The Effect of Civilian Casualties on Wartime Informing: Evidence from the Iraq War

• Author: Andrew Shaver,Jacob N. Shapiro
• Date: 01 August 2021
• DOI: 10.1177/0022002721991627
• Published date: 01 August 2021
• Subject Matter: Articles

Article

The Effect of Civilian

Casualties on Wartime

Informing: Evidence

from the Iraq War

Andrew Shaver

1

and Jacob N. Shapiro

2

Abstract

Scholars of civil war and insurgency have long posited that insurgent organizations

and their state enemies incur costs for the collateral damage they cause. We provide

the first direct quantitative evidence that wartime informing to counterinsurgent

forces is affected by civilian victimization. Using newly declassified data on tip flow to

Coalition forces in Iraq we find that information flow goes down after government

forces inadvertently kill civilians and it goes up when insurgents do so. These results

confirm a relationship long posited in the theoretical literature on insurgency but

never directly observed, have strong policy implications, and are consistent with a

broad range of circumstantial evidence on the topic.

Keywords

asymmetric conflict, civilian casualties, conflict management, civil wars

The effect of civilian casualties on wartime informing to counterinsurgent forces is of

great interest to scholars of sub-state violence, many of whom have long highlighted

the importance of information in insurgency campaigns. Most twentieth-century

counterinsurgency theorists writing on the wars of decolonization argued that

1

Department of Political Science, University of California, Merced, CA, USA

2

Department of Politics, School of Public and International Affairs, Princeton University, NJ, USA

Corresponding Author:

Jacob N. Shapiro, Department of Politics, School of Public and International Affairs, Princeton University,

Princeton, NJ 08540, USA.

Email: jns@princeton.edu

Journal of Conflict Resolution

2021, Vol. 65(7-8) 1337-1377

ªThe Author(s) 2021

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0022002721991627

journals.sagepub.com/home/jcr

obtaining information on rebels from non-combatants was critical for government

forces and that protecting the population from insurgents was critical to gaining that

cooperation (Trinquier 2006; Galula 1964; Taber 1965; Clutterbuck 1966; Thomp-

son 1966; Kitson 1971). More recently, Kalyvas (2006) argued that indiscriminate

violence against civilians is counterproductive because it can turn civilians against

the party causing them harm.

1

Berman, Shapiro, and Felter (2011) explicitly model

the relationship between harm and informing as part of a three-sided game in which

civilians punish insurgents who create excessive costs by sharing information with

government forces. Within this literature scholars have long posited that insurgent

organizations and their state enemies incur costs for the collateral damage they

cause: insurgent harm produces more informing and harm by government forces

produces less.

That basic argument can be restated as a general theoretical question and a more

specific policy question. The general question is whether civilians strategically

decide whether to provide informa tion based on the behavior of insurg ents and

counterinsurgents. The specific policy question is whether civilian victimization

affects information flow.

Indirect empirical evidence of affir mative answers to both questions has been

observed in a number of conflicts. US government officials cited the potential for

civilian casualties to harm cooperation from civilians as one reason for imposing more

restrictive rules of engagement in Afghanistan in 2009. Condra and Shapiro (2012)

show that in Iraq insurgent violence went up after Coalition-caused civilian casualties

and down after insurgent-caused ones, consistent with an informational reaction to

abuse. Lyall, Blair, and Imai (2 013) show that self-reported vic timization by the

International Security Assistance Force (ISAF) correlates with lower support for

ISAF and higher support for the Taliban, and Lyall, Imai, and Shiraito (2013) show

that manipulating the ethnicity of harm described in vignettes can shift expressed

approval about informing. Shapiro and Weidmann (2015) provide evidence that

plausibly exogenous increases in cell phone coverages led to lower insurgent violence

during the Iraq war, which they attribute to greater information flow to counterinsur-

gents as the presence of mobile telecommunications lowered the risks of informing.

What is missing in these papers is direct evidence on information flow to counter-

insurgents. None of these papers directly measures rates of informing. As one recent

paper puts it, “despite its central role in civil war dynamics, the act of informing is

still poorly understood, due mostly to the classified nature of informant tips” (Lyall,

Imai, and Shiraito 2013).

Using newly declassified data on weekly province-level tips collected by Iraqi

and Coalition force during the Iraq war, we provide the first direct test of the

influence of civilian harm on wartime informing. Specifically, we test the effects

of government and insurgent civilian victimization on information flows to the

government. We cannot test the effects of civilian victimization on information

flows to the insurgents, a topic mentioned in both insurgent memoirs and interviews

1338 Journal of Conflict Resolution 65(7-8)

(Berman, Felter, and Shapiro 2018), because there are no data we are aware of which

measure tip flows to insurgents in Iraq.

Our data on information flow from the population to government forces span all

thirteen provinces which experienced substantial violence over a sixty-week period

from June 2007 to July 2008.

2

We combine the data on tips with administratively-

collected geolocated data on combat violence and press-based data on civilian

casualties collected by Iraq Body Count which has been used in a broad range of

previous work on Iraq (Alkhuzai et al. 2008; Boyle 2009; Hicks et al. 2009; Berman

et al. 2011; Hicks, Dardagan, Serd´an, et al. 2011; Condra and Shapiro 2012; Lewis

et al. 2012; Diwakar 2015). We buttress our quantitative analysis with anecdotes

from rich qualitative documentation on intelligence collection in Iraq obtained

through Freedom of Information Act (FOIA) requests.

Exploiting plausibly exogenous variation in civilian casualties occurring during

combat incidents due to the randomness inherent in weapons effects we find a robust

relationship between indiscriminate violence and informing.

3

In our baseline

unweighted model, an additional insurgent-caused civilian casualty leads to approx-

imately .2 more tips. In a weighted version of the same regression an additional

insurgent-caused civili an casualty leads to approxima tely .24 more tips, and an

additional Coalition-caused civilian casualty leads to approximately .64 fewer tips

in the next week. As we show in extensive robustness checks, other reasonable

specifications provide larger estimates. Controlling for a broad range of possible

confounders, such as longer term trends in combat, mostly strengthen the estimates.

These effects, while modest in magnitude, are substantively significant. In the med-

ian week in which insurgents caused civilian casualties, they killed four civilians,

predicting one additional tip to Coalition forces. That is a substantial number

(roughly 5%of the weekly mean) since single tips often resulted in raids that led

to the capture of both large numbers of weapons and prominent insurgents. These

results are partially consistent with prior work indicating that informing should be

more sensitive to government-caused civilian casualties (e.g., Condra and Shapiro

2012; Lyall, Blair, and Imai 2013): the drop in tip flows following a single

government-caused casualty is roughly much larger than the increase following an

insurgent one, but is only statistically significant in weighted regressions.

From a scientific standpoint, these results provide the first direct statistical evi-

dence for a relationship scholars and practitioners have posited for more than fifty

years. Our estimates reflect a causal effect to the extent that spikes in civilian

casualties from week-to-week are conditionally independent of the next week’s

trend in informing. We carry out a large series of robustne ss checks to address

identification concerns, showing that across a range of specifications the results are

remarkably stable. That stability lends credence to a causal interpretation. Specifi-

cally, we analyze the relationship between changes in civilian casualties and sub-

sequent changes in information sharing at the province-week level, conditioning out

all time-invariant province-specific factors to identify off short-run within-province

variance. Controlling for local trends in combat violence, past informing behavior,

Shaver and Shapiro 1339