Adeed Dawisha and Karen Dawisha's excellent article "How to Build a Democratic Iraq" (May/June 2003) contains a series of valuable recommendations and admonitions. The few failings of their article lie less in what they advocate or reject than in what they leave out. The following recommendations draw on the experiences of other democratizing countries to fill in several gaps in the Dawishas' analysis. Together they aim to enhance the likelihood that democracy takes root and survives in Iraq.


One of the biggest dangers facing postwar Iraq is the prospect of its becoming a classic "petro-state" (like Nigeria or Venezuela), in which vast revenues from the sale of oil accrue to a shaky national government. Such states are characterized by massive corruption, fiscal profligacy, and vicious zero-sum competition for control of oil revenues. Politicians who gain power in these countries typically invest heavily in repression to retain it.

Privatizing Iraq's oil industry would not avert this danger because the immense revenue that the government would receive from royalties, taxes, or auctions could be misspent in the same way as are revenues from the sale of oil. Instead, oil revenues should be driven as far from the federal government as possible, down to states, localities, and even individuals. Dawisha and Dawisha dislike the idea of giving each Iraqi citizen a bank account, into which equal shares of Iraq's monthly oil revenues could be deposited. But this system, or something like it, should be the goal. The result would be to put the country's wealth directly into the hands of Iraqis, to remove discretionary authority from the national government, and to force the government to rely on taxation for income -- which would force it, in turn, to report and justify its expenditures.

Another key component of the transition process will be to create an effective electoral system. To ensure that early standard-setting elections are free and fair, foreigners could step into the role of conducting and overseeing them. But once the job is turned over to the Iraqis, they would benefit from the establishment of a virtual fourth branch of government charged with administering elections. The heads of this supreme electoral authority could be chosen by a supermajority of the legislature and granted sufficient funding from domestic or foreign sources to ensure their technical competence. Mexico transformed its electoral system, which was once riddled with fraud, into one of the cleanest in the world by creating just such an autonomous Federal Electoral Institute. Iraq should follow suit.

Once an electoral system is actually established, politicians respond with remarkable predictability to the incentives it produces. The trick is to get those incentives right. Dawisha and Dawisha cleverly advocate the creation of multimember districts as a way of moderating ethnic tensions. But they do not go far enough in specifying how these districts should be drawn, and we know enough from the experiences of countries such as Brazil and Taiwan to guess at the potential effects of electoral design. To prevent the emergence of any permanent political majority, Iraq should be divided into a small number of very large districts, with boundaries that match those of Iraq's three major ethnic groupings (Kurdish, Sunni Arab, and Shi'ite Arab), and possibly a separate district for Baghdad. Not all of these districts would need to have the same number of members, but seats within them would be based on a strict share of the vote. Thus, in a district with 40 members, a party or candidate that received 2.5 percent of the votes would earn 1 seat.

Such a system would fragment chauvinistic parties designed to advance the interests of one major ethnic group by activating new types of division -- based, for example, on class, sector, tribal affiliation, or policy platform -- and by diluting the salience of primary ethnic identities. Ultimately, this structure would help prevent the hegemony of any one group. If such a system encouraged too much fragmentation, districts could be made slightly smaller -- perhaps one for each existing state, with at least five members in each -- but the same principle would apply.


Dawisha and Dawisha rightly advocate a parliamentary system for Iraq, but one potential pitfall of such a system is the power it grants to the bureaucracy. In presidential regimes, directly elected chief executives exercise direct-line authority over cabinet officers, who exercise direct-line authority over bureaucrats. In parliamentary regimes, indirectly elected prime ministers and cabinet officers exercise indirect authority over professional civil servants. Bureaucrats under parliamentary systems are thus less accountable to elected representatives than they are under presidential regimes.

In postcolonial India, the effectiveness of parliamentary rule was dramatically enhanced by a clean, competent, British-trained civil service. Lamentably, Iraq has no such inheritance; after decades of Baath Party rule, its own bureaucracy is politicized and corrupt. As a consequence, additional mechanisms will be needed to prevent and redress bureaucratic abuses. One such mechanism is that most Nordic of institutions, the ombudsman. Properly staffed offices of ombudsmen at the national and provincial levels would offer Iraqi citizens the opportunity to register their grievances and seek effective administrative redress. Moreover, the office of ombudsman has to date proved quite adaptable to countries without a history of such institutions.

A democratic Iraq will also require a competent, well-trained, well-equipped, and honest police force that is separate and distinct from the military. Colombia's special antinarcotics force, built from scratch with U.S. funding, offers an example of how such units can be constructed even under the most challenging of circumstances. Other developing countries, including South Africa, have had some success in creating effective neighborhood or community police forces. As with ombudsmen, the construction and vetting of such constabularies should start immediately, allowing the new cops to earn their stripes during the period of occupation.

A related issue is how to structure the military. The danger of a budding democracy's falling to military rule looms especially large in a country such as Iraq, with its long history of coups and attempted coups. At the same time, the experiences of Argentina, Brazil, South Korea, and other countries over the last two decades suggest that states with a legacy of civil-military relations as bad as Iraq's can still establish full civilian control. Although there is no single formula for doing so, the following steps seem prudent in this case. First, Washington should limit the size of the military overall, especially the army. Second, it must integrate Shi'ite and Kurdish elements, including the Supreme Council for Islamic Revolution in Iraq and the Kurdish Pesh Merga, to ensure that the Iraqi officer corps is not dominated by Sunni Arabs. Next, the officer corps must be purged of Baath Party loyalists or those who have been involved in massive violations of human rights. Intelligence functions must be separated from the military and placed under the control of civilian agencies; any companies that the military currently runs must also be privatized. Military promotions must be made a matter of professional standards and seniority, rather than connections to the executive branch, and the legislature must be given final authority over high-level promotions and budgets. Civilian courts should get formal, ultimate control over military officers. The post of the civilian commander in chief should be specified in the constitution, and all cabinet officers, including the minister of defense, should be civilians. The military's official mission should be explicitly confined to external defense, rather than internal security or law enforcement, and military curricula must be altered to reflect this limited role. Finally, Iraq's rebuilders should encourage the training of civilians in military matters so that they can oversee budgets, procurement, and the like.


In most emerging democracies, mass media are controlled by politicized state monopolies (as in eastern Europe, Africa, and Asia) or private oligarchs who trade favorable coverage for political influence (as in Latin America). The result in both cases is a systematic bias in favor of incumbents and the wealthy, inadequate scrutiny of official misconduct, a dearth of civic and popular voices, and an absence of public forums where political options can be peacefully debated. These problems can be mitigated by placing state-owned media under the direction of professionalized, politically insulated boards, appointments to which require supermajorities in the legislature. Creating a system for allocating private broadcasting concessions that is likewise insulated from direct executive control would further strengthen press freedom. Because the freest press regimes in the developing world involve a mixture of public and private ownership, and because many local private investors may lack sufficient capital, both domestic and foreign private ownership should be allowed. Finally, a variety of protections for the press -- including laws on freedom of information and the protection of the confidentiality of journalistic sources -- can be enshrined in the constitution and the legal code. These laws hardly need to follow a U.S. format; various countries and international organizations are capable of rendering expert advice.

The challenges to democratization in Iraq are huge. Yet even if Iraq becomes a semiauthoritarian regime -- of the type that currently governs Russia -- certain democratic institutions may endure. Building these institutions now will substantially enhance the odds that Iraqis emerge from Anglo-American occupation better off than they were before.¶

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Chappell Lawson is Associate Professor of Political Science at the Massachusetts Institute of Technology.
  • More By Chappell Lawson