The recent International Conference on Afghanistan in Kabul reaffirmed the plan to begin "handing over responsibility for security" to Afghan forces in 2011. The conditions for that to happen first need to be put in place, though, namely sufficient public security and a state apparatus that contributes constructively to establishing that security. It is doubtful whether that will be possible. There has been a steady increase in violence of late – not just against foreigners but also against the local population. Short-term military successes in fighting the insurgents are within reach, but sustainable nation-building that guarantees its citizens a basic level of security, well-functioning institutions and public services and a state that enjoys the requisite legitimacy is not possible with the means available and within military time-frames. The violence in Afghanistan will only die down permanently if its structural causes can be eliminated.

Once the International Security Assistance Force (ISAF) begins its gradual withdrawal, development cooperation will be called upon even more to help establish a well-functioning state. But will it be able to deliver? Bilateral and multilateral donors are increasingly at pains to learn the right lessons from their involvement in such diverse countries as Sudan, Somalia, Sierra Leone, Rwanda, East Timor, Nepal, Haiti, Iraq, Kosovo, Cambodia, Congo, Bosnia and Afghanistan. The experiences gathered are less depressing than cases such as Afghanistan suggest. But they also highlight that despite the very best intentions there is a considerable risk of doing the wrong thing. Above all, the right questions need to be asked:

On what foundation can the state and the relationship between the state and society be re-established? Nation-building must and can only succeed if it comes from within the affected society. To that end it is important to get both the country's influential élites and the general public on board. Foreign assistance can provide valuable support, in the form of lowering the hurdles to participation, providing forums and creating incentives. However, where outsiders set the targets themselves they risk alienating important actors. We all know now that it was a mistake to exclude "the Taliban" from the process of rebuilding Afghanistan that began in 2001. Given the current situation, President Karzai has no alternative but to continue seeking to build bridges to the "moderate Taliban". It is up to Afghans themselves to decide whether Karzai will prove to be the best representative of the new Afghan state in the long run. Efforts to establish the new state on a social basis as broad as possible at the very least deserves external support. That means that external actors need the relevant know-how. They need to understand the processes at work in a society, beyond formal institutions as well, in order to be aware of the various actors' networks and sources of legitimacy, and to be able to assess what a society expects of the state. Only then can one gauge whether the nation-building process is sufficiently inclusive and whether it is likely to be successful.

Do our goals concur with those of nation-building? This is the acid test when it comes to assessing how serious the international community is about its involvement. Despite assurances to the contrary, foreign security interests for many years dictated other countries' involvement in Afghanistan. Although some goals are legitimate per se (containing the threat of terrorism, reducing drug cultivation), the outcome has been a sobering one: The successes are either fragile (international terrorism) or failed to materialise (drug cultivation). In contrast, the methods chosen have placed additional burdens on the nation-building process. They have not made foreign involvement more popular and have given rise to questionable military alliances. Over the coming years it should be wisdom, rather than mere altruism, that tells us that we should focus on achieving the goals of nation-building. Only a well-functioning state that is sufficiently anchored in society can responsibly contain security risks to the outside world too.

Have the various roles been assigned properly? International actors are best placed to successfully support nation-building in post-war societies if they are involved in their capacity as moderating "third force", negotiating peace solutions and helping to lay the foundations for a new state. For the time being most of the international actors in Afghanistan cannot take on that role since ISAF is in effect a war party. Over the coming years, development policy must try to extricate itself from its joint liability and redefine its role (even though the insurgents are currently demonstrating that they are indifferent to such subtly differentiated roles). Neutrality also means that external actors should not tie themselves to specific actors, such as the current government and to its political success. The key is to offer to help and to establish conditions on which that help will be based – whereby the standard should not primarily be compliance with the wishes of the international community, but rather loyalty to the domestic peace and nation-building process.

Is there a comprehensive development policy strategy? Where external actors support nation-building by means of diverging strategies, there is a very great risk that different interest groups in a society will feel encouraged to turn their back on the basic consensus that was once established and was inevitably linked to compromises. The most important task of the international coordination process is to avoid sending the wrong signals and to go about sending the right ones.

Can development cooperation ever be successful in fragile states given these demands? The answer is that there is no alternative. Over the coming decades the focus will increasingly be on poverty reduction, especially in fragile states. Whilst one group of traditional "developing countries" has a good chance of permanently escaping the vicious circle of poverty and a weak state, another group risks being drawn deeper and deeper into it. Maximum possible know-how about the context and better knowledge management will increasingly become the key bottlenecks of future development cooperation.

Dr. Jörn Grävingholt, Department „Governance, Statehood, Security“, German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE).