Making Democracy Stick

An ambitious strategy of democracy promotion is poised to be a major pillar of U.S. foreign policy for many years after 9/11, just as Cold War containment, trade liberalization, and development assistance were pillars of American policy in the decades after 1945. The strategy of democratization must begin with the moral proposition that “the call of freedom comes to every mind and every soul,” as President Bush said in his second inaugural address. But if the strategy is to succeed, we have to ask and answer some hard questions about what obstacles exist to achieving stable democracies and how they can be overcome. That the strategy faces challenges is not doubted, least of all by some of its leading advocates. Bush acknowledged “many obstacles” to democratization and called it the “concentrated work of generations.” British Prime Minister Tony Blair has said that “democracy is hard to bring into countries that have never had it before.” Even Natan Sharansky, author of a relentlessly optimistic appeal for democratization, says that in places like Iraq, democracy faces “a very difficult transitional period.”[1]

Visiting Scholar Gerard Alexander  
But these champions of democratization emphasize obstacles to transitions to democracy rather than obstacles to the stability of democracies afterward. Bush and Blair and authors like Sharansky and Joshua Muravchik repeatedly reject the notion that fully functioning democracies may face more structural obstacles even after they are inaugurated. They especially reject two long-standing claims: that stable democracy requires certain cultural preconditions and that stable democracy is possible only above certain per capita income levels.

There are, indeed, solid grounds for rejecting both: Several democracies have endured in what are, by the standards of these claims, inhospitable cultural and economic contexts. But more often than not, the reasoning of the democratization advocates goes farther, implying that no societal attributes are necessary preconditions for stable democracy.  Sharansky, for example, sweepingly rejects the “idea that certain peoples are incapable of democratic self-rule” and the notion that “there are certain cultures and civilizations that are not compatible with democracy.” [2] Consistent with this, while some programs of the National Endowment for Democracy (NED)–the main U.S. entity tasked with promoting democracy–also seek to strengthen existing democracies, most recent U.S. policies are designed to help tip countries from the authoritarian category to the democratic. [3] That tipping is seen as the biggest challenge.

These advocates offer a powerful justification for their optimism: the universal hunger for liberty. President Bush’s letter introducing his 2002 National Security Strategy proclaimed that “People everywhere want to be able to speak freely; choose who will govern them; worship as they please.” At a November 12, 2004 press conference, Bush said he believed that successful democratization among Palestinians “can happen, because I believe people want to live in a free society.” Standing at his side, Blair said that “given the chance, [Iraqis will] want to elect their leaders. Why wouldn’t they? I mean, why would they want a strong-arm leader who’s going to have the secret police, no freedom of speech, no free press, no human rights, no proper law courts? The people want the freedom.” The NED’s “Statement of Principles and Objectives” states that the idea of democracy is “intrinsically attractive to ordinary people throughout the world. . .an ideal that billions of people in all parts of the globe revere and aspire to.” Sharansky says succinctly that “all peoples desire to be free.” [4] These champions seem to be saying that where there is this much will, a way will be found to create stable representative institutions–indeed, that will may be the way, especially once people are offered the opportunity.

But there are compelling reasons to believe that certain structural conditions threaten democracies in ways that cannot be overcome simply by a desire for self-rule. If America’s democratization strategy is to fulfill its early promise, we have to identify these obstacles and decide how they can be overcome. Recent history provides a powerful lesson of what will happen if we don’t.

An unsatisfying record

After 1945, western governments launched what became a massive program to spur economic growth in less developed countries. Like today’s democratization project, this program was announced in speeches–such as Harry Truman’s 1948 inaugural address–which called forth the energies of great nations and fired the imaginations of millions. Like today’s emerging democratization project, this program was simultaneously noble and self-interested, since world poverty was seen as the cause of several ills of global reach, including disease and political instability. And the economic development program certainly became the concentrated work of generations. It spawned vast bureaucracies. One expert estimates that advanced industrialized countries gave $1 trillion (in constant 1985 dollars) in official development assistance to developing countries between 1950 and 1995 alone [5]– a figure that does not include many more billions in subsidized loans, forgiven debt, and private charitable donations. The program to spur economic development ranks as one of the most ambitious collective projects in history.

Yet it’s difficult to call it a success. Some aid recipients have become wealthier, but many remain desperately poor, and a number have even deteriorated economically. Around a billion people are still estimated to live on less than $1 a day, and many more on less than $2. There is bitter disagreement over why this is. But we can agree that the universal hunger for wealth was not sufficient to cause growth. Sustained growth has been achieved in a wide enough range of circumstances for us to conclude with confidence that no country is permanently barred from it. But it has also failed to occur so often that we have to conclude that some conditions must be necessary for translating the desire for growth into its accomplishment. And it has proven difficult to identify these conditions. This is why development strategies have shifted repeatedly, from Truman’s call for the sharing of “industrial and scientific techniques,” to funding for heavy infrastructure, to later liberalization, to encouragement of reformist political leadership, to recent attempts to address dire poverty and achieve the rule of law. Since growth spurts have fizzled repeatedly in some countries while igniting again and again in some others, we can also say that the conditions necessary for growth must be deeply entrenched and are not things that routinely fluctuate, such as individual national leaders. Sustained growth is possible in any country; but it appears impossible under certain conditions, and some of these conditions may characterize certain countries for many years.

In the same way, stable democracy has occurred in so many circumstances that we can infer that no country is permanently barred from it. But democracies have also collapsed after being launched in so many other cases that we must also conclude that certain conditions must be crucial for stable democracy. These conditions also seem deeply entrenched, because democracy, like growth, has proved serially unsuccessful in a number of countries but highly stable in others. The launching of a democracy is invariably a great thing, but it is not a great indicator of what will follow. Charles Krauthammer captured this point in a Washington Post column (March 18, 2005): “We do not yet know whether the Middle East today is Europe 1989 or Europe 1848”–a reminder that even the countries which have finally created very stable democracies once could not (in some cases, for another century after 1848). If stable democracies require certain underlying conditions, then time is not on the side of a new democracy that is launched in a country where those conditions aren’t met or are insufficient.

If this is right, and if the issue of underlying conditions is not addressed head-on, then there is a very real risk that half a century from now, people will judge the current democratization project much as we, today, judge the economic development project: as a noble but largely unsuccessful effort, one that left in its wake at least as much disillusionment as accomplishment. The architects and advocates of democratization have a big stake in ensuring that Natan Sharansky is not seen one day as the Gunnar Myrdal of a different historical project and that a future P.T. Bauer does not emerge to show how democratization went terribly wrong. Everyone has a huge stake in getting the democratization project right from the start. The first step is to identify the underlying conditions that stable democracies rely on and to figure out how those conditions can be promoted.

What stable democracy requires

Democracy is often portrayed as a set of individual rights. Natan Sharansky tests for its presence by asking, “Can a person walk into the middle of the town square and express his or her views without fear of arrest, imprisonment, or physical harm?” [6] In the parlance of economists, rights like these are usually “nonrival goods”: Their exercise by any one citizen has virtually no effect on their exercise by any other. But democracy has two other features, concerning contestation and capacities. And these features make stable democracy dependent on underlying conditions.

First, a regime is a democracy only if contested elections result in governments that produce economic and other policies binding on all citizens. Contestation means that parties are able to win but are willing to lose. In other words, opposition parties have to be able to compete effectively with those currently in power. They must have the credible potential to hold incumbents accountable. And voters and parties must be willing to lose elections. Second, a regime is a democracy only if policies and individual rights actually are the law of the land, only if they are effectively enforced. This means that a stable democracy requires a government with the capacity to enforce both the rules of the game and the policies produced through those rules against violation or nullification either by abusive agents of the government itself or by private actors, whether common criminals or would-be warlords.

A democracy is present only if this willingness to lose and these capacities to challenge and enforce are present. A democracy is stable only if these features of democracy are all renewed on a regular basis. Three broad characteristics of countries shape this willingness and these capacities. The first two concern the balance of power between state and society; the third concerns a factor within society itself. First, the government has to be powerful enough to enforce rights and policies. But, second, it must not be so powerful that officeholders can become unaccountable even to majorities of their citizens. And third, the major political groups must not be so mutually threatening that they would rather overturn the democratic game than lose an election to their adversaries.

It’s no coincidence that democracies have proven most stable where these three underlying conditions have been met. These include today’s North Atlantic region and Australia and New Zealand. Sturdy democracies also emerged–to the surprise of some–in several East Asian countries, including Japan and later South Korea, where strong states are counterbalanced by vibrant societies and economies based on growing middle classes and where party systems are strikingly compact rather than polarized. These conditions also more or less characterize some(but not all) Eastern European countries. They also appear in India, where an effective state structure left by the British governs a society that is exceedingly diverse but has not been powerfully polarized ideologically. In all these cases, governments have authority, but robust societies–whether wealthy or not, and whatever their cultural origins–accountability; and major parties are typically not divided by differences worth toppling democracy for. These are the building blocks of stable democracy.

Champions of democratization would be right to suspect, and even bet, that these essential raw materials exist in countries where authoritarian rulers may nonetheless be clinging to power. In such countries, if only dictators could be toppled, democracies would be likely to endure. These conditions existed in several East Central European countries even while they were still occupied by Soviet troops. They existed in Spain by the late 1960s, and in Mexico probably by the 1980s, even though well-institutionalized dictatorships maintained firm grips on power there. Not surprisingly, democratic processes stabilized soon after being launched in these countries. There are surely more cases like this, waiting to be tipped into the democratic category. Malaysia and Singapore are plausible candidates. These countries are the low-hanging fruit of the ongoing democratization project.

But to say that this applies to all countries is to assume that the necessary underlying conditions are universally present. This is not the case. And where it is not, new democracies risk eroding with time rather than enduring.

Three unfavorable conditions

A new democracy is undermined by one of three main problems: severe political polarization, an over-powered state, or a too-weak state.

Polarized conflict: In all democracies, policy disagreements make voters and activists resent losing elections. But history repeatedly shows that democracy is at dire risk from “losers” only when the major political groups in a country have agendas so mutually threatening that their dearest-held interests and values are at stake. Then, losers can face an unpalatable choice. In democracy, they fear an overwhelming threat when their adversaries win. But they might plausibly hope that a sympathetic authoritarian regime would protect their interests and values while targeting its repression primarily against their adversaries. Authoritarian “entrepreneurs” (say, in the military) have often offered losers such protection in return for support. Many coup plots have been hatched in just this way.

This explains why several times in the twentieth century, large sectors of the population supported military coups in Southern Europe and Latin America, countries in which parties on the left and right were bitterly divided over property rights and religion.[7] In the post-ColdWar period, Haiti’s economic elite backed the military overthrow of elected President Jean-Bertrand Aristide in 1991 when they detected a gathering social revolution from the left. The next year in Algeria, moderate and secular Muslims and economic elites rallied behind a military coup rather than lose an election to Islamists.[8] In each case, people constituting a fifth, a third, or more of the population made a choice that observers can find baffling: They backed the rise of a regime they knew would rule them by force.

If polarization is deep-running enough, any new democracy is at risk. Because the conflicts in Haiti  and Algeria, for example, were the result of durable features of these societies, there was every reason to believe that even if the 1991 and 1992 coups had  been speedily reversed, the major political groups would simply have resumed an unwillingness to lose, elections. We see exactly that in Haiti, where international intervention reversed the military coup there in 1994. From then until 2004, Haiti resembled less a stable democracy and more an armed camp in which major groups eyed each other with the same wariness as before. [9] In 2004, armed groups successfully mounted a de facto coup. In polarized conditions, time is not on a new democracy’s side.

Where does this jeopardize the current democratization strategy? Polarization has famously ebbed in some regions, especially Europe. But in others it has not, and new instances of polarization have arisen. Classic and sometimes bitter struggles over land ownership and income are far from resolved in parts of Latin America. And the rise of political Islamism since the 1970s may have the same polarizing effect in parts of the Muslim world as the rise of socialism did in nineteenth- and early twentieth-century Europe, leaving more moderate Muslims alarmed at the prospect of ceding power. Wherever agendas are this mutually threatening, major groups remain unreliable democrats.

Over-powered governments: In all democracies, incumbents have advantages like patronage and publicity. But this advantage widens dangerously for any prospective democracy when incumbents control the central or even sole source of wealth in a society that is otherwise poor.[10] This reaches extreme forms in countries like Saudi Arabia, where revenue from government-controlled oil represents the taproot of income on which the entire population is ultimately dependent. When this happens, regimes often dominate their societies not merely through heavily funded security forces but also through patronage spending on such a large scale that few sectors of society remain truly independent of cooptation by and pressure from the regime. Many (in some cases, most) citizens receive some benefit at the partial discretion of the rulers–public or private employment, income support and social services, diverse consumer subsidies, licenses of all kinds–that they might find reduced or withdrawn if they supported a true opposition movement. These governments may not be strong internationally, but they tower imperiously over all other organizations in their own societies.State power on this scale does more than make existing authoritarian regimes resilient; it also jeopardizes any democracy that might be installed in such a setting. Human nature being what it is, even democratic officeholders are likely to resort to the use of these same patronage tools as soon as they come under the pressures of public opinion and competitive elections. The resulting gross imbalance of power would mean that truly independent sectors might well be too anemic to police the state/society boundary that is indispensable to democracy. Concern over precisely this problem has sparked discussions of how to prevent oil revenue from being used abusively by future elected leaders of Iraq.[11] It also fuels acute worries over whether Hugo Chavez is using his economic power to subvert democratic processes in Venezuela.

This hazard to democracy exists everywhere that governments in poor countries directly receive large-scale international “rents.” This is most visible in oil-awash countries in the Persian Gulf(and Libya). But other political systems in the Middle East are heavily affected by it as well, as in the cases of transit rents from Syria’s oil pipeline fees, the international aid showered on the Palestinian Authority, and Egypt’s canal income (as well as its massive annual subsidy from the United States). A number of countries in sub-Saharan Africa have such impoverished economies that even rents from mineral exports and from foreign aid (which can comprise half the government’s budget) give officeholders enough patronage power to severely distort the viability of an independent opposition. Stabilizing a new democracy in the shadow of such a state would remain an uphill battle.

Weak governments: Ironically, democracy is also undermined when the state is too weak, as Francis Fukuyama has recently reminded us.[12] When the government’s capacity to enforce its will does not extend to substantial parts of a country, democracy is undermined in several ways. There is no consistent enforcement of individual rights. The policies produced by democratic authorities are not, in fact, the law of the land. Instead, people are forced to live with binding political decisions made by whoever does exercise local power, typically through undemocratic means. Finally, a too-weak government may become incapable of defending itself from these substate challengers. We call these “failed states” and their tormentors “warlords.” In these cases, the capital city might host a government that was elected, but the country cannot meaningfully be called a democracy. It may do no good to temporarily keep the warlords at bay and hold new elections. If the central state lacks the power to enforce the rules of the political game, then regional power brokers will resume undermining the new regime as soon as outsiders look away. The relevance of the “democratic” label will erode along with it. 

Instead of getting rarer with time, weak states like these have proliferated in the post-Cold War world. This has inspired extensive discussion of failed states in large parts of sub-Saharan Africa, especially countries racked by civil wars in West and Central Africa, and in Afghanistan, despite ongoing improvements there. By the same standard, a major concern about Iraq is not that the central government in Baghdad lacks the will to impose order on the Sunni triangle but that it could conceivably lack the capacity to do so. Wherever this capacity is a serious problem, new regimes that manage to stabilize are unlikely to deserve the label “democracies.”

These three problems can also interact–lethally. Polarization can accompany an over-powered state. It appears that in Algeria, not only were truly independent “civil society” groups weak in the face of a state in command of all the country’s export earnings, but many of them sided with the 1992 coup out of concern that the Islamists were even more illiberal than the military. And a state can be simultaneously weak and over-powered. In several African countries and in Afghanistan for most of its modern history, governments have often been too weak to enforce the rules of the democratic (or any other) game but too strong to be challenged effectively by a democratic opposition.[13] The result is that elected leaders can degenerate into enclave rulers jealous of any opposition. Presidents of several African countries have turned out to be little more than the mayors of their capital cities.

Rolling up our sleeves

This survey suggests that stable democracy faces daunting challenges in large parts of the Middle East and North Africa (because of over-powered states and some polarization), sub-Saharan Africa (simultaneously weak and over-powered states), and perhaps parts of Latin America (lingering polarization) and Central Asia and Burma (over-powered states). But this enables us to ask the authentically optimistic question: What can we do to promote underlying conditions that are more favorable to stable democracy?

Easing polarization: The international democratic community can try to help address theunderlying sources of polarized conflict and mitigate its short- and medium-term effects. At the very least, democratic politicians can be pressured to moderate. Many Western governments pressed leftist leaders to avoid radical rhetoric and policies in several transitional cases in Southern Europe and Latin America in the 1970s and ’80s. Decision-making institutions can be designed to encourage power-sharing and compromise. Of course, rules can be revised. And advice can be rejected: The U.S. government repeatedly advised Jean-Bertrand Aristide to moderate after his original election, after his 1991 overthrow, and after his reinstallation in 1994. Both these strategies, therefore, have trouble making all sides to a fierce conflict enter into stable commitments to democracy. But temporary patches can provide time for more ambitious projects to accumulate the effects of profound change.

In that spirit, the international democratic community can attack the sources of underlying conflicts. This will usually be laborious. Conflicts over religion might be eased through programs aimed at systematically encouraging moderate religious leaders and teachings. Western countries have yet to mount a systematic moderating strategy of this kind in the Muslim world, which would require local interlocutors. This approach would attempt to counter the Saudi government’s global program of funding and promoting its radical variant of Islam. Conflicts over property and income might be eased in agrarian settings through peaceful land reforms–it is time to revisit the socioeconomic “hearts-and-minds” components of older counterinsurgency strategies–and in other settings through programs that enable social mobility, including micro-lending, tax reform, and deregulation.

Weakening over-powered states: The international democratic community can help create conditions favorable to stable democracy in countries with over-powered states by empowering sectors that are truly independent of state control. This notion lies behind Western programs aimed at “strengthening” civil society in Africa and the Middle East, which have funded nongovernmental organizations that provide social services and even those of a more political(pro-democracy) nature. But this will not result in viable opposition sectors as long as these NGOs are highly vulnerable to being coopted by the resource-rich state. One researcher has recently warned that while Middle Eastern ngos focusing on social services are “sometimes viewed [by outside donors] as a potential counterweight to state power,” they are in fact “largely an extension of” state power because they are heavily funded by the regimes as well, and their members–such as trade unionists and professionals–rely on policies made at rulers’ discretion.[14] So far, NGOs like these have not acted as effective forces for liberalization and democratization.

Only a shift in the balance of resources can fuel a truly independent opposition. This requires that more sectors of the population develop sources of wealth independent of regime control. Three main options are on offer. The first is large-scale economic redistribution that disperses existing economic resources away from rulers and their economic allies.[15] Leaving aside how this might be brought about, this approach leaves unclear why future regime revenues would not be used gradually to reassert regime control over the same assets, if under a new guise. A second option is to steer the future stream of rents away from regime leaders altogether by setting up an oil fund with specific spending tasks or directly paying each citizen his “share” of oil revenue on the model of the Alaska Permanent Fund.[16] But there is a problem here. In this scheme, the law creating the fund is now the guardian of democracy. But what will guard this guardian from a power-grab by officeholders? This is not an issue in Alaska, where control of the fund is guaranteed by an already robust democratic process; it is another matter when democracy is supposed to be guaranteed by the legal status of a fund.

A third option is for new wealth to be created in these countries, which would make international rents less important to begin with. This may sound an odd prescription for, say, oil rich countries. But on the whole, their economies have been and remain poor outside of the oil sector, and it is poverty that magnifies the political importance of international rents.[17] Broad based and diversified growth might create an independent middle class and business community of the kinds that have served East Asian democratization so well. There is a chicken-and-egg aspect over how to proceed: The policy reforms needed to achieve such economic growth–reforms ultimately intended to create the basis for strong democracy–might be unfeasible so long as authoritarian rulers are determined to monopolize power and its material sources. But the record suggests that even partial juridical and economic liberalization can allow enough broad-based growth to create further pressures for liberalization. For example, the Financial Times (September 13, 2005) reports that private enterprises have gone from being a negligible share of China’s economy to what might be a majority of its gross domestic product today. The most optimistic scenario, though not the inevitable one, is a virtuous circle between reforms, growth, and more reforms.

Strengthening weak states: This may be the most resilient obstacle to creating underlying conditions favorable to stable democracy. We know frustratingly little about why some states are too weak to create order, enforce rights, and implement policies. Correspondingly, the “international community knows how to supply government services; what it knows much less well is how to create self-sustaining indigenous institutions,” as Fukuyama understatedly puts it.[18] The sociologist Robert Nisbet argued that people learn to be part of political society by participating in smaller communities first; this seems to call for a policy of building states by first strengthening civil society. But Nisbet was generally describing “Goldilocks” situations, where the state was neither too strong nor too weak, neither domineering nor incapacitated. But when states are over-powered, civil society groups are vulnerable to being co-opted by regimes; and when states are under-powered, they become vehicles for power brokers emerging from un civil society, who have their own agendas and who use intimidation and force to get their way. The most common alternative strategy for dealing with weak states is to pour international resources into building up state capacities–including bureaucracies and security services. This has often been ineffective; creating a police force and a courthouse does not necessarily create order or justice. Worse, international projects of state-building risk creating a repressive state powered by virtue of aid rents rather than being nourished by roots in the local society. This may be the result of the European strategy of heavily funding the Palestinian Authority. The result may even be an enclave state that stifles dissent in a few targeted zones but is powerless outside them, a state that is too strong in its headquarters and still too weak everywhere else.

Fukuyama notes the sobering possibility that at least in some cases, international projects of economic development and nation-building “have actually eroded institutional capacity over time.”[19] As a result, the international democratic community is likely to pursue two agendas–development and democratization–that may at times be at loggerheads with one another. For now, the test case for U.S. policy toward failed states is post-Taliban Afghanistan, where elections accompanied the gradual buildup of a military force and a bureaucracy apparently loyal to the central government and the incremental degrading of the power of regional warlords. But the pudding is not ready for eating. We will not know whether weak-state status has been overcome until the government in Kabul is no longer dependent on foreign aid for its budget and NATO troops for its security. Until then, we need to pay as much attention to state building in Afghanistan as we’re paying to democratization in Iraq. Until we know more about how to overcome failed-state status, we may have to accept that many countries in sub-Saharan Africa and parts of the Muslim world may be the greatest source of frustration for democratizers for some time to come.


The truest words about democratization were spoken when President Bush referred to this project as “the concentrated work of generations.” And there is every indication that he meant it. The American call for democratization has been universal, but serious pressure for it has been more selective, and force has been used against only two dictatorships, both of which posed direct threats to the wider international community. To prosecute the war on terror, the administration has cooperated extensively with authoritarian regimes in Kuwait, Pakistan, and Saudi Arabia, presumably on the assumption that they will remain on the scene at least for some time to come. The main change since 9/11 has been that the Bush administration is not indifferent as to regime type. It actively prefers democracy whenever that is feasible. It urges democratizing reforms with unprecedented consistency. And it explicitly envisions a world in which all countries one day are governed democratically. To realize that vision, we don’t need to know where democracy will take root next. We need to know under what conditions it can ever take root.

Gerard Alexander is a visiting scholar at AEI.


1. Council on Foreign Relations, “The Power of Democracy,” remarks by Natan Sharansky(November 9, 2004).

2. Natan Sharansky, The Case for Democracy (PublicAffairs, 2004), 6, 16. Sharansky does not discuss a single social condition that might make democratic stability less likely. See also Joshua Muravchik, Exporting Democracy (AEI Press, 1991), especially chapter 6.

3. For example, the Iraq project was intended in large part to overcome lingering regional unfamiliarity with democracy and doubts that democratization is feasible. In his second inaugural address, Bush sought to overcome despair among democratic activists currently living under authoritarian regimes. And the United States has ramped up rhetorical and other forms of pressure on dictatorships to persuade rulers to cede at least some ground. Symbolically, the NED’s most recent “strategy document” emphasizes challenges of stability in new democracies far less than did the version adopted a decade earlier. Their content can be compared at: and at:

4. Sharansky, Case for Democracy, 17.

5. William Easterly, The Elusive Quest for Growth (mit Press, 2001), 33.

6. Sharansky, Case for Democracy, 40–41.

7. This is the theme of Gerard Alexander, Sources of Democratic Consolidation (Cornell University Press, 2002).

8. William Quandt, Between Ballots and Bullets: Algeria’s Transition from Authoritarianism (Brookings Institution Press, 1998).

9. Robert Fatton, Haiti’s Predatory Republic: Haiti’s Unending Transition to Democracy (Lynne Rienner, 2002).

10. See, for example, Michael Ross, “Does Oil Hinder Democracy?” World Politics 53:(2001).

11. See Nancy Birdsall and Arvind Subramanian, “Saving Iraq from Its Oil,” Foreign Affairs(July-August 2004).

12. Francis Fukuyama, State-Building: Governance and World Order in the 21st Century (Cornell University Press, 2004).

13. Barnett Rubin discusses this pattern in The Fragmentation of Afghanistan (Yale University Press, 2002).

14. Amy Hawthorne, “Is Civil Society the Answer?” in Thomas Carothers and Marina Ottaway, eds., Uncharted Journey: Promoting Democracy in the Middle East (Carnegie Endowment for International Peace, 2005), 93.

15. David Waldner recommends this for Iraq in “The Last Time Iraq Tried a Parliamentary System, It Ended in Failure, Under Circumstances Not Unlike Today’s,” Newsday (January 23, 2005).

16. This has been recommended by, among others, Fareed Zakaria, The Future of Freedom(W.W. Norton, 2003), and Vernon Smith, “The Iraqi People’s Fund,” Wall Street Journal(December 22, 2003).

17. Andrew Schrank, “Reconsidering the Resource Curse: Selection Bias, Measurement Error, and Omitted Variables,” paper presented at the 2004 annual meeting of the American Political Science Association, Chicago.

18. Fukuyama, State-Building, 42.

19. Fukuyama, State-Building, 100.