Our Programmes



Sign up to our newsletter.

Follow ODI

State-Building, Aid & Security in the Context of Failed States Samuels

Time (GMT +01) 16:00 17:30


Dr Kirsti Samuels, State-Building & Constitutional Law Consultant, UNDP Somalia


Karin Christiansen, ODI


Diana Cammack, ODI

Dr Kirsti Samuels presented the findings of three evaluations of state-building interventions, and particularly the role of constitutional processes in securing stable and peaceful states. She also spoke of findings from her current experience leading UNDP in its support of constitutional processes in Somalia.

The three evaluations she had been involved with recently are:

  • A project with the International Peace Academy in New York, which focused on seven country studies over 2 years. The seven countries examined were East Timor, Afghanistan, Somalia, Occupied Palestinian Territories, Haiti, Bosnia Herzegovina and Liberia. Seven cross-cutting themes of state-building were looked at, including security, legitimacy, institutional building, rule of law, public finance and economic policy. The edited volume will be produced in the next 6 months.

  • A second project with the International Peace Academy focusing on constitutional choices in post-conflict settings. This project looked at governance choices within constitution making processes in post-conflict societies and the impact of these chosen structures on peace and stability 5-10 years down the track. Countries examined include Bosnia, Fiji, Lebanon, Northern Ireland, South Africa, and Uganda.

  • A project with International Institute for Democracy and Electoral Assistance (IDEA), based in Stockholm. This evaluation looked at the processes supporting constitutional development in post-conflict societies, and which seemed to result in more stable political processes. It considered 12 case studies. The report and case studies are available on the IDEA website.

General findings emerging from the studies

Initial findings paint a less than positive picture about the success of recent state-building strategies. For instance, despite received wisdom having it that the efforts to rebuild East Timor have been successful, without an adequately trained or empowered judiciary, the result has been an overwhelmingly strong Executive, which is counter-productive. Similarly, repeated state-building attempts in Haiti have had only limited successes - for instance at one point, a functioning police force. However, even limited successes can be undermined by political elites without interests in supporting the outcome.

Referring to the international donor community and UN institutions, Dr Samuels said

'We [the international donor community] really don't know how to do it…Too often institution-building results in buildings, desks and granting official titles without the culture necessary to ensure they work...'

From her studies, Dr Samuels feels there is a need to really focus on changing political culture and supporting long term and broad based processes of public dialogue. Constitution-making is one opportunity for such inclusive and participatory processes. The evaluations show that constitution making processes that supported such dialogue resulted in more democratic constitutions, greater public support for the constitution, and encouraged a process of healing. Constitution building processes should not be approached as legal, technical assistance exercises. Changing political culture is one of the most difficult developmental processes. It should involve extensive civic education, hard political dialogue and consensus building - processes which are often overlooked in favour of technical solutions that are easier to implement, budget for, and report against. But if (as in South Africa) there is an extended and staged process of dialogue (e.g. up to five years), ther eis hope that the resulting government is more representative, more consultative, they reform, they debate, and they construct, and the people at least try to hold them to account.

While constitutional processes are not the 'magic bullet' everyone seeks, and the nature of conflict in conflict torn societies can change and undermine state-building efforts even years down the track, nonetheless, extended processes of dialogue around political processes is the area where Dr Samuels feels there is most hope.

"In the end the only thing that works is long term education - broad based, and particularly of women. You won't get a democracy by training politicians. You will if you empower people to hold their politicians to account."

Another lesson emerging was that international support should support home-grown solutions as much as possible - some of the most interesting political developments in Somalia and Somali-land occurred in areas where there was little or no international attention.

Power sharing between warring factions is often focused on in peace processes, and then carried over to the constitutional process. While power sharing is seen as crucial to peace in the initial stage, 10 years down the track, case studies show it hasn't actually brought stability. As Dr Samuels said

'The games that players continue to play when they are in power together in a power-sharing environment - which is to undermine each other…threaten to bring down the government…means you either get stagnation, as in the Lebanon model, or everything falls apart, as in the Northern Ireland model'.

On the other hand, federal structures are seen as much more stable - there is a lot less ongoing negotiation on a daily basis, the implication being that processes are allowed to develop and then negotiate power institutionally rather than inter-personally.

A surprising finding was that the use of electoral structures to moderate political divisions is, in Dr Samuels' words, 'An extremely risky strategy'. A strategy that forces people, for instance, to vote for multiple parties in order, one of which is from a different ethnic group to that of the voter, has extremely unpredictable outcomes. In Fiji, for instance, it resulted in votes going towards the most extreme parties.

Discussant: Karin Christiansen

Multiple donors in countries made 'game theory complicated'. For instance, 43 official donors in Uganda, with different funding streams and objectives, fundamentally challenged the government in its ability to manage the aid industry. This was true elsewhere too, and made medium term goals that supported local decision making difficult.

Karin Christiansen suggested there are four habits of aid that create these difficulties:

  • Faddishness: 'We invent a new silver bullet every year or so'.

  • The fallacy of composition: i.e. the theory that if we all do something well as individuals, it will add up to something greater than the sum of the whole.

  • 'Pattern switching': i.e. wanting to abstract a pattern of state building experienced in Westminster to other countries. 'A 200 yr messy process that resulted in a political settlement here won't produce a political settlement elsewhere'.

  • Participation: "we don't really know how to do it".

Ms Christiansen argued that the aid industry, and donor governments more broadly, need to be much more realistic about what can be achieved. Constitutional development won't create a broad based programmatic politics let alone poverty reduction. In particular, she argued, funding should be separated from hard conversations. 'We bribe people into saying what we want'. We need to engage in a much more realistic discussion of the nature of politics in countries and the limits of our control and influence. 'We should worry less about them and more about us - or rather how we behave. Take responsibility for our own actions in these contexts'.

Questions from the floor

Q: What makes you think that in Somalia what you are doing will work any better than any similar processes that have been tried by the UN and others before? How will the interest or informal politics be involved and formalised equitably?

A: You have to look at ongoing and underlying processes. Peace agreements seem to be quick fixes to elite interests e.g. one warlord ends up being Minister for Diamonds, and one Minister for Roads. Such appointments don't really address the needs of the civic population. But if you put pressure on those elites to go and speak to the people, they often find it is educational and enlightening.

Q: I am surprised by how well Somaliland is doing - with almost no aid industry altogether. Many people are surviving on remittances. Could you expand on whether UNDP will build on constitutional process there, successful, that is not recognised?

A: Somaliland went from chaos like Somalia but created something recognisably constitutionally constructed. Interesting lessons exist in how they evolved their process. From a 'strong man' state they reverted to traditional mechanism of elders meetings, which through processes of negotiations with warring factions ended up producing a bicameral parliament. The situation is different in Somalia because traditional structures were destroyed by colonisers in a way they weren't in Somaliland.

Q: This is an extraordinary exercise in donor self-flagellation - is the conclusion that there should be some political selectivity - that we should only fund or support legitimate burgeoning countries? How would choices be made?

A: Many people say the aid industry should get out. But no one can suggest a better strategy. The World Bank of course started to think about performance based aid - well performing countries. But they realised this wasn't responsible, so LICUS strategies were born.

Q: What are the implications of fragmented approaches? Do we need to create a dominant donor or agency to drive things in particular areas?

A: Not sure one dominant donor is the right approach - but that there is a dominant process and that people are really having hard discussions over prioritisation. You can't have 400 different consultation processes - this is very dangerous. Hope is a very fragile thing in these situations. If you disappoint people's expectations, you can create quite a lot of resentment, disenchantment, alienation. Developers need to get much better at job creation, and we could learn from China - e.g. throwing lots of people at infrastructure development. While there is a risk of creating a boom and bust, at least it gives people the capacity and the means to hold people to account - not just education but livelihoods.


In the sixth meeting in the 'State-Building, Aid and Security in the Context of Failed States' series Dr Kirsti Samuels presented the findings of three evaluations of state-building interventions.