AID , POWER , AND GRIEVANCES : LESSONS FOR WAR AND PEACE FROM RURAL AFGHANISTAN

Recent studies present contrasting findings on how reconstruction and development aid affects security in wartime contexts. Some research has found that aid projects decrease violent incidences, while other work has found both no effect or even evidence of a positive relationship. Addressing this mixed empirical picture, this article examines the complex intra-communal dynamics spurred by the distribution of aid in rural Afghanistan. Drawing on original interviews conducted in a community of Marjah district, Helmand province, the analysis indicates that development aid helps to elevate previously relatively less powerful individuals into positions of community leadership. This newly generated class of local leadership subsequently develops relationships to the community that differ from their predecessors since their social position is rooted in new sources of power. As a result, intra-community tension increases. These findings help to specify the conditions under which the delivery of aid may not help to win “hearts and minds” of locals and may potentially promote conflict. In addition, the analysis underscores how consideration of antecedent social conditions and temporal processes can help to refine our understanding of the wartime relationship between aid and security. Two of this century’s most costly and long-lasting wars, NATO’s intervention in Afghanistan and the American-led invasion of Iraq, have interwoven development with war-making, in part to win over the “hearts and minds” of the affected populace and help forestall insurgencies. This has given rise to novel instruments of wartime development such as military-civilian Provincial Reconstruction Teams, as well as massive budgets. After more than a decade of these programs, researchers have begun to examine their efficacy. The empirical results are inconclusive. For example, a widely cited paper on aid and conflict in Iraq found that the Commander’s Emergency Response Program (CERP), which enabled U.S. military officers in the field to provide local-level public goods, reduced levels of violence there. Afghanistan’s National Solidarity Program, a government-run endeavor that distributes grants to Community Development Councils, likewise was found to decrease violence, but only in areas where levels of violence were already relatively low. In contrast, one analysis of CERP in Afghanistan found a large inverse association of aid with violence, but reports that the finding nonetheless was not statistically significant. A recent re-analysis of Afghan data, over a longer time period and broader geographic area, concludes that CERP aid had no effect on violence at all, positive or negative. And fieldwork in Afghanistan suggests that development aid has generated social tension and conflict. These mixed findings cast doubt on current arguments that link the delivery of aid, winning hearts and minds, and the subsequent reduction of insurgent activity. In all, the evidence from Afghanistan suggests that development projects do not increase the benefit of participating in the formal labor market, which would in turn reduce the payoff of participating in rebellion. Moreover, the conditional implementation of these projects does not seem to incentivize the cooperation of the populace, a dynamic referred to in the literature as the information-sharing theory. To address the mixed empirical picture of development efforts and (violent) conflict, this article examines the effect of outsider-provided aid on a rural Afghan community at the local level. The focus is not on patterns of violence but on the complex intra-community dynamics unleashed by foreign actors’ development activities. The ultimate goal of analyzing what happens as a result of development aid, and why, is to take steps toward constructing new theory that identifies the mechanisms by which reconstruction and development work affects insurgency—beyond rates of violence. Drawing on original interviews with residents of one southern Afghan community, I find that foreigners’ aid often flowed to community members with limited pre-aid access to communally accepted sources of power. This occurred for two related reasons. First, rather than the community’s previously powerful residents, these individuals were involved in advising THE ECONOMICS OF PEACE AND SECURITY JOURNAL KARELL, Rural Afghanistan p. 44 Vol. 10, No. 2 (2015) | doi:10.15355/epsj.10.2.43 The Economics of Peace and Security Journal — ISSN 1749-852X — http://www.epsjournal.org.uk © EPS Publishing, 2015. All rights reserved. For permissions, email: ManagingEditor@epsjournal.org.uk outsiders during the planning and design stages of aid projects and, second, they then gained access to resources through the projects’ implementation stages when outsiders collaborated with local vendors. Subsequently, the aid helped to elevate these residents into positions of leadership, thereby generating a new cadre of local leadership. This new leadership relates differently to their communal neighbors than do their predecessors, primarily because both their initial empowerment and their continued support relies more on external material resources than had been the case before. As a result, intra-community tension increased. That outsiders collaborate with leaders of peripheral communities, and, through the provision of resources such as aid, affect how these leaders interact with their communities is not a new phenomenon, of course. For example, during the British conquest of southern and eastern Africa, their military and commercial leaders manipulated and allied with various local elites. During the Afghan jihad against the Soviets, some Afghan fighters used external resources to become warlords. More recently, local Afghan commanders were charging NATO forces USD10,000 per month for the collaboration of their militias. In various African states, foreign aid since the 1970s has induced elites and potential civil society actors to be more attentive to donors and to pursue rent-seeking rather than to be responsive to the concerns of their communities. Building on these experiences and findings, this article provides a novel formulation of how complex, local-level resource-based relationships between outsiders and peripheral communities, as well as among the community members themselves, connect to the general dynamics of aid provision, wartime development, and (violent) conflict. It does so by taking the rare step of examining the heterogeneity in power and influence within local peripheral groups and tracing how outside forces can first generate local elites and then come to believe that these actors are indeed the relevant indigenous local elites to work with. This social construction of new elites does not necessarily generate less responsive community leaders, I argue, but rather reshapes the community’s social order, thereby affecting in ways in which leaders are attentive to their neighbors. These new manners of interaction between leaders and residents can, in some contexts, induce grievances within the community and increase the likelihood of conflict at the local, rather than regional or national, level. Data and context: The case of Marjah The analysis draws on 15 in-depth, semi-structured interviews conducted in one community in the Marjah district of Helmand province in November and December 2014. Due to constraints on discussing sensitive matters, such as the sources of power and the allocation of resources, the respondents comprise a convenience sample of the community. To identify potential respondents, the community’s leaders were first contacted and asked to consent to the administration of the study. To minimize disruption to the community, the leaders were then asked to nominate possible respondents. If these nominees agreed to participate, a private meeting was arranged to conduct the interview. While this type of sample has obvious weaknesses, it generated data on social processes and previously unidentified (possible) causal relationships that can facilitate the conceptualization of new variables of interest, advancement in the design and implementation of future systematic studies, and the interpretation of existing findings. Concern regarding respondents’ possible biases and the validity of their responses is mitigated by three considerations. First, I am in fact primarily interested in respondents’ subjective understanding of what transpired in their community over the last several years. Personal perspectives on social dynamics carry crucial implications for various outcomes such as support for the central government and, as discussed later on, grievances and potential support for an insurgency. Second, the findings reveal that respondents were largely dissatisfied with the prevailing direction the community leadership took. This is opposite to biases that might be expected of individuals nominated by the leadership. Third, without being prompted, respondents independently recounted communal events in remarkably similar ways. For example, every respondent described a recent community-wide election and correctly reported the outcome. Additionally, two key respondents involved in a recent local scandal separately depicted the event in a similar manner. Such instances of overlapping narratives increase confidence that respondents are, first, aware of community events and issues, and, second, accurately report their understanding of the social dynamics to the researcher. Due to social restrictions, all interviewees were male. They ranged in age from about 20 to 60, with an average age of around 40. Most were farmers; one was a teacher and another was a police officer. Six used the honorific haji, taken after completing the hajj, which sometimes indicates that an To address the mixed empirical picture of the relation between economic development aid and violent conflict, this article examines the effect of outsider-provided aid on a rura

These mixed findings cast doubt on current arguments that link the delivery of aid, winning hearts and minds, and the subsequent reduction of insurgent activity.In all, the evidence from Afghanistan suggests that development projects do not increase the benefit of participating in the formal labor market, which would in turn reduce the payoff of participating in rebellion.Moreover, the conditional implementation of these projects does not seem to incentivize the cooperation of the populace, a dynamic referred to in the literature as the information-sharing theory. 2 To address the mixed empirical picture of development efforts and (violent) conflict, this article examines the effect of outsider-provided aid on a rural Afghan community at the local level.The focus is not on patterns of violence but on the complex intra-community dynamics unleashed by foreign actors' development activities.The ultimate goal of analyzing what happens as a result of development aid, and why, is to take steps toward constructing new theory that identifies the mechanisms by which reconstruction and development work affects insurgency-beyond rates of violence.
Drawing on original interviews with residents of one southern Afghan community, I find that foreigners' aid often flowed to community members with limited pre-aid access to communally accepted sources of power.This occurred for two related reasons.First, rather than the community's previously powerful residents, these individuals were involved in advising outsiders during the planning and design stages of aid projects and, second, they then gained access to resources through the projects' implementation stages when outsiders collaborated with local vendors.Subsequently, the aid helped to elevate these residents into positions of leadership, thereby generating a new cadre of local leadership.This new leadership relates differently to their communal neighbors than do their predecessors, primarily because both their initial empowerment and their continued support relies more on external material resources than had been the case before.As a result, intra-community tension increased.
That outsiders collaborate with leaders of peripheral communities, and, through the provision of resources such as aid, affect how these leaders interact with their communities is not a new phenomenon, of course.For example, during the British conquest of southern and eastern Africa, their military and commercial leaders manipulated and allied with various local elites.During the Afghan jihad against the Soviets, some Afghan fighters used external resources to become warlords.More recently, local Afghan commanders were charging NATO forces USD10,000 per month for the collaboration of their militias.In various African states, foreign aid since the 1970s has induced elites and potential civil society actors to be more attentive to donors and to pursue rent-seeking rather than to be responsive to the concerns of their communities. 3uilding on these experiences and findings, this article provides a novel formulation of how complex, local-level resource-based relationships between outsiders and peripheral communities, as well as among the community members themselves, connect to the general dynamics of aid provision, wartime development, and (violent) conflict.It does so by taking the rare step of examining the heterogeneity in power and influence within local peripheral groups and tracing how outside forces can first generate local elites and then come to believe that these actors are indeed the relevant indigenous local elites to work with.This social construction of new elites does not necessarily generate less responsive community leaders, I argue, but rather reshapes the community's social order, thereby affecting in ways in which leaders are attentive to their neighbors.These new manners of interaction between leaders and residents can, in some contexts, induce grievances within the community and increase the likelihood of conflict at the local, rather than regional or national, level.

Data and context: The case of Marjah
The analysis draws on 15 in-depth, semi-structured interviews conducted in one community in the Marjah district of Helmand province in November and December 2014.Due to constraints on discussing sensitive matters, such as the sources of power and the allocation of resources, the respondents comprise a convenience sample of the community.To identify potential respondents, the community's leaders were first contacted and asked to consent to the administration of the study.To minimize disruption to the community, the leaders were then asked to nominate possible respondents.If these nominees agreed to participate, a private meeting was arranged to conduct the interview.While this type of sample has obvious weaknesses, it generated data on social processes and previously unidentified (possible) causal relationships that can facilitate the conceptualization of new variables of interest, advancement in the design and implementation of future systematic studies, and the interpretation of existing findings. 4oncern regarding respondents' possible biases and the validity of their responses is mitigated by three considerations.First, I am in fact primarily interested in respondents' subjective understanding of what transpired in their community over the last several years.Personal perspectives on social dynamics carry crucial implications for various outcomes such as support for the central government and, as discussed later on, grievances and potential support for an insurgency.Second, the findings reveal that respondents were largely dissatisfied with the prevailing direction the community leadership took.This is opposite to biases that might be expected of individuals nominated by the leadership.Third, without being prompted, respondents independently recounted communal events in remarkably similar ways.For example, every respondent described a recent community-wide election and correctly reported the outcome.Additionally, two key respondents involved in a recent local scandal separately depicted the event in a similar manner.Such instances of overlapping narratives increase confidence that respondents are, first, aware of community events and issues, and, second, accurately report their understanding of the social dynamics to the researcher. 5ue to social restrictions, all interviewees were male.They ranged in age from about 20 to 60, with an average age of around 40.Most were farmers; one was a teacher and another was a police officer.Six used the honorific haji, taken after completing the hajj, which sometimes indicates that an To address the mixed empirical picture of the relation between economic development aid and violent conflict, this article examines the effect of outsider-provided aid on a rural Afghan community at the local level.Focusing on the complex intra-community dynamics unleashed by foreign actors' development activities, I find that foreigners' aid often flowed to community members with limited pre-aid access to communally accepted sources of power.Thus empowered, new community leadership emerged and intra-community tensions increased.
individual possesses a certain amount of resources, enabling him or her to complete the hajj, and often garners one greater respect in one's community.All interviewees were residents of one community in Marjah.Situated in central Helmand, Marjah was made a district in 2011/12 and had a recorded population of 27,000 in 2014.Once a sparsely populated area of clay desert, Marjah became more densely populated in the wake of the development of major irrigation infrastructure, such as the damming of the Helmand river and the construction of a series of canals, as well as government land (re)distribution schemes from the 1950s through the 1970s. 6arjah's development into an agricultural region produced a complex and distinct social context.Most significantly, a tribally and ethnically heterogeneous community resulted from the arrival of several waves of settlers.Specific details of how this community experienced the 2001-2014 war are difficult to come by.For example, the Afghan Country Stability Picture (ACSP), a database on reconstruction and development projects funded by the U.S. military, USAID, Afghanistan's Ministry of Rural Rehabilitation and Development (MRRD), and other organizations from 2002 to 2009, records 54 projects in Marjah.Of these, all but one were funded by organizations labeled as "other," and only three were recorded with a project start or end date.Alternative databases of aid projects, such as those provided by the MRRD or the U.S. Special Inspector General for Afghanistan Reconstruction (SIGAR), only contain data aggregated at the national or provincial levels.An exception is a database from the Combined Information Data Network Exchange (CIDNE), which records 62 planned CERP projects completed in 2010, totaling USD2,778,267.The numbers suggest that Marjah received more CERP aid than did other districts in Helmand, primarily in regard to cost per project (see Table 1). 7hile CERP projects were only one type of extra-local aid provided across Afghanistan, they have been of particular interest to scholars of aid and insurgency.Explicitly designed to enable external actors, in this case military commanders, to engage relatively rapidly in modest-sized projects tailored to local needs, and thereby to emphasize collaboration between outsiders and locals, it was at the same time hoped to decrease the appeal of insurgency.Specifically, the projects' effects on intra-community dynamics of interest here flowed through a range of interactions: Military units' identification of projects in collaboration with local government officials and community leaders (sometimes through shuras, or gatherings with civic notables); implementation of capacity-building workshops; competitive tendering processes with local actors; payment disbursement in installments; site visits for quality control; ceremonies at project completions; subcontracting by Afghan partners; and the influence of the financial capital by itself.A rival program, the National Solidarity Program, also emphasized links between external actors and local communities, albeit with less direct contact.Compelled to elect Community Development Councils, communities then planned projects and applied for grants from the government. 8hile Marjah's distinctiveness cautions against applying the findings to Afghanistan as a whole, the district's economic and social conditions resemble those found in many rural conflict zones, as well as in most of Helmand-a pivotal province during the NATO war and a region of continued importance for the opium trade and stabilization efforts.For example, most of Marjah's residents, as Helmand's, engage in the agriculture economy.Additionally, the community's social heterogeneity mirrors the tribal fragmentation depicted in ethnographic accounts elsewhere in the province.Furthermore, over the course of the war, Marjah experienced levels of violence comparable to other districts in the province (Table 1).Its levels of CERP aid are greater than the median, but in a case study, high values of the primary independent variable of interest are useful for generating insights into underspecified processes and causal relationships.In turn, these can be used to construct variables and hypotheses in subsequent studies. 9

Findings
Other than preceding instability, many recent economics and political science studies of aid and conflict in Afghanistan overlook antecedent social conditions.In contrast, the analysis that follows shows that pre-existing social fundamentals and expressions of power play an important role in how aid delivery affects a community.In particular, the communal allocation of power among residents interacts with the methods of aid distribution, shifts sources of power within the community, and thereby changes who holds power and how the newly powerful use their newfound social positions.

Antecedent conditions of power
Interviewees recollected that before the influx of aid, power was typically based on familial lineage and inherited material and social resources.Not uncommon in rural Afghanistan, such sources of local power appear alongside positive reputation, reliable provision of goods and services to neighbors, and ability to protect the interests of the local community from the state and other interest groups-although, of course, these complementary sources of power are more easily achieved with material wealth, social rank based on lineage, and kinship networks. 10hen outside actors distributed aid in Afghanistan, such antecedent conditions sometimes benefitted powerful individuals.Primarily these individuals gained from outsiders' relatively common practice of consulting with local leaders to identify and plan projects.Consultation was especially likely if powerful individuals had used their sources of power to gain government positions or to establish close ties with officials.Moreover, during the implementation phase of projects, when outside actors contracted with local vendors and disbursed funds in stages, powerful actors sometimes were able to use the advisory positions to steer aid to their own networks. 11n Marjah, however, pre-existing sources of power curbed powerful individuals.The interviews suggest three reasons for this.First, previous power holders often were targeted and killed by the Taliban or other armed groups, which sometimes saw members of leadership families as obstacles to their own goals.Nearly all respondents, for example, noted that in the mid-2000s, the Taliban had assassinated the most respected leader of their community.Second, access to existing sources of power appeared to weaken some powerful individuals' incentives to take the risk involved in accepting external aid since cooperating with outsiders, such as monetary payments for conducting a job, could potentially cause tension with neighbors and result in less respect for a power holder's existing sources of power.For example, one respondent recounted how in 2010 foreign forces "were pointing to some fields [of opium] and asking residents to go burn them, saying 'we will give you many dollars in return.'That person [who burned the fields] did not care about the owners of those fields nor whether they agreed to it or not.They were just doing that work and were getting money from [doing] it." 12early, the community's established leaders would find it challenging to engage in such work: Doing so would have risked undercutting their own carefully cultivated positions.Thus, the process of aid distribution increased the possibility that powerful individuals might not gain from the actual provision of aid.Recall that outsiders often consulted with community leaders during planning and design stages but implemented projects with different local actors.In the case of CERP, for example, non-commanders, such as contracting officers and paying agents, dealt with vendors.This practice created an opportunity for community leaders to be left out of the actual acquisition of outside resources, even if they had chosen to take the risk of accessing the aid. 13inally, pre-existing sources of power were unfamiliar to foreigners.Thus, resources were channeled to locals whom the foreigners could best, or most easily, understand and relate to.Those locals were not always the previously widely-acknowledged community leaders.That outsiders might misidentify community leaders in rural Afghanistan is not new.In the 1990s, after a U.N. directive had been issued specifying that aid would be delivered through shuras, local militia commanders staged such assemblies to portray themselves as community leaders.In some instances outsiders purposefully misidentified leaders if it served their interest to favor one faction over another.In Marjah, interviewees perceived that foreigners typically neglected to work with the respected leaders of their community, especially during the advisory stage of project design, and instead collaborated with residents who "introduced themselves as elders of the village to the foreigners," such as officers of the Afghan Local Police (ALP).Indeed, many such residents were previously farmers and far removed from established leadership roles.As one interviewee put it, "the unimaginable became reality because a person who could not handle a farmer's responsibilities became [a member of the] ALP." 14 Antecedent conditions of power, combined with the practices of aid distribution, thus influenced who received aid.Residents of Marjah holding social positions based on established sources of power, such as lineage, were less likely to access aid.In contrast, residents without ties to the established sources of power were more likely both to work with foreigners as well as to hold positions that were more familiar to foreigners.Consequently, a cohort of relatively less powerful residents gained access to externally provided aid.

Gaining and leveraging of resources
Once relatively less powerful residents had gained access to resources from foreigners they worked to increase and consolidate their newfound social positions.The most successful of the newly empowered residents did so by nominating themselves during district council elections.All interviewees described the recent elections in similar terms: Two residents who had become wealthier through ties with foreigners nominated themselves, arguing that this indicated that they were able to continue bringing external aid to the community.To some residents, these attributes were vital skills for community representatives. 15hese two men won the elections, even as this outcome proved unpopular among the interviewed respondents.One interviewee explained the result by saying, "most of our people are illiterate, so most of them accepted [the victorious candidates'] lies and trusted them." 16Another emphasized how the victors' source of power decreased their legitimacy: "[The election winners] still do not deserve to be the people's representatives because they were peasants and farmers previously.They became the elders because of the assistance and money of the foreigners, and they got rich because of the foreigners." 17 course, as the newly empowered residents used aid resources to strengthen their power-for example, by being elected to the district council or entrenching themselves further in the ALP-they appeared even more like legitimate leaders to the foreigner actors providing the aid.This generated a reinforcing process in which they received even more aid over time.As one respondent explained: "I can tell you that there is a factory producing elders, and every person tries to become an elder to earn money.They get the support of a few people and then present themselves to the foreigners as an elder, which then gets them projects from the foreigners." 18 other words, foreign aid helped to empower a group of once relatively less powerful residents, enabling them to occupy leadership positions.It did so by first providing them with resources and then enabling them to portray themselves as able to gain resources from outside the community, akin to community patrons.This resulted in a self-fulfilling prophecy: After gaining and consolidating the reputation, or even formalized positions, of community leadership, aid continued to flow to these residents precisely because foreigners believed them to be community's leaders. 19

Emergent conditions of power and grievances
The generation of a new group of elites need not, in itself, be detrimental to a community or an obstacle to stability.In the community under study, however, the process produced social consequences that respondents widely saw as negative.The most significant consequence, besides the concentration of aid resources in the hands of a few, was the new elites' handling of intra-community dispute resolution. 20early all respondents explained how community members historically relied on leaders to resolve especially contentious interpersonal disputes.This practice, in their view, had recently become unjust and less reliable with the new wave of leaders.Relative to the preceding leaders, the ascendants-with power rooted in foreign aid rather than in inherited deference and respect of their neighbors-showed little interest in dispute resolution.Because they did not rely on prestige and status gained from resolving intra-community strife, the emerging leaders no longer engaged in civic matters as their predecessors did.
Moreover, when the new leaders did become involved in dispute resolution, most respondents noted that they imposed fees.For example, one interviewee said that the current elders "... solve only disputes that will benefit them and in which they have some personal interest.They do not participate in the meetings (shuras) of the poor and the weak.If they do, they will not work honestly to solve the dispute [in shuras meant to resolve disputes].[But the worst] thing is that elders are now asking for money to solve the disputes.Previous elders would not take money but the present elders receive money as if they are working on commission." 21other interviewee said, "In regards to solving disputes, a very big difference has occurred [since the introduction of foreign aid].Previously [leaders] were solving disputes for the sake of God but presently ... the decisionmakers ... are accustomed to having more money and the elder takes some money when solving disputes." 22e result of the shift in the sources of power, from lineage and inheritance to the ability to acquire material resources, is that the current leaders do not see much value in previous social practices because the old sources of power do not support their current social positions.Furthermore, when they do engage in these social practices, their behavior differs from their predecessors' so as to expand their current source of power, the acquisition of wealth.
All respondents conveyed dissatisfaction with the perceived shift in leaders' behavior.In the majority of interviews, respondents repeatedly lamented how "honest elders have decreased enormously," and several respondents linked the change in their community's leadership to broader social issues in a negative manner.For example, one interviewee observed, "present elders do not think of the m o t h e r l a n d , c o u n t r y , a n d people-they only think of their own pockets [to fill]."Another interviewee explained: "Foreign aid has impacted the traditional justice system by appointing or creating careless and dishonest people as tribal elders and heads of villages.Instead of solving people's problems, these elders instead only see to their own pocket [to enrich themselves] and how to take away money from others.This is the reason that people no longer believe in the elders ... Previously, poor and rich people were equal to the elders, but now it is not like that.Elders now favor the rich people (maleks) [in dispute resolution] even if the poor people are not at fault at all.[The elders] put all the faults and mistakes on the poor people [to ensure the payment of greater fees from the rich]." 23e process is summarized in Figure 1.First, antecedent conditions that create, structure, and distribute power within a community.The introduction of outside aid then generates new potential sources of power.In one case, depicted in the upper path of the figure, previously powerful individuals are able to participate in one of these potential sources: They advise outsiders on projects.If they are also able to subsequently control the implementation of projects, the pre-existing sources of power will likely be reinforced.
In contrast, three other scenarios are possible, each resulting in previously less powerful community residents gaining power.These scenarios are depicted in the lower three paths of Figures 1.In the first scenario, previously powerful individuals advise on projects but are not able to control the implementation.In the second, it is the previously less powerful individuals who advise outsiders on projects.In this scenario, previously less powerful individuals have accessed the first new source of power-advising on projects-possibly because their social positions are more familiar to foreigners.In the third scenario, previously less powerful individual access a second new source power-the implementation of projects-by directly co-opting this process.In the case of Marjah, the second and third scenarios occurred.
After previously less powerful individuals gained access to new sources of power, acquired resources, and consolidated their power, outsiders perceived them as legitimate leaders even though it was the intervention of outsiders that helped to elevate these residents in the first place.Aid thus continued to flow to these new leaders.This is depicted in the lower-right of Figure 1 with the arrow pointed to the left.
In these latter scenarios, the conditions of power changed: The new leaders' power is based on accessing externallysourced aid.At this point, respondents reported that the newly formed elites approached typical leadership roles differently than did their predecessors.The most frequently mentioned difference regarded how leaders handled intra-community dispute resolution.The new leaders were said to be less engaged and, when they were engaged, they were imposing fees for what previous leaders had treated as a community service.Rather than villagers being dependent on local elites for protection and public goods, aid fostered elites' dependency on external sources of revenue and perhaps even the resources of villagers. 24

Discussion and conclusion
With an eye toward possible effects on peace and security, this article examines how reconstruction and development efforts in Afghanistan affected the internal social dynamics of rural communities.Drawing on original interviews conducted in one  community in central Helmand's Marjah district, potentially generalizable processes can be theorized to create testable hypotheses.Indeed, the broader goal of this article is to take initial steps toward developing new theoretical and empirical approaches to studying the relation between aid and security in wartime contexts.
The analysis highlights three stages of a process that unfolded in the community under study.First, aid flowed to specific residents depending on their pre-existing access to established sources of power and on aid distribution procedures.Specifically, residents with limited connections to locally distinct sources of power typically accessed externally provided aid to a greater degree than their more powerful neighbors.Second, this aid then helped to elevate residents into leadership positions.The emergent elites were subsequently seen by foreigners as the true community leaders and, as a result, even more aid flowed to them and their growing networks.Third, this outcome increased social tension and grievances in the community.Of note is that the new elites' approach to typical leadership roles varied from their predecessors', most likely because their power was rooted in different sources.
Specifying the social process makes several contributions to the study of aid and security in Afghanistan and beyond.Regarding Afghanistan, it theorizes social dynamics that have remained implicit in existing qualitative research which finds that development efforts increase instability.In addition, the analysis illuminates the differential effects that externally provided aid can have across levels of social organization.For instance, the reforms and programs during the last decade have led to, among other things, the founding of provincial councils.These councils engendered increased trust between local communities and the government, primarily because local residents were selected to serve in these governmental bodies.In contrast, my research suggests that external aid, programs, and reforms aimed at the local level can carry adverse effects. 25eyond Afghanistan, I take a constructionist perspective on how elite resource capture can contribute to violence.Rather than taking powerful actors as given, it accounts for the generation of such actors as well as for their role in promoting grievances and, potentially, conflict.Indeed, while instability is not the primary outcome of interest, the analysis links events in contemporary rural Afghanistan to recent findings on conflict between the central state and peripheral groups by developing an empirically based, potentially generalizable depiction of how centrally-condoned aid can generate grievances in a peripheral community.Grievances, in turn, are increasingly (re)credited with driving conflict over the last two and a half centuries.In brief, the relevant literature finds that when members of peripheral communities view themselves as excluded from a central, governing coalition, or when they perceive this coalition as illegitimate, they are more likely to organize along nonstate networks and to engage in antistate activities such as ethnic group-based civil war.Conflict has often resulted when centrally supported elites establish the political exclusion of peripheral communities.Returning to the case of Marjah, this article indicates that residents who did not access aid, as well as the preceding, and now declining, community leadership, both express grievances akin to political exclusion: They feel left out of the emergent elites' relationship with centrally-condoned aid. 26t the same time, the article indicates that the relationship between aid and security may be more complex than typically conceptualized.Aid can have a multi-directional effect on violence.By supporting less powerful individuals, aid may have successfully secured the compliance of residents who otherwise would have participated in potentially violent, but not antistate, activity, such as drug smuggling.As a result, one type of violence may have decreased while concurrently provoking an insurgent, antistate response.Alternatively, aid may have empowered residents who, through their newfound leadership positions, are able to become involved in drug smuggling and related violence, again increasing one type of violence while decreasing the populace's cooperation with insurgents and that type of violence.In a word, aid may enable some residents to become more violent while decreasing the violence committed by other residents. 27he article leads to several policy implications regarding the delivery of aid in contexts like contemporary rural Afghanistan.Most generally, expanding the reach of a central state by winning the "hearts and minds" of local elites may undermine local government: External support for local elites can inadvertently delegitimize them among their community members.In addition, if aid providers aim to deliver resources conditionally, careful consideration should be given to what type of behavioral return to make aid conditional on, as well as on whom to impose conditionality.For example, the analysis suggests that, first, the behavior of residents induced through the provision of aid increased intra-community tension and, second, that aid providers were imposing conditionality on actors they themselves elevated to positions of influence.A third policy implication regards the social level at which aid should be directed: Reforms at the provincial level (e.g., provincial councils), may provide the necessary independence between local communities and the state, whereas interventions at more local levels may detrimentally disrupt foundational intra-community relations.After all, Afghanistan was most peaceful during the Musahiban dynasty (1929-1973) when the 15.The ability to deliver outside resources to a community has been a relatively common source of power across Afghanistan, evoking the roles of an arbab, although, as Barfield (2013)  notes, in the past, these representatives of villages to the "outside" were often appointed by the government, corrupt, and widely seen as a necessary evil.19.For a specific biographical account of foreign aid generating a new elite and an affiliated patronage network in another southern Afghan province, see Gopal (2014).20.Concentration of aid resources: According to one respondent, "foreign aid was good for some [residents] ... [But]  the foreigners thought that the community and villages were just following a few people and they were mistaken.The villagers were not happy about the way the projects were given to the same people every time" (Respondent 1, 40 years old; 28 November).
21. Respondent 10 (25 years old; 2 December).central state intervened only minimally in local affairs, and otherwise largely remained "over the horizon." 28inally, the article addresses the study of aid and security in general.First, the processual model presented here can be refined and assessed for generalizability through research in other Afghan communities as well as in comparable contexts elsewhere.More directly applicable to empirical approaches, however, are the specific insights provided by the analysis.For example, it suggests that models of interactions between incumbents, populace, and insurgents may be improved by not assuming homogeneity in power, identity, and solidarity among and within the populace and perhaps endogenizing the population's formation of social boundaries and elites.In addition, future analyses of insurgency may better capture real-world processes by considering factors such as sources of power, the duration of time that leaders have been in power, leaders' background, residents' satisfaction with community leaders rather than with the state, and the effect of introducing and then withdrawing aid.More broadly, the analysis emphasizes how antecedent social conditions and time can shape the outcomes of interest.After all, it is in the nature of social processes that they produce different consequences as they unfold.

Figure 1 :
Figure 1: Changes in power and aid in a rural Afghan community.

Notes
I thank Hannah Brückner, Travers Child, John O'Brien, Amiri Rahmatullah, and anonymous reviewers who helped to improve the final version of this article.All remaining errors are mine.The article was made possible by research support provided by the Division of Social Science, New York University Abu Dhabi.

Table 1 : Summary statistics
Notes: Only districts with at least one casualty per year are included in the tally of casualties across Helmand, 2003-13.Sources: CIDNE; Global Terrorism Database.