Skip to navigation – Site map

A Comment on ‘Democracy Promotion at a Local Level : Experiences, Perspectives and Policy of Swiss International Cooperation’ by Martin Dahinden

Olivier Roy
This article is a translation of:
Une réaction à « Democracy Promotion at a Local Level : Experiences, Perspectives and Policy of Swiss International Cooperation » de Martin Dahinden

Full text

1The article ‘Democracy Promotion at a Local Level : Experiences, Perspectives and Policy of Swiss International Cooperation’ by Martin Dahinden presents a fine and complex analysis of what could constitute a concrete and efficient democratisation aid policy; it highlights possible limitations of such a policy, the complexity of levels of intervention and the need to implement this policy within the context of a given civil society and local power dynamics.

2Such democratisation aid clearly arises from an assumed voluntarism (it is practically a moral and political obligation) and while the perspective is meant to be pragmatic and realistic, there is nevertheless an idealistic dimension. One does not negotiate on values inscribed in the Swiss Constitution nor on the democratic model that one wishes to introduce. This model is based on a precise two-tiered definition of democracy: majority rule (elections) and the rule of law (separation of powers, constitution). Since free elections can bring actors to power who do not favour the rule of law, it is important that both pillars be raised simultaneously. The rule of law, freedom of expression, respect for human rights, an independent judicial system and impartial administration are essential, non-negotiable conditions when one speaks of democracy.

3The first problem that  therefore arises is that instead of examining the local political culture an imported model is being used - even though this definition of democracy is supposed to be a universally acknowledged one.

4That said, gone are the days of direct intervention (from Dayton to Bagdad by way of Kabul), which were inspired by the 1945 model (implanting key-in-hand democracy in countries assumed to lack or to have forsaken democratic values). This is because direct intervention has failed, at least relatively, but above all because it implies prior direct military intervention that the West is no longer willing or able to pursue. However, abandoning the project of exportation, whose failure has clearly been underscored by Mr Dahinden, has not yet led to a revival of cynical Realpolitik (though one might ask whether the West's failure in Syria is not facilitated by a certain dose of cynicism: since most of the actors are a potential threat to the West, let the massacre continue ).

5The problem is therefore to help establish a precise model of democracy, relying in principle on local structures, since direct management is out of the question. When the countries concerned are themselves actively involved in a democratisation process, the approach is straightforward: in such cases, technical assistance may be desired (at least officially) by the states in question and may also enjoy a broad consensus within the political class as well as among government officials or local actors. The matter then becomes a purely technical one, free from major political considerations. The ‘know-how’ shared within the framework of cooperation has a multiplier effect and serves to accelerate the democratisation process (this was the case of the former communist countries in Europe). However, when neither the local authorities nor the dominant political culture favour democratisation, the provision of democratisation aid must be dealt with  differently.

6As pointed out in thearticle , in case democracy cannot be exported, and a state is not interested in a democratisation process, or a state does not exist  (failed state), democratisation must take place through a relatively autonomous ‘civil society’, a term that is somewhat overused nowadays. On the positive side,  experience has shown that there is almost always a popular demand for democracy. However, this does not mean that democracy automatically ensues; we recall that in every single election that followed the collapse of a dictatorship - Iraq, Afghanistan, Kyrgyzstan, Tunisia - or was held in the wake of sudden electoral openness - Algeria in 1991 - there was a major voter turnout despite a very real danger in some cases (Afghanistan, Iraq). And in each case, the election was followed by disappointment due to a lack of institutions and competent policymakers. In other words, the process lost its dynamic. Today, the areas concerned are mainly the former communist territories, the Balkans, the Caucasus and Central Asia, as well as the Middle East, i.e. countries where Muslims form the majority, countries where Muslims form a strong minority as well as countries in conflict with Muslim countries. Islam is therefore often perceived as a specific obstacle to democratisation.

7The problem  therefore is one of finding ways to encourage a democratisation movement in areas where the states or local power brokers (warlords) oppose it. Reference to the Swiss model is particularly interesting in that this model is based on the idea of building democracy from the bottom up rather than pursuing a more or less ideologically driven state project; this model is well-suited for fragmented societies (multi-ethnic, tribal or geographically isolated societies) where development may be experienced at different rates from one zone to another while nevertheless having an osmosis effect. Within this context, the pursuit of micro-projects makes perfect sense but raises other issues.

8Since the model used is rarely indigenous, democratisation aid implies the need to select local actors who share the ‘western’ vision of democracy or  who may potentially act as instruments of democratisation through their personal activities (e.g. independent journalists). This therefore implies the creation of interfaces (NGOs, foundations, development agencies) that identify and reinforce local structures  and serve as a means of circumventing state authority or powers that may not be in favour, or may even be against, a democratisation project.  

9In other words, such a vision of democratisation involves social engineering, by selecting and encouraging specific actors and groups who will receive aid. This is done with the aim of enhancing their influence and role in society (support for the weakest groups such as women, etc.). The insistence on ‘gender’ is interesting (and relatively new to western aid with the concept of empowerment, i.e. increasing one’s capacity to exert local influence over collective decisions). However, gender mainstreaming at times is perceived as both an attack against the local culture and wishful thinking: while no assessment has yet been made of the situation in Afghanistan, local power structures do not seem to have changed as a result of measures aimed at giving women greater influence at the local level, e.g. by making gender parity mandatory in village councils responsible for aid distribution. Civil society is supposed to be relatively autonomous, which implies that when civil society is oriented in a given direction (reinforcing weak actors, for instance), this should have an almost mechanical effect on that society’s capacity for democratisation. And yet, there are doubts as to whether civil society organisations (CSOs) are truly rooted and have the capacity to drive democratisation policies even at the local level.

10The article  does not elude the problem: any action taken to reinforce and include ‘weak’ groups invariably has local political consequences because it calls existing power relationships into question and may lead to reactions of hostility (assassinations or abductions) or circumvention (maintaining appearances but leaving the true power structures intact). The risk is to create islands of democratic practices, akin to an Indian reservation where independent journalists, female entrepreneurs and human rights activists would live in a loosely integrated space on the margins of society and under western protection, whether such protection is provided directly or at arm’s length (i.e. the latter could be the case of the young Malalai in Pakistan).

11The challenge is therefore considerable. However, another problem also needs to be considered: democratisation is certainly more than a mere technical approach. It is an eminently political endeavour that calls into question power relationships at both the local and national level. A democratisation policy may reach a deadlock over local and national political stakes because the dual focus placed on civil society and local actors of democratisation may fail to consider the political game unfolding within the given society. Every society is political and local figures have national relays. The Arab Spring also showed that ‘objective’ actors of democratisation may not have a democratic vision of society, as was the case for instance of the Salafi mouvements s in Egypt, which enjoy strong support among voters; this voter base mistrusts the elitist Muslim Brotherhood and perceives the Salafis as being closer to the populace and therefore more ‘democratic’ (a vote cast in favour of the Salafis is not necessarily a pro-Sharia vote, since the Muslim Brotherhood also supports Sharia rule). How is one supposed to deal with local actors who enjoy popular support but who also favour the Sharia over democracy?

12The key, as highlighted in the report, is to work at the local level. However, as we have seen, the ‘political’ actors, i.e. those who vie for local power are not necessarily the ‘right’ actors from the western standpoint. This is because their power is derived from the use of force (e.g. Afghani warlords) or because their position within civil society is the result of practices that are inherently undemocratic (notable figures, clan or tribal chiefs, aristocracy in all its forms). It is therefore not uncommon to see a western actor call for the removal of these bad actors, if necessary through legal means (accusing warlords of war crimes, for instance), which automatically expels them from the political arena. However, these individuals have no intention of disappearing quietly. We must therefore re-examine the main premise underpinning all action taken in favour of democratisation, namely that “there is no democracy without a democrat”: it is a matter of chicken and egg, since democracy is not a feature, it is a system. One can only work with actors who are committed to ‘real’ local political life, who therefore also have their own interests at stake. It is important to build democracy at the local level but to also ensure that this democracy is a political one, i.e. where all actors indeed have real ‘power’ and not merely democratic ideas. In this sense, the creation of a municipal democracy should be a priority objective. It was the practice of local democracy, for instance, that convinced  the leaders of the Turkish AKP to play the democratic game, regardless of their own personal convictions. This is clearly expressed in the article : democracy is primarily a matter of politics.

Top of page


Electronic reference

Olivier Roy, « A Comment on ‘Democracy Promotion at a Local Level : Experiences, Perspectives and Policy of Swiss International Cooperation’ by Martin Dahinden », International Development Policy | Revue internationale de politique de développement [Online], 4.3 | 2013, Online since 11 November 2013, connection on 21 September 2017. URL : ; DOI : 10.4000/poldev.1653

Top of page

About the author

Olivier Roy

Professor of political science at the European University Institute (Florence, Italy).

Top of page


Creative Commons Attribution-NonCommercial 3.0 Unported License.

Top of page
  • Logo The Graduate Institute of International and Development Studies
  • Logo DOAJ – Directory of Open Access Journals
  • Logo ERIH PLUS | NSD
  • Les cahiers de