T
tu
o
S

l

S
R
e
T
S
UN
S
io
D

R
e
T
F
UN

N
o
io
T
UN
l
UN
C
S
e
–
e
D

T
C
io
º
N
o
C

D
e
M
R
UN

R
o
F

N
o
io
T
tu
o
S

l

)

UN
C
Q

(

S
io
S
sì
l
UN
N
UN

e
v
io
T
UN
R
UN
P
M
o
C

e
v
io
T
UN
T
io
l
UN
tu
Q

.

3

l

e
B
UN
T

–
S
UN
S
io
D
e
H
T
H
T
io

w
N
o
io
T
C
N
tu
N
o
C
N

j

io
e
T
UN
T
S
e
l
B
UN
P
UN
C
UN
F
o
e
C
N
e
S
B
UN
e
H
T

)
1
(

:
S
R
e
T
S
UN
S
io
D
R
e
T
F
UN
N
o
io
T
UN
l
UN
C
S
e
–
e
D
T
C
io
º
N
o
C
D
e
M
R
UN
o
T

S
H
T
UN
P
e
e
R
H
T

S
T
H
G

io
l

io

H
G
H
N
o
io
T
tu
o
S
e
H
T

l

:
E
T
O
N

–
S
io
D

e
H
T

N
e
e
w
T
e
B

P
UN
l
R
e
v
o

o
N

H
T
io

w
D
e
N
B
M
o
C

io

T
N
e
M
N
R
e
v
o
G

e
H
T

N
o

S
T
C
UN
P
M

io

R
e
T
S
UN
S
io
D

e
S
R
e
v
D
UN

)
2
(

;
)
N
M
tu
o
C

l

T
F
e
l
(

S
l
e
B
e
R

e
H
T

N
o

S
T
C
UN
P
M

io

e
S
R
e
v
D
UN

io

G
N
v
UN
H

R
e
T

N

io

T
N
e
M
N
R
e
v
o
G

e
H
T

N
o

S
T
C
UN
P
M

io

R
e
T
S
UN
S
io
D

e
S
R
e
v
D
UN

)
3
(

D
N
UN

;
)
N
M
tu
o
C

l

l
UN
R
T
N
e
C
(

sì
R
T
N
tu
o
C

e
H
T

F
o

e
C
N
e
D
N
e
P
e
D

l
UN
R
tu
T
l
tu
C
io
R
G
UN

H
G
H

io

UN

D
N
UN

S
UN
e
R
UN

T
C
io
º
N
o
C

D
N
UN

R
e
T
S
UN

e
G
UN
R
e
v
o
C

D
N
UN

sì
C
N
e
T
S
io
S
N
o
C

T
C
e
F
R
e
P

e
H
T

.
)
N
M
tu
o
C

l

T
H
G
io
R
(

sì
T
R
e
v
o
P

D
UN
e
R
P
S
e
D
w
D
N
UN

io

S
UN
e
R
UN

T
C
io
º
N
o
C

D
N
UN

R
e
T
S
UN
S
io
D

e
H
T

N
e
e
w
T
e
B

P
UN
l
R
e
v
o

o
N

H
T
io

w
N
o
io
T
C
N
tu
N
o
C

j

.
UN
T
UN
D

l
UN
C
io
R
io
P
M
e

e
H
T

D
N
UN

l
e
D
o
M
e
H
T

N
e
e
w
T
e
B

T
ª

G
N
o
R
T
S

UN

D
N
UN

R
e
w
o
P

sì
R
o
T
UN
N
UN
l
P
X
e

H
G
H

io

e
T
UN
C
io
D
N

io

)
S
w
o
R

e
e
R
H
T

T
S
UN
l
(

N
o
io
T
tu
o
S

l

e
H
T

F
o

)
0
0
.
1
(

S
e
R
o
C
S

H
T
UN
P

S
io
H
T

sì
B

D
e
R
e
v
o
C

sì
l
N
o

e
S
UN
C
(cid:4)
D
e
N

io
l
R
e
D
N
tu

R
o
F

T
N
e
io
C
ª
F
tu
S
(cid:4)
(cid:6)

–

R
o
(cid:4)
(cid:3)

D
N
UN
(cid:4)
*

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 73

insurgent groups,” which the United Liberation Front of Assam rebels ex-
ploited to boost their ªghting force.93 This shift of power relations between the
government and the rebels resulted in the latter scaling up their attacks, Guida-
ing to a rise in conºict intensity.

In the second path, the rebels were neither weak nor negatively affected by
the disaster, and the armed conºict zone overlapped signiªcantly with the
impactrebel * weakrebel * sovrapposizione). As in the ªrst path,
disaster-affected area (
such a situation allowed the rebels to gain power (relative to the govern-
ment) by boosting their recruitment efforts, broadening their support bases
(per esempio., when delivering aid to the affected areas more efªciently than the gov-
ernment), and strengthening their capabilities (cioè., Egypt 1994, Tajikistan
1992). In the Bangladesh 1991 case, the government was concerned that the re-
bels could exploit anti-Bengal and anti-government sentiments in the disaster-
affected Chittagong Hills Tracts after Cyclone Gorky. It therefore preemptively
intensiªed repressive violence (often against civilians) to quell unrest.94

The third path covers three cases, but only Uganda 1999–2001 is not covered
by any other path, and indeed, the third path disappears if Uganda 1999–2001
is not part of the sample. The third path thus only explains the Ugandan case
and can be interpreted as follows: The Lord’s Resistance Army depended on
the local population (rebeldepend) for food and funds, and these resources were
increasingly scarce because of government offensives and the 1999–2001
drought that affected this population (sovrapposizione). The Lord’s Resistance Army
thus increased violence against civilians to loot food from local villages and
refugee camps in order to ªght a relatively strong government (capablestate),
resulting in higher conºict intensity.95

Tavolo 3 displays the QCA solution for armed conºict de-escalation. Again,
the solution has a perfect consistency (1.00), implying that it explains all seven
outcome cases in the sample. The solution consists of three separate paths
quasi-sufªcient for the outcome.

The ªrst path has the highest explanatory power, as it covers six out of seven

93. Partha Das, Dadul Chutiya, and Nirupam Hazarika, Adjusting to Floods on the Brahmaputra
Plains, Assam, India (Kathmandu: International Centre for Integrated Mountain Development,
2009), 22.
94. Naomi Hossain, "IL 1970 Bhola Cyclone, Nationalist Politics, and the Subsistence Crisis Con-
tract in Bangladesh,” Disasters 42, NO. 1 (Gennaio 2018): 187–203, https://doi.org/10.1111/disa
.12235.
95. Anthony Vinci, “Existential Motivations in the Lord’s Resistance Army’s Continuing Con-
ºict,” Studies in Conºict & Terrorism 30, NO. 4 (2007): 337–352, https://doi.org/10.1080/10576
100701200173.

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

International Security 47:4 74

de-escalation cases (ªve of them uniquely) and is straightforward to interpret:
the rebels were adversely affected by the disaster (impactrebel), hence staging
capablestate) was unable to exploit
fewer attacks, and the rather weak state (
this situation, resulting in an overall decrease in armed conºict activity. For ex-
ample, IL 2010 ºoods in Pakistan inundated nearly 20 percent of the country’s
territory, posing severe logistic constraints to the Tehrik-i-Taliban (TTP) rebels.
The TTP’s taxes and voluntary contributions also declined in the affected ar-
eas, and the insurgents diverted resources and personnel to relief actions.96
The Pakistani Army could not beneªt from the weakening of the rebels be-
cause it was “overstretched with ºood relief and military operations” in other
areas and also faced mobility constraints in the ºooded territory.97

In the second and third paths, the disaster strikes a vulnerable society char-
acterized by either strong agricultural dependence (agridep) or high poverty
rates (poverty). Allo stesso tempo, the disaster does not overlap with the conºict
zone ( sovrapposizione), yet it has signiªcant adverse effects on the government
(impactgov). Per esempio, the path covers cases where the government had to
deploy 40,000 security forces to the affected disaster area (Bangladesh 2007) O
incurred a GDP loss of 3.8 per cento (Burundi 2005–2006).98 The rebels in both
Bangladesh and Burundi were unable to exploit these situations because the
disasters did not overlap with the conºict areas, and the rebels either were al-
ready very weak (the Purbo Banglar Communist Party–Janajuddha faction in
Bangladesh) or were affected by the disasters (Party for the Liberation of the
Hutu People–National Liberation Forces in Burundi, which had to cope with
higher food prices and less material support from the drought-affected popu-
lation). Just like the situation described in the ªrst path, no conºict party
gained in relative power, but both faced reduced capabilities to wage violence.
To summarize the QCA results, disasters facilitate armed conºict escalation
only in countries that are already vulnerable to climate-related extreme events
(cioè., because of strong economic dependence on agriculture and prevalent
poverty). They mostly do so by enabling rebel groups to recruit more soldiers

96. Muhammad Feyyaz, “Political Economy of Tehrik-i-Taliban Swat,” Conºict and Peace Studies
Journal 4, NO. 3 (2011): 1–22; Siddiqi, “Climatic Disasters and Radical Politics.”
97. C. Christine Fair, “Pakistan in 2010: Flooding, Governmental Inefªciency, and Continued In-
surgency,” Asian Survey 51, NO. 1 (Febbraio 2011): 97–110, esp. 104, https://doi.org/10.1525/
as.2011 .51.1.97.
98. International Federation of Red Cross and Red Crescent Societies (IFRCRCS), Bangladesh: Cy-
clone Sidr, Information Bulletin no. 01/2007 (Geneva: IFRCRCS, 2007); World Bank, Republic of Bu-
rundi: Addressing Fragility and Demographic Challenges to Reduce Poverty and Boost Sustainable Growth
(Washington, DC: World Bank Group Publications, 2018).

l

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

P

:
/
/

D
io
R
e
C
T
.

M

io
T
.

e
D
tu

/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D

l

F
/

/

/

/

4
7
4
5
0
2
0
9
0
9
8
6

/
io
S
e
C
_
UN
_
0
0
4
5
9
P
D

.

F

B
sì
G
tu
e
S
T

T

o
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

Rise or Recede? 75

and broaden their support base, but only when the rebels are not negatively af-
fected by the disaster and have sufªcient capabilities. In such situations, IL
disaster facilitates an at least temporary shift of power toward the insurgents.
This ªnding provides support for the power differential mechanism underly-
ing H3: ªghting will escalate if the disaster beneªts one conºict party relative
to the other. The sample revealed only one case where the rebels suffered ad-
verse disaster impacts and the government did not (Somalia 2010, a de-escala-
tion case). Therefore, testing whether governments and rebels exploit their
respective opponents’ disaster-related weaknesses with equal frequency re-
mains a task for future research.

Disasters also facilitate armed conºict de-escalation only in highly vulnera-
ble countries. A de-escalation happens when one conºict party is adversely
affected by the disaster and the other conºict party cannot exploit this vulnera-
bility, because it is either weak in general or temporarily affected by the disas-
ter as well. This outcome provides qualiªed support for the power differential
meccanismo (and H3): armed conºicts de-escalate when both parties suffer
signiªcant adverse impact from disasters. But the more common de-escalation
scenario is when just one party (usually the rebels) is affected by a disaster
and its opponent (usually the government) is too weak to capitalize on
this opportunity.

My ªndings make four contributions to the wider literature on climate
change, disasters, and armed conºict. Primo, structural factors like ethnic exclu-
sion and agricultural dependence feature prominently in existing studies on
conºict onset and incidence. My results indicate that ethnic exclusion is less
relevant for disaster-related changes in conºict intensity. While acknowledg-
ing the relevance of poverty and agricultural dependence as necessary condi-
tions for these shifts, this article highlights the importance of more dynamic
and local conditions. In particular, I show how disasters can shape armed
conºict intensity by (temporarily) changing the distribution of relative power
between the government and rebels (power differential mechanism). Future
research should explore how climatic changes, disasters, and environmental
stress in general inºuence power relations between conºict parties.99

Secondo, in contrast to mainstream debates about climate change, disasters,
and increased conºict risks, this analysis shows that climate-related disas-

99. Scholars of civil war, by contrast, have written extensively on how power relations shape mili-
tary and bargaining actions before, during, and after armed conºicts. See Blattman and Miguel,
“Civil War”; Cederman and Vogt, “Dynamics and Logics of Civil War.”

l