Elections in a Dangerous Place
The cover of Paul Collier’s new book, Wars, Guns, and Votes: Democracy in Dangerous Places (Harper Collins, 2009) is graced by a picture of Darfurian rebels. It is apt. Although the book says little specifically on Darfur–or indeed Sudan–the whole argument is highly relevant to the Sudanese predicament, and especially the 2010 elections.
Collier problematizes the operation of democracy in the states that contain the “˜bottom billion’ of the world’s population. Although he makes passing reference to countries such as East Timor, it’s clear that Collier is really talking about Africa, and in particular about African countries at risk of violent conflict or just emerging from conflict.
Collier shows that it is simplistic and dangerous to assume that a post-conflict election represents an exit from the risks of war. He gives the example of the Democratic Republic of Congo, where the peacekeeping mission was due to leave on October 30, 2006, the day after the run-off round of the elections. In the run-up to the elections there was a misleading veneer of civility to politics, but as soon as the results started coming in, that veneer was stripped away. Rather than the peacekeepers flying out, more foreign troops were rushed in to maintain the fragile political order. Collier argues that peacekeeping operations need to remain for at least a decade, and to be replaced with an “˜over-the-horizon’ security guarantee when the blue helmets actually leave. The only demonstrably effective exit strategy, he writes, is aid-assisted economic recovery.
The best parts of Collier’s book are his reports on a series of quantitative research exercises, undertaken by him in partnership with other scholars or research students, into the increasingly impressive political science dataset that allows us to pose questions such as, how are democracy and dictatorship associated with the risk of armed conflict? In relation to this particular question, Collier writes that the data show an ambivalent relationship. For countries with incomes below $2,700 per capita, democratic elections increase the risk of conflict. For richer countries, it’s dictatorship that is more dangerous.
Collier investigates the choice confronting the ruler of a “˜bottom billion’ (read: African) country who is under pressure to democratize. Should this ruler follow the path of governing well and seeking to win a free and fair election on the basis of his record, he faces many challenges””not the least of which is the improbability that his opponents will play this game. A range of alternative strategies includes lying to the electorate, scapegoating a minority, bribery (retail and wholesale), intimidation, excluding one’s opponents from standing, and miscounting the votes””or a combination of the above. He supplements this with empirical (quantitative) analyses of electoral strategies pursued in several countries (Sao Tomé and Principe, Benin and Nigeria) and political narratives for the conduct of elections in Kenya, Cote d’Ivoire and Zimbabwe. In these countries, all evidence points to the reality that the most irregular electoral strategies are the most effective. Candidates who try to play by the good governance rules are likely to lose. He produces evidence that suggests that in poor and unstable countries, individuals with a criminal bent are more likely to enter politics (and succeed).
While bribery and miscounting votes are the preferred strategies of the incumbents, violence tends to be the strategy preferred by the weaker challenger. For example, while there is good evidence that President Kibaki stole the Kenyan election through manipulating the count, his challenger Raila Odinga pursued a divisive and dangerous strategy that threatened or provoked violence.
Among the many nuggets in Collier’s book is an account of how voting not only tends to follow ethnic lines, but in post-conflict elections, voters tend to polarize. An example of the latter is Northern Ireland, where a peace agreement designed to reinforce the moderates ended up turning out the votes for the extreme parties on both sides. Noting that “˜within the post-conflict decade there is no safe period,’ he points to post-conflict elections as special moments of danger.
As with many books on such topics, Collier’s analysis of what goes wrong is stronger than his recommendations for putting them right. His proposals for security guarantees (to head off the risks of coups) are misplaced. Better, I think, to follow the underlying logic of his argument about the functions of elections in conflict-prone countries. The rationale for democracy is twofold: it is supposed to provide channels for civic representation and thereby remove the grievances that would otherwise give rise to rebellion, and it is supposed to legitimize the government. Collier’s analysis suggests that elections in poor and divided countries frequently do neither. He shows that representation is secured on the basis of ethnic identity and bribery, and that those who get elected are usually those most skilled at manipulating the system in a manner that is, at best irregular and at worst criminal. The outcome is neither representation nor legitimacy””though the option of dictatorship is, Collier assures us, worse.
A very important, if under-developed, side argument in Wars, Guns, and Votes is that what drives rebellion is not grievance but feasibility. In a country where an insurrection can be sustained militarily (because of a weak government and favorable terrain) or financially (due to the “˜right’ sorts of natural resources or a generous neighbor or diaspora), he implies that political entrepreneur will, sooner or later, start a revolt. The implication is that violent rebellion will not be stopped by addressing “˜root causes’ but by making it more difficult to rebel. This intriguing line of argument deserves further elaboration, because if correct it indicates that the security agenda, as such, should be the post-conflict priority.
But perhaps the most significant missing component in Collier’s analysis concerns the political functions of patronage and ethnicity. If these are as deeply entrenched as he indicates, and it will take decades of economic development for different forms of government to emerge, then should we not turn to harnessing these realities as the building blocks of post-conflict order? Implicit in Collier’s analysis is the observation that the negotiated power-sharing arrangements of a peace agreement are more durable than the elected regimes that follow. Perhaps this is because negotiated agreements are based on accommodating power and material interest””dividing executive positions and the associated access to resources””whereas the liberal democratic model of elections provides for a fair division of legislative seats, but with the “˜winner’ taking all the executive. Meanwhile the vernacular definition of democracy is that everyone should have a place at the table. Why not””as Richard Dowden proposes“”provide for executive positions and national resources to be allocated in accordance with the vote? And why not address the problem of the concentration of power by having a collective presidency, in which representatives of key minorities have a veto over key executive powers?
What does this tell us about the Sudanese elections? Collier’s book only allows us to draw general conclusions, without considering the special circumstances of Sudan, such as the ICC arrest warrant against the President and the 2011 vote on unity or secession for the south. But the generalizations based on comparative evidence are important.
The first lesson is that, as in any other country, the first post-conflict elections can be very dangerous. While the run-up to the elections may be calm””perhaps with the exception of places such as Darfur where anti-government armed groups may want to disrupt them””the real risk of violent lies in the aftermath. The fact that so many positions are up for election multiplies the risk, as governorships and state assembly seats may be more bitterly contested than national office.
The second lesson is that the elections””although delayed beyond the CPA timetable””are still much too early to be a plebiscite on the CPA. It is only through the decades-long process of social and economic rehabilitation that the success of the peace settlement will become apparent and the risks of a new war will be avoided. If the purpose of the exercise is to ascertain whether the Government of National Unity is “˜making unity attractive’, it is a foregone conclusion that the answer will be no.
The third lesson is that the best chance for minimizing the hazards is a pre-election pledge between the main parties in the Government of National Unity, to an ongoing partnership, in which the major issues and resource allocations are all decided in advance of the election. This implies that the less there is a real choice in front of the electorate, the safer the result will be”””˜safer’ for the parties also means safer for the citizens.