Skip to content
Dilemmas facing agencies in the urban centres of Afghanistan

The issue comes more clearly to the fore when such power holders take over urban areas and therefore have responsibility for the administrative structures of the state within those areas. This article considers how this question has been played out in the context of Afghanistan.

The Afghan conflict has tended to be characterised as one in which people moved in their millions from the villages of Afghanistan across the border to the sanctuary of refugee camps and urban settlements in Pakistan and Iran. However, not all those who fled their villages in response to Soviet military action left the country. Many sought refuge in caves in the mountains and, in recent years, and certainly since the collapse of the Soviet-backed Government of Muhammed Najibullah in April 1992, many more made their way to the urban centres of Kabul, Mazar-i-Sharif, Herat, Kandahar and Jalalabad.

Each successive phase of the conflict has led to further displacement of the urban population, both within and between urban centres, with the result that one cannot talk in terms of a stable urban population. It is difficult to know whether the population of Kabul is nearer to half a million or a million at any one time although most of the estimates given are around the million mark. A study undertaken by ICRC in December 1996 found that a significant proportion of the population of Kabul had moved several times as a direct result of the conflict. Similarly, monitoring undertaken by UNHCR at the main entry points to the city found a high level of movement back and forth.

Foremost among those who have left are professionals working for the government administration. With such a significant outflow of professionals, agencies cannot plan on the assumption that those who are currently working in the government administration will necessarily remain, or even that their own staff will stay. Those who are left behind in Kabul are likely to be among the more impoverished because those with the means to leave for Pakistan, Iran or other parts of Afghanistan have already done so. This reality has combined with a deteriorating economic situation within the capital to produce a population close to destitution.

The dilemmas faced by agencies

In seeking to meet what are significant humanitarian needs, UN agencies, ICRC and NGOs have needed to determine appropriate criteria for their interventions and for their targeting of aid. Early interventions in the urban sector, such as those in Mazar-i-Sharif and Jalalabad, were clearly targeted at the internally displaced, in direct response to the outflows from Kabul following the major rocketing episodes. Such targeting, however, is not easily effected. In Mazar, for example, although a couple of small camps were established, most of those displaced disappeared into the wider population and could not easily be accessed.

Another dilemma arose for agencies when 200,000 people were displaced from the Shomali Valley to Kabul in early 1997. Humanitarian agencies were reluctant to make provision for a population which had apparently been forcibly removed from their settlements as an act of war, lest this encourage further such action, and agreement was reached that ICRC would simply provide temporary shelter for those who could not find accommodation with relatives, in the hope that people would quickly find their own solutions and return to the Shomali Valley as soon as possible. This policy proved to be reasonably effective and, by the late summer of 1997, about half of those displaced had returned. There remained the question, however, as to whether the remaining 100,000 or so should be specifically provided for and, if so, how, given that these were absorbed within the population. Such targeting was also difficult to justify when so much of the population was nearing destitution and when most had experienced displacement in one form or another.

Different agencies dealt with the issue of establishing criteria and targeting in various ways. ICRC, with its IDP-specific mandate, has taken the view that it should provide for the wider population, on the assumption that the displaced would also benefit (with the obvious exception of the temporary refuge referred to above). UNHCR, with its refugee-specific mandate, has sought to focus on the provision of support to displaced people returning primarily from Jalalabad to Kabul, to assist them in the repair of their damaged houses. The European Commission’s DG1 has opted to concentrate on the rehabilitation of the agricultural areas surrounding Kabul and other urban areas so as to strengthen the capacity of the agricultural hinterland to receive people returning from the cities.

Prior to the suspension of ECHO funding for Kabul in July 1998, most NGOs in Kabul had been funded by ECHO which had been providing between 10m and 35m ECU per year since 1995. For them, the question of whether people had suffered displacement had not been a major issue. More important had been the question of how best to provide for a population which was nearing destitution in an economic environment which offered very few opportunities and where those opportunities were seriously constrained by the policies of the power holders, such as the controls imposed on the employment of women.

A number of options have been explored including: large-scale relief programmes benefiting up to half a million people but for which the process of drawing up lists of beneficiaries has inevitably been problematic; food-for-work; and income-generating programmes for targeted individuals (difficult in a collapsed economy when there is little or no market for new produce). However, relief distributions have inevitably ended up being the primary solution, both in Kabul and other urban areas.

Dealing with the power holders: the provision of public services

Agencies have also had a responsibility to assess whether basic health, water supply, sanitation and education services are available to the urban population. This has required evaluation of the capacity of the governmental and municipal administration, and also of their willingness to commit resources to these services.

In the case of Afghanistan, a number of major dilemmas have arisen in this context since the advent of the Taliban. These arose primarily because the Taliban chose to reduce, as an instrument of policy, the services available to one section of the population – women. Women were denied access to education and faced obstacles in their efforts to obtain equal health care. Women were also not permitted to work outside the health sector and this has made it difficult for agencies to employ women in order to gain access to women for the provision of services. In addition, the Taliban were entirely focused on the war effort and wished to commit all their resources to this objective. Funding for public services was not, therefore, a priority.

Agencies therefore faced two further dilemmas: whether or not to accept responsibility for the provision of those public services which were permitted by the Taliban and also whether to quietly acquiesce at the denial of equal access to services or to advocate for a reversal of Taliban policy. They also had to consider to what extent their assumption of such responsibility was strengthening the power holders or reinforcing their policies. Among the questions they had to address were the following:

  • Would the power holders be forced to provide public services on a greater scale if the agencies withdrew? If this was unlikely, should the agencies take the view that humanitarian needs dictated that they remain, regardless?
  • Were they engaging in institution-building? How could this be defined? If an agency repaired and repainted a hospital damaged by conflict, was this institution-building or a necessary step to provide basic health care in the immediate future to a population at risk? If equipment was provided, did the same argument apply? Should agencies repair damaged pipes and water pumps? Were these essential to immediate public health needs? Should agencies work in cooperation with the Ministry of Public Health, the Municipality and the Water Supply Department, sitting on joint committees, in order to plan the most effective use of resources or should they operate parallel services? To what extent was there a relationship between the severity of need of the population and the question as to how far agencies should work in conjuction with the state administration? Was it easier to take a more principled position if the severity of need was less?
  • Were steps they took now going to make the situation worse in the future? If they supported spontaneous home schools for girls now, would this provide justification for future policy for girls to be educated at home? If they helped to repair and equip a hospital for women, would this later be used to require women to only attend this hospital?
  • Would advocacy to urge a change in policy, so that equal access to all services might be achieved, have any impact?


Questions such as these are likely to arise increasingly in situations such as Afghanistan where there is no national government and where blocks of territory are held by particular military elements. Agencies have no choice but to engage with these elements, if only to ensure the security of their staff and their operations. However, if a point is reached where a military force begins to regard itself as the administrative authority for the area under its control, the agency has to reappraise its situation. It will, no doubt, have to seek permission to continue to operate, involving complex interaction with the new authority. The point is reached where the agency has to decide whether or not to involve the authority in decisions relating to the provision of public services.

In a normal situation, where there is an internationally-recognised government, this would not be an issue. However, in a situation where there is no such government the agency has to make a very difficult decision as to whether, in its view, the new authority is demonstrating a reasonable level of responsibility in relation to the provision of services and should, therefore, be regarded as having a legitimate right to exercise control over the provision of those services.

In so far as it may be providing most of the resources to provide the services, the agency potentially has some leverage to negotiate an arrangement whereby services are provided on terms which it finds acceptable. However, with a power holder such as the Taliban, which does not attach priority to the provision of public services and, at the same time, feels very ambivalent about the presence of foreigners, there is little or no leverage. In Kabul, where funding on a large scale has been provided and where twenty or more agencies have been operating, the humanitarian assistance community has been able to exercise very limited leverage. In Herat, Kandahar and Jalalabad, where agencies are only able to ensure that health and public health services are provided at a very basic level, the leverage is even less,

The relationship is inevitably an awkward one. If the target population has clear humanitarian needs, the agency will probably deal with the power holder to the extent that its cooperation is necessary, or conducive, to the most effective use of resources. This is, however, a difficult balancing act: excessive cooperation may be regarded by the power holder as implicit recognition of its authority; minimal cooperation may undermine service provision and might also weaken what remains of a government administration.

A further issue for agencies with a developmental mandate, particularly those within the UN system, is that the governmental structure provides the natural counterparts for their interventions. This is less likely to be the case for agencies with a relief focus. These will look for whatever structures are likely to be most effective and, while these could include government departments or para-statal bodies such as the Red Crescent, it is likely that ICRC and NGOs, rather than government departments, will play the major role in implementing relief programmes.

One of the key arguments for working with and resourcing the administration in such situations is that, if a legitimate government did emerge, it would benefit from a functioning administration. Those espousing this view argue that if agencies chose to work in parallel with the administration rather than with it, government staff would almost certainly be attracted by the higher salaries and better operating conditions of the agencies and the administration would be further depleted.

The outcome in Afghanistan

In practice, in the urban areas of Afghanistan, both UN agencies and NGOs have worked closely with the Ministry of Public Health, the Municipalities and the Water Supply Departments. They have taken primary responsibility for the provision of health and public health services but have also provided some resources to the administrative structures of the state to enable them to function more effectively and to play a contributory role. Thus, for example, UNCHS (United Nations Centre for Human Settlements) has provided vehicles to the Municipality to facilitate waste clearance, NGOs have refurbished hospital buildings, and incentive payments have been made to health service workers to encourage them to give time to public services rather than work exclusively in private practice.

Agencies have thus worked in partnership with the administrative structures of the state, at times on the basis of contractual relationships in which both parties agreed to provide certain resources. However, there have been a couple of examples, of late, where the UN has entered into contracts with administrative structures of the state, or with NGOs nominated by the Taliban, to take full responsibility for the provision of particular services (such as vaccination programmes in particular regions or aspects of urban-based relief programmes). This new development in the relationship between the UN and the Taliban has encouraged the Taliban to regard NGOs as competitors for external resources and to think that, if NGOs were removed, the UN would enter into further contracts with the administrative structures of the state or Taliban-nominated NGOs. This would appear to have been one factor in the negotiations which took place between the Taliban and NGOs in July 1998, which resulted in the expulsion of NGOs from Kabul.

However, this ambivalence towards NGOs does not only relate to their position as potential competitors. Western NGOs have also been regarded with suspicion because of the risk that they might undermine the efforts of the Taliban to achieve a spiritual cleansing of the country. Isolated incidents of culturally inappropriate behaviour on the part of aid workers have tended to reinforce this perspective. Further, the Taliban have viewed the multiplicity of NGOs with concern and have felt uneasy at the very limited control which they have had over their activities in a situation in which they have been aiming to impose strict controls on the behaviour of the population. They have thus sought to bring NGOs under tight rein. In addition, NGOs have been seen as potentially sympathetic to opposition elements, by virtue of their historical links or because particular Afghan members of staff are thought to have certain sympathies or links. The fact that certain NGOs have worked in active support of governmental structures in the past, while the Mujahidin Government was in power, and are now equivocal in their support for the same with the Taliban in control, is a further factor.

This relationship has been played out in a situation where poverty is acute and where public health problems have placed large numbers of people at risk. At times, the complexities arising from the relationship have left donor governments, UN agencies, ICRC and NGOs effectively powerless to meet the needs of the population.

The current situation in the urban centres of Afghanistan is likely to be reproduced within urban centres held by non-governmental power holders elsewhere in the world, particularly if there is a civil war situation. It is far from clear whether the international community has got to grips with how it should most appropriately work in such situations.


Peter Marsden has been employed as Information Coordinator for the British Agencies Afghanistan Group for the past ten years. He is also a Research Associate at Queen Elizabeth House and has contributed to the Queen Elizabeth House study on the 'Economic and Social Consequences of Conflict’.

He has recently produced a book entitled The Taliban: War, Religion and the New Order in Afghanistan, published by Zed Books, 7 Cynthia St, London N1 9JF. Tel:+44 171 837 8466, Fax: +44 171 833 3960. Email:

This site is registered on as a development site. Switch to a production site key to remove this banner.