Lang : English
  • English
  • Français
  • हिन्दी
  • Português
  • Español
close× Call Us

Brief 08: Hearts and Minds in Afghanistan

The researchers of this study have found that while Afghani individuals harmed by the coalition soldiers are far more likely to support the Taliban, those harmed by the Taliban are not more likely to support coalition forces.

Link to Full Study

Category: Conflict and Violence

Date of Publication: Monday, August 18, 2014

EGAP Researcher: Jason Lyall, Graeme Blair, Kosuke Imai

PDF: PDF icon Brief08HeartsandMindsAfghanistan.pdf

Click to Download the Data

Geographical Region: Middle East

Research Question:

The explicit goal of many counterinsurgency programs, such as those of coalition (ISAF) forces in Afghanistan, is to win the “hearts and minds” of the people.  But how can combatants achieve these goals?

Preparer: Kolby Hanson



More than a decade into the war in Afghanistan, the Taliban insurgency still enjoys strong popular support in many areas of the South and East.  The researchers conducted surveys in five provinces in those regions (Helmand, Khost, Kunar, Logar, and Urozgan) to understand the support better.

Measuring civilian support poses a number of methodological problems.  Civilians may be dishonest, perhaps fearing that their safety or wellbeing might depend on giving a certain answer.  In addition, official reports of violence and aid may be unreliable and biased due to corruption and poor monitoring, meaning that determining the effectiveness of these programs could be very difficult.


Research Design:

In order to elicit honest levels of support, the researchers used a series of four endorsement experiments.  Rather than asking civilians whether they supported the Taliban or ISAF, the survey asked whether they support four policies, which they prefaced with “The Taliban supports,” “The ISAF supports,” or “Some people support.”  (The policies are in fact all endorsed by both sides.)  This does not measure the individual’s support for each side, but provides an estimate of support for a group within a region’s population.

The survey also asked individuals what types of Taliban and ISAF actions they have been exposed to. For instance:  Have they or their communities received aid from either side?  Have they, members of their families, or members of their communities been victims of violence by the ISAF or by the Taliban?  Were they approached afterwards in an attempt to make amends?  Using these questions, the researchers identify what actions appear to increase or decrease support for one side or another without relying on dubious official statistics.


As we might expect, civilians who have been harmed by ISAF soldiers (or whose family or friends have been harmed) are far more likely to support the Taliban and less likely to support the ISAF than those who have not.  However, those harmed by the Taliban are only slightly more likely to support the ISAF and less likely to support the Taliban.

Meanwhile, aid-related programs designed to court civilian support appear to be less effective.  Afghans who have received ISAF aid or who have interacted with ISAF soldiers are no more likely to support the coalition and those who have received Taliban services or aid are no more likely to support the Taliban.  However, harmed individuals who have been approached by ISAF soldiers (offering a formal apology and financial restitution) are much more likely to support the ISAF over the Taliban – nearly as much as those who have not been harmed in the first place.

The study’s logic, however, might be reversed.  It could be that the Taliban use violence against different civilian populations than coalition forces do.  For example, the Taliban might use it to reward supporters while the coalition might use it to court pro-Taliban supporters.  This is particularly true with the restitution program, which targets only a small number of individuals, who might be much more or less likely to support the coalition regardless of suffering or restitution.  Similarly, it could be that civilians attribute blame for an incident to the group they oppose.  Although the study’s method cannot tell us for sure, the authors do provide substantial evidence that these effects are not taking place.

Policy Implications:
  • The study bolsters one of the central principles of counterinsurgency – that civilian casualties dramatically undermine efforts to win civilian hearts and minds, at least for coalition forces.  The surveys also find that ISAF violent event reports often do not align with civilian reports of violence, suggesting that civilian violence might be more widespread and damaging than the ISAF believes.
  • The study also calls into question whether aid programs actually win civilian support.  Civilians who have received aid are no more likely to support ISAF forces than those who have not.
  • The study also suggests that personal interaction, and especially apologizing in person for civilian casualties, might be a good method of reducing the impact of civilian casualties.  Although it is a seldom-used tactic (only 16% of harmed individuals report being approached by coalition soldiers afterwards), it does appear to be a way to regain civilian support in the wake of violent events.