Escalation in Northern Kosovo: Causes, Dangers and Prospects
Escalation in Northern Kosovo: Causes, Dangers and Prospects
Report / Europe & Central Asia 3 minutes

Kosovo: Let’s Learn from Bosnia

After almost three and a half years working in Bosnia to implement the Dayton Peace Agreement, the international community will soon face the prospect of establishing a presence in Kosovo.

Executive Summary

After almost three and a half years working in Bosnia to implement the Dayton Peace Agreement, the international community will soon face the prospect of establishing a presence in Kosovo. The model proposed at Rambouillet was very similar to that set up at Dayton, but the situation now is very different.  This report examines the international effort in Bosnia to see whether lessons can be drawn for Kosovo and other possible future international administrations.

At Dayton it was assumed that military implementation would be harder than civilian, so the two were kept rigidly separate, and the role allotted to civilian implementation was to give a ‘helping hand’ to the Bosnian authorities, who were assumed to want to co-operate with implementation.  The Rambouillet agreement preserved the same division between civilian and military implementation as Dayton.  Yet experience in Bosnia has shown that a helping hand has not been enough since the Bosnian parties have not in fact co-operated among themselves or with the internationals.  In response, the High Representative in Bosnia has gradually taken on extra powers, but his ability to use them effectively is limited because he still has no force at his disposal to back him up.  However many powers he appears to have, he will always face this problem, even if a so-called ‘protectorate’ is established.  Currently, influence over aid is his only real power.

Civilian implementation in Bosnia has further suffered from diffusion of responsibilities between implementing agencies, who have too often been rivals rather than partners.  If the international community is going to become involved in collaborative efforts such as Bosnia and Kosovo, it needs to learn to work as a community.  Putting organisational autonomy first is not the best way to do this.  Rather, the international effort should become more like a government, with a single head and clear chains of authority within a single hierarchy.  This should apply to both civilian and military implementation if possible.  The report explores possible mechanisms for achieving this, and advocates the creation of a non-political inspectorate to ensure that work is being done to an acceptable standard.

The aid community has been less effective than it could because of a similar failure to create a culture of co-operation.  The report advocates that leading donors set up a commission among themselves to try to work better together.

With refugees the main lesson from Bosnia is that refugees do not return when local authorities do not want them.  This could still happen in Kosovo if existing authorities remain in place.  But if they do not, as seems more likely, then the task will be a more traditional one of reconstruction, complicated by the destruction of documents and deliberate planting of land-mines.  Minesweepers will have to precede the return of refugees in all areas, and a Property Claims Commission will be needed to sort out claims to property in the absence of documentation.

If Serbian troops and administrations withdraw, the KLA is the organisation best placed to fill the vacuum left behind, but they have yet to show the will and experience to run a democratic society along Western lines.  If the international community wants to work for a democratic and pluralist Kosovo it must prevent the establishment of provisional authorities who may be as hard to work with as elected authorities have been in Bosnia.  The only way to do this is to adopt wide-ranging executive powers which might indeed justify the title of ‘protectorate’, and deploy military force as a clear declaration of intent to use those powers.  This model would also permit elections to be delayed until the conditions were right for them.

The report concludes that, although one can still be agnostic about models which might have worked better in Bosnia, conditions in Kosovo are right for a protectorate-style model of administration, and in fact that the Dayton/Rambouillet ‘helping hand’ model is unlikely to be effective in a Kosovo now devastated and depopulated.

Sarajevo, 17 May 1999

Subscribe to Crisis Group’s Email Updates

Receive the best source of conflict analysis right in your inbox.