City Research Online

Abstract

In recent years, donors and governments have allocated substantial amounts of aid to conflict-affected communities in the hope that some of it will "win the hearts and minds" of the local population and reduce violence. The logic is that aid can help to foster more positive attitudes toward the government by providing valuable goods and services. The belief is that, in turn, the local population will be less likely to support the insurgency and to instead engage in cooperative behaviours that could reduce violence. For example, communities may no longer be willing to provide food, shelter, and other forms of support to insurgents; members of the community may be less likely to participate in violent insurgency; and communities may be more willing to share information with the government regarding the activities and whereabouts of insurgent groups.

The previous research on aid and "winning hearts and minds" has primarily studied this relationship using violence-related outcomes. Scholars have subsequently attributed any observed reduction in violence to winning hearts and minds without providing compelling evidence for this explanation beyond its compatibility with the data. This research project has therefore sought to overcome this limitation by re-investigating the hearts and minds mechanism using survey data from the conflict in Afghanistan between 2008 and 2013 as a most-likely case. Afghanistan represents a "most-likely" case for examining the winning hearts and minds given the substantial financial resources that were allocated via Commander's Emergency Response Program (CERP) and the longevity of its programme, spanning 12 years (2004 to 2016). The empirical results from this research demonstrate that, contrary to widespread belief, district{level aid is negatively associated with individual{level perceptions of the national, provincial and local government. These results are robust to the inclusion of a number of relevant control variables, as well as alternative model specifications and variable measurements.

The findings of this dissertation is aligned with the research of a number of scholars who argue that aid not only fails to secure the "hearts and minds" of local populations but may, in fact, promote violent insurgency. This body of research point to a limited consensus in the previous literature for the effect that aid has on winning hearts and minds, showcasing the complex relationship between aid and attitudes. I contend that much of the existing research on aid and counterinsurgency has inferred the success or failure of the "hearts and minds" approach primarily through violence related outcomes, often interpreting reductions in conflict as evidence of improved civilian attitudes. However, such interpretations frequently rest on indirect inference rather than direct empirical validation. In response, this dissertation sought to address these limitations by re-examining the relationship between aid and local attitudes beyond violence-related outcomes alone, using the conflict in Afghanistan as a most-likely case through which to assess the broader validity of the "winning hearts and minds" hypothesis using attitudinal data from the Survey of the Afghan People (SoAP).

These results were obtained using a multilevel logistic regression, testing the effect of reconstruction aid on winning hearts and minds. Utilising three approaches to modelling, namely (1) a bivariate model between aid and attitudes toward government, (2) a model including the individual-level controls, and (3) a model which includes all individual- and district-level control variables, this research concluded that, across all three modelling approaches, the coefficient for aid is both negative and statistically significant. In other words, reconstruction aid is linked with a decrease in attitudes towards the government at the national, provincial and local levels. That is, individuals residing in districts with more aid projects are also less likely to report that the government is doing a good job. This is
a result that is the complete opposite of what we would have expected to observe should the hearts and minds mechanism hold true for the conflict in Afghanistan.

From this research, a clearer picture of the complexity of the case of Afghanistan emerges, one that shows the complex relationships between the presence of international aid and reconstruction projects, and their effect on the perception of government by the Afghan population. My empirical research has found evidence against the notion that aid can help to win over hearts and minds in the context of counterinsurgency in Afghanistan. This is a surprising result, as the mechanism itself has been a widely accepted and has informed attitudes across both academic and military fields. This includes the main guidance found in the "Commander's Guide to Money as a Weapons System", which describes how winning hearts and minds should become a military aim in order to facilitate defeating the insurgency.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HN Social history and conditions. Social problems. Social reform H Social Sciences > HT Communities. Classes. Races J Political Science > JA Political science (General) J Political Science > JC Political theory J Political Science > JV Colonies and colonization. Emigration and immigration. International migration J Political Science > JZ International relations U Military Science > U Military Science (General)
Departments:	School of Policy & Global Affairs > Department of International Politics School of Policy & Global Affairs > School of Policy & Global Affairs Doctoral Theses Doctoral Theses

Export

Downloads