DAHL'S TWO IMPEDIMENTS TO DEMOCRACY: THE UNEXAMINED CAUSAL LINK

On the occasion of the first Madison Award, the recipient, Robert Dahl, gave a lecture now lost, leading me to begin with a lecture he gave two years earlier, “On Removing Certain Impediments to Democracy in the United States.” In this lecture Dahl identified two significant impediments to democracy: First, the “country's commitment to only a modest interference by government in the conduct of corporate capitalism” (Dahl Reference Dahl1977, 3). Second, “The framers' deliberate creation of “a framework of government that was carefully designed to impede and even prevent the operation of majority rule” (5).

Despite identifying these two impediments, Dahl did not point out that they were causally related. When the power of corporate capitalism is increasing, for whatever reasons, a structure of government designed to “impede or prevent” action that might stop or reverse this change becomes a contributory cause of the trend's continuation.

In addition, what neither Dahl nor his audience knew when he offered this diagnosis was that both problems were about to get worse—very much worse. This intensification of the two impediments to democracy that he identified began in the mid-1970s, when he gave this lecture and accepted the Madison Award, and it has continued ever since.

Consider Dahl's first impediment to democracy, the US “commitment to only a modest interference by government in the conduct of corporate capitalism.” This commitment had, in his view, the consequence “particularly adverse to democracy” that the ensuing economic order generated great differences “in political resources, skills, and incentives within the demos itself” (8). He reasoned that “a country committed to procedural democracy must either place effective limits on the extent to which economic resources can be converted into political resources, or else ensure that economic resources are much more equally distributed than they are in the United States at present” (16, emphasis mine). He thought that efforts to keep economic resources from being converted into political resources had “largely failed.” Thus, he was convinced: “It is time—long past time—to consider the other approach” of distributing the economic resources more equally. “The question of distribution of wealth and income,” he concluded, should be “high on the agenda of national politics” (16).

Yet even as Dahl spoke, the conditions that might have led to a more equal distribution of wealth and income were changing around him. Many of these developments had a turning point in the 1970s, as Jacob Hacker and Paul Pierson show in their recent book, Winner-Take-All Politics.

In 1976, when Dahl gave his lecture, the richest 1% of the American families received 9% of the total pretax income of the country. This was the inequality that Dahl thought needed redressing. By 2007, that top 1% was getting not 9%, but 24% of the nation's pre-tax income. The very top 0.1% was getting more than 12% of the total income.Footnote ¹

The capacity of the wealthy to convert their money into political influence has also increased greatly since the 1970s, both because the political system needs more money and because it can get more money. Electoral campaigns needed more money after the mid-1970s because television advertising became more necessary, high-quality TV ads became more expensive, polling became more frequent, and consulting became more hi-tech. Most importantly, electoral success is positional. What counts is not how much you can spend but how much you can spend relative to your opponent. Campaigns are like an arms race. In the 10 years between 1974 and 1984, real campaign expenditures by the Democratic and Republican candidates for the House more than doubled. Between 1974 and 2010, they more than quintupled.Footnote ²

Lobbying has experienced a parallel arms race since the mid-1970s. If I have two skilled lobbyists and my competitor hires three, I need at least three, if not four or six, to keep my edge (Reich Reference Reich2007). The amount spent on lobbying has nearly doubled in just the last decade (Hacker and Pierson Reference Hacker and Pierson2010, 114). Some of that money only increases the positional advantage of one corporation vis à vis another. But some of it increases the positional advantage of corporate capital vis à vis other interest groups, such as consumers and workers.

In the decade after Dahl spoke, “corporate capital” also grew far stronger organizationally. Between 1974 and 1980, the Chamber of Commerce doubled its membership and tripled its budget. The number of firms with registered lobbyists in Washington grew by a factor of 14. The number of corporate PACs quadrupled. The expenditures of corporate PACS in congressional races increased nearly fivefold between the late 1970s and the late 1980s.Footnote ³ In general, the number of organizations active in the Washington pressure system took a significant leap in the 1990s, growing by 47% from 1991 to 2001 and by 19% in the five-year period from 2001 to 2006.Footnote ⁴

This money had effects. Larry Bartels, author of the tellingly titled Unequal Democracy, has calculated that the Republican advantage in financing presidential campaigns added more than 3% to the Republican presidential vote in the 1972 and 1976 elections, and almost 7% in the next three elections (Bartels Reference Bartels2008, table 4.11). In 1976, when Dahl gave his lecture, the three main Republican party groups outspent their Democratic equivalents by about 3 to 1. Ten years later, they were outspending the Democrats by 5 to 1. When the Democrats caught up, they did so in part through an appeal to a particular sector of corporate capital—the financial sector. Hacker and Pierson report that in the 2007–2008 election cycle, the Democratic Senatorial Campaign Committee “raised four times as much from Wall Street as its GOP counterpart did.” Growing financial dependence on Wall Street since the 1970s encouraged Democrats to support repeal of regulations on the financial industry, to protect hedge-fund managers' incomes, and to roll back stock option reform.Footnote ⁵ The Supreme Court's Citizens United decision in 2010 has made the power of corporations even greater.

Since the 1970s, technological advances such as standardized containerization, cheaper air transport, fiber-optic cables, and the Internet have also reduced the costs of doing business overseas and importing overseas products (Reich Reference Reich2007). As it became easier for business to threaten both unions and government with exit, corporate power increased.

American labor unions also suffered a turning-point defeat in 1978, just months before Dahl received the Madison award. Despite a Democratic president and a Democratic majority in both houses of Congress, the unions lost their last major attempt to reform labor law—in large part because of Dahl's two impediments to democratic action. Business outspent labor 3 to 1, and a filibuster prevented action in the Senate. Since then, National Labor Relations Board sanctions against businesses attempting to prevent labor organizing have dramatically weakened. As corporate power has increased, the power of its major organized opponent has declined.Footnote ⁶

Finally, in the decades after 1975, as the financial sector succeeded in rolling back the regulations adopted in the 1930s, the big banks grew and became “too big to fail.”

Since the mid-1970s, these forces have produced cumulative feedback effects. Political victories for corporate capital create greater inequalities in resources, which in turn lead to further political victories. The effects of these trends have extended far beyond what Dahl imagined when he warned against the first impediment to democracy in the United States, the “country's commitment to only a modest interference by government in the conduct of corporate capitalism.”Footnote ⁷

The second impediment to democracy that Dahl identified—the system of deadlock embedded in the US constitution—has also gotten far worse since he spoke. The structure and mores of the Senate have evolved since the mid-1970s from more “relational” interactions (Wawro and Schickler Reference Wawro and Schickler2010, 299) to more “individualist” ones (Sinclair Reference Sinclair and Oppenheimer2002, 244). At the same time the political parties have become increasingly polarized. As a result, the number and forms of obstruction, including filibusters and threats to filibuster, have skyrocketed.Footnote ⁸ Dahl had often argued against the many veto points in the US constitutional system, perhaps most forcefully in his 1956 Preface to Democratic Theory, a work that inspired James McGregor Burns's 1963 Deadlock of Democracy. But he was unaware when he spoke both of how much worse that system would become and of how closely his two impediments were entwined.

Deadlock, James McGregor Burns pointed out, produces “drift”—the unimpeded trajectory of any externally caused phenomenon. Hacker and Pierson make this point central to the analysis in Winner Take All Politics. When external conditions—in this case the combination of increasing business organization and lobbying capacity with union decline and globalization—produce a set of trends—in this case toward increasing economic and political inequality—political deadlock does not preserve the status quo. Rather, it preserves the momentum of exogenous change. Dahl's second impediment leads directly to his first.

Political stasis has the same result on a global scale. When external conditions produce trends such as global warming and nuclear proliferation, political stasis does not preserve the status quo. Instead, stasis allows the forces that created the problem to make it worse.

Many American political scientists have suggested possible cures for the increasing political inequality in the United States and the increasing stasis in Washington. These proposals include higher taxes on the very wealthy, campaign finance reform, ending the TV arms race through regulation, delegating redistricting to independent commissions and changing the filibuster rules. Although such proposals, and the macro-political changes they might effect, are manifestly important, I do not comment on them here.

Instead I want to focus on the impediments I see in western, particularly American, ideas about democracy and relatedly, on impediments in our political theory. In some important strands of contemporary political theory, a stress on resistance plays into the dynamic of stasis in governing and thus—ironically, given the goals of the theorists—facilitates the drift toward greater concentration of corporate and financial power.

THE RESISTANCE TRADITION

What I call the resistance tradition runs deep in western political thought. The separation of church and state in Christianity, attributed to Jesus in the Gospels and cemented later by Augustine, placed these two powers in potential opposition. When the Investiture Controversy of the eleventh and twelfth centuries set the material interests of the church against those of the emperor, the churchmen—who were then the intellectuals of Europe—developed a social contract theory from the implicit assumptions of various practices in Europe at the time. They framed that theory as one of resistance to the emperor.

The earliest well-enunciated version of social contract theory comes from the monk Manegold of Lautenbach in 1085, toward the beginning of the Investiture Controversy. He wrote of a king or emperor that

Manegold also argued, perhaps with some sardonic glee, that if the owner of some pigs

Manegold's formulation puts the people in the role of a principal, hiring an agent—the swineherd or king—to do a job for them and sacking the agent if he does a bad job. It also puts the people in a surveillance role, watching to see if their hired hand does a good job and suspecting that he might not.

Social contract theory need not play a resistance role. When the Mayflower pilgrims landed on Plymouth Rock and signed a document to “covenant and combine ourselves together into a civil body politic,” they did so to accomplish collective ends. Yet in western political theory the social contract became preeminently the centerpiece of a resistance tradition that would flower in the sixteenth century among Protestant writers and become a cornerstone of early liberal democracy. Resistance to the state became a deep part of Europe's DNA. It played a founding role in the nascent understanding of democracy in the United States.

The theory of the separation of powers—an integral part of Dahl's second impediment—illustrates this metamorphosis. Aristotle had enunciated the idea that to govern well a polity needed the contributions of different kinds of government (by the best, by the few, and by the many), in a mixed polity with balance among the parts. By the sixteenth century, however, this formulation had evolved into Machiavelli's suspicion-based theory of the separation of powers, in which the role of each part is to “watch the other.” When Montesquieu elaborated this theory, he further developed its resistance aspect (rather than its balance aspect), arguing that the watching and checking function protected individual liberty. By the time Madison and the other framers absorbed the theory, the Aristotelian idea of a mixed government in which a little bit of this and a little bit of that produced the right balance for governing had evolved into a resistance idea—based on protecting each individual from state oppression.

The theme of resistance to government resonates throughout American political theory, from Madison through Thoreau and Emerson to the present day. George Kateb reminds us, for example, that Emerson saw democracy as “a device to limit rule, to neutralize legislation, … to chasten administration” (Kateb Reference Kateb1995, 190).

In the last 30 years, some of the most exciting and engaging political theorists on both sides of the Atlantic have tapped into this strand in western political thought. In American political thought, Benjamin Barber penned, in Strong Democracy, what I consider our generation's most compelling call to common action. Yet Barber argued that political representation—which lies at the base of all democracies capable of far-reaching action—is “incompatible with freedom” (Barber Reference Barber1984, 146). Sheldon Wolin, a theorist of great stature, praised the “moments of commonality” in which individuals “concert their powers” and “collective power is used to promote or protect the wellbeing of the collectivity” (Wolin Reference Wolin1994, 24, 11). Yet he also concluded that such moments will always be “episodic, rare” (11). In Wolin's view “institutionalization marks the attenuation of democracy,” so that democracy can only be “a rebellious moment” (19, 23).

More recently, Philip Pettit, who is deeply concerned with common action, nevertheless grounds his civic republicanism on “non-domination”—institutionally “constraining the will of the powerful” both in the private realm through laws and norms and in the public realm through the separation of powers and other forms of making “law relatively resistant to majority will” (Pettit Reference Pettit1997, 173). The formal dispersal of power, with the many channels it gives citizens to contest decisions, is the institutional foundation for realizing non-domination politically.

Michel Foucault's evocations of resistance have inspired writers from James Scott to Judith Butler, as well as the larger movement represented by Chantel Mouffe, the early work of William Connelly and Bonnie Honig, and others who see the essence of the political as agonism, or struggle—the antithesis of settled decision. Even Jürgen Habermas, whose deliberatively based theories have attracted sharp criticism from agonists, originally conceived the “public sphere” as a sphere of resistance to state domination in his seminal Structural Transformation of the Public Sphere.

Resistance writing rightly underscores the importance of vigilance against overweening power. But the cumulative tendency of these constant warnings is to denigrate “good enough” government, oppose coercion in all its forms, and stress the illegitimate features of any actual instance of coercion at the expense of potentially legitimate features. Such warnings have played a particularly strong role in the United States, from the standard uncritical assessment of the separation of powers in civics texts to a generalized hostility to government action of any kind.

A THEORY OF DEMOCRATIC ACTION

Let me suggest another approach, based on two premises. First, we need coercion to solve collective action problems. Second, there can be no such thing as fully legitimate coercion. Therefore we need theories that can guide public action and help improve democratic legitimacy incrementally. We can move toward the ideal of democratic legitimacy without discrediting every state that falls short of it. Resistance has an important place in any democracy, and at times the goal of democratic citizens should be stopping the state. But a theory of democratic action must also work to make institutions more and more democratically legitimate, while recognizing the good that can be accomplished by democratic coercion that is, at best, only imperfectly legitimate.

Let me clarify my use of the terms “legitimate,” “power,” and “coercion.”

By “legitimate” and “legitimacy” I mean normative democratic legitimacy, that is, the legitimacy of acting rightly according to democratic ideals. I do not mean sociological legitimacy, that is, legitimacy in the eyes of a public.

By “power” I mean, in general, the actual or potential causal relation between the preferences or interests of an actor or set of actors and the probability of an outcome.Footnote ⁹ This causal relation can work through the simple capacity to act (I turn on the light switch—my preferences cause the outcome). When the outcome requires other human beings, the causal relation can work through genuine persuasion on the merits. Neither of these is coercive power.

By “coercion” or “coercive power” I mean the threat of sanction or the use of force. I do not view the words “coercion” or “coercive” as inherently negative. It is true that both coercion and the threat of sanction will always have a negative valence. Punishment would not be punishment if those being punished did not want to avoid it. But a relationship, for example between two people, may be better when the partners have equal capacity to sanction each other than when neither has that capacity. When you care for someone, you give that person the capacity to sanction you and to threaten sanctions. Deeply interdependent social relations are, I believe, all built in part on mutual coercion.

In short, by coercive power I mean “power over,” not the nicer forms of power such as “power to” (that is, capacity) or “power with” (cooperative power). Although Arendt (Reference Arendt1970) described power as the human ability to act in concert, this is not what I mean. Those forms of power are admirable in their place. But to solve collective action problems we also need coercive power, based on the threat of sanction or the actual use of force.Footnote ¹⁰

Why is this the case? The answer has to do with the nature of goods that are non-excludable. Their character is such that everyone can use them without in any way contributing to providing them. Examples include law and order, common defense, roads without tolls, an educated workforce, clean rivers, and breathable air. Collective action problems arise whenever we want to produce such goods. All of these things, and thousands of other desirable collective outcomes, are nonexcludable, or largely so. By their nature, goods like these cannot be parceled out only to those who work or pay to bring them into being. It is only goods like these—nonexcludable goods—that produce the collective action problem, which is, at bottom, a problem of non-contribution.

Coercion is not always necessary to solve collective action problems and get people to contribute to producing a nonexcludable good. Sometimes we can produce such goods through voluntary acts of solidarity. Everyone can voluntarily chip in to build a road, defend the country, produce schooling for the poor, or abstain from overfishing. But in most cases we also need coercion around the edges to give those who are tempted to free ride on the contributions of others an external incentive to contribute. The need for coercion to solve collective action problems is, in my view, the primary reason for government. Coercion also helps human beings achieve justice together through government.

My second premise is that no actual instance of coercion can fully meet the criteria of democratic legitimacy. Over the years, democratic theorists have worked out democratic criteria for moments of both genuine commonality and genuine conflict. The criteria for moments of commonality specify, among other things, that deliberations leading to consensus should ideally take place in conditions free from coercive power (free, that is, from the threat of sanction and the use of force).Footnote ¹¹ In reality, however, the conditions for deliberation are never fully free from coercive power. As for conflict, the democratic criteria for moments of conflict specify, among other things, that ideally, decisions should be based on the equal power of each participant.Footnote ¹² In reality, however, power is never fully equal in democratic negotiation or even in majority rule, where the agenda always derives from an unequal process. Therefore the coercion that actually existing democracies deploy to implement their decisions will never be completely legitimate.

In short, a political theory of democratic action demands a corresponding theory of imperfect legitimacy. Legitimacy is not a dichotomy—a thing you either have or do not have. It is a continuum from more to less.

A political theory of democratic action should not neglect the goals of resistance theory. Every means of approximating relatively legitimate coercion has its underside. Every exercise of coercive power puts those on the receiving end of that power at risk. But simply blocking the exercise of power is often a bad solution. One version of resistance theory, attractive to the framers of the American constitution and to many since then, holds that if you put enough institutional veto points in place, the little that gets through is bound to promote the common good.Footnote ¹³ This approach privileges stopping the work of the government. It may have been appropriate in a simpler world, where it might reasonably be said that the government is best which governs least, and in a more decentralized world, where the scope of government action did not need to be as great. In a more heavily interdependent world, a democracy needs more collective power to solve the growing number of collective action problems. It can safely allow more collective power through the grid if it reduces the worst effects of that power in other ways.

Starting with the aims of the power itself, a democracy can organize itself to make the power that surges through the system more likely to promote the common good—for example, by reforming campaign finance, reducing corruption, attracting more public-spirited individuals to office, and bringing stakeholders into constructive negotiation with one another. Democracies can also devise targeted safeguards for the vulnerable—for example, by legislation such as the Voting Rights Act of 1964. Democracies can encourage both constructive and critical organization in civil society—for example, by facilitating unionization, subsidizing investigative journalism, and protecting internet access—so that new ideas feed into state power and people can organize effectively when resistance is necessary.

More generally, when a good has mixed positive and negative features—and I consider coercion such a mixed good—one should not always block or automatically resist the good, but rather look for practices and institutions that reduce its undesirable effects, protect the vulnerable, compensate the losers, and facilitate ongoing changes for the better.

In the tension between resistance and action, context is critical. Tyrannical regimes demand resistance. Deeply corrupt regimes cannot justly claim legitimacy. But when the threat of tyranny is relatively weak and corruption relatively limited, the need for collective action is often greater than the need for resistance. I do not pretend here to offer guidance to political movements on their choice of tactics, many of which are appropriately aimed at resisting particular injustices or at drawing attention to unsolved problems, such as rising inequality or global warming, even when the protesters do not agree on a plan of action (think “Occupy Wall Street”). I am arguing instead for something deeper: a shift in emphasis within democratic theory, from a long-standing promotion of resistance to the greater embrace of coercion, even while recognizing that the coercion can never be more than partially legitimate.

NEGOTIATION AND DELEGATION

Where might a democratic theory that recognizes the central role of coercive action turn its analytic gaze? Two promising and underexplored areas are negotiation and uncorrupt delegation supplemented by citizen deliberation. Theorists trying to make headway on these problems could benefit from working closely with empirical scholars of conflict resolution, comparative government, and perhaps other fields.

To explore the normative complexity of negotiations, we could begin with Denmark. In 2002, two development economists coined the phrase “getting to ‘Denmark’” to describe the goal of helping impoverished countries deliver key public services.Footnote ¹⁴ Francis Fukuyama adopted “getting to Denmark” to describe the historical paths for acquiring an effectively functioning, accountable state under the rule of law (Fukuyama Reference Fukuyama2011, 14 ff, 431 ff.). Denmark is small, homogeneous, and defended primarily by the armies of others. Like many Nordic states it has a culture that may not be duplicable. Its social welfare model has the inevitable imperfection of requiring significant barriers to entry. Yet Denmark could nevertheless serve as one model in an exploration of negotiation—specifically the contribution of different forms of democratic negotiation to relatively legitimate coercion. The outcomes of the Danish political process match what its citizens want relatively closely, and the process itself, although not based on the majority rule of alternating parties, has strong claims to democratic legitimacy.

Regarding outcomes, Denmark has the most equal income distribution of any advanced industrialized country. Robert Kuttner reported in 2008 that “Denmark's financial markets are clean and transparent, its barriers to imports minimal, its labor markets the most flexible in Europe, its multinational corporations dynamic and largely unmolested by industrial policies, and its unemployment rate of 2.8 percent the second lowest in the OECD” (Kuttner Reference Kuttner2008, 78). In its Index of Economic Freedom, the Heritage Foundation gives Denmark a score of 78.6 out of 100, or eighth place in the world, better than the United States in ninth place. Denmark has universal health insurance, good child-care and generous unemployment compensation. It has the world's second highest tax rate and spends 50% of its GDP on public services.Footnote ¹⁵

How did Denmark's democracy become capable of such effective action? First, after a series of reforms in the early and mid-nineteenth century, Denmark is now tied with New Zealand and Singapore for the distinction of being the least corrupt country on earth in Transparency International's Corruption Perceptions Index.Footnote ¹⁶ Second, because its list system of proportional representation currently produces eight parties in parliament and no single party has had a parliamentary majority since 1909, passing laws requires negotiation and compromise among parties. This system generates a more cooperative form of negotiation than in the US Congress. Third, Danish democracy has little separation of powers on the national level (although we cannot be sure this is related to its effectiveness). It has a parliamentary system with a unicameral legislature and extremely limited judicial review. Finally, Denmark has evolved a form of effective and far-reaching decentralization in which local elected bodies serve as responsive service-deliverers but not powerful veto points.Footnote ¹⁷

The result of this noncorrupt and negotiated system? Denmark's citizens have, according to the Eurobarometer, greater trust in their national parliament and their national parties than the citizens of any other country in Europe. They are more “satisfied with the way democracy works” in their country than the citizens of any other country in Europe. Staggeringly, 94% of the Danish citizens are at least “fairly satisfied” with the way democracy works in their country.Footnote ¹⁸ To this sociological legitimacy, add some normative legitimacy from two features. First, Denmark's citizens are actively engaged in their politics. Without any compulsory voting, the turnout in the general elections since 1960 has averaged 85%. In The Economist's 2010 Democracy Index, Denmark has the third highest score in the world, after Norway and Iceland.Footnote ¹⁹ Second, the very process of negotiation adds democratic value by drawing out the reasons and justifications advanced by the different parties.Footnote ²⁰

Danes have also shown their capacity for resistance when needed. In 1943, when the German army occupied Denmark, the public denunciation of the German plan for deporting the Jews involved the King, the universities, students, the Danish state church, the Supreme Court, the trade unions, the employers' confederation, the farmers' organizations, the heads of ministries, and all of the political parties except the small pro-Nazi National Socialist Workers' Party of Denmark (Kirchhoff Reference Kirchhoff1995). Denmark's corporate entities were actually the foci for resistance.

I am not saying that the United States can model itself on Denmark. That would be absurd. Nor am I saying that Denmark is a perfect polity. Its protections for its own citizens are arguably related to its relative homogeneity and its barriers to immigration. Finally, I do not have sufficient empirical data to judge the relative merits of the different systems that are less prone to deadlock, whether majority-rule Westminster systems or well-structured systems of negotiation, or the roles of different kinds of veto points, which in some contexts may promote, rather than hinder, common democratic action (Birchfield and Crepaz Reference Birchfield and Crepaz1998). I am saying that these questions need entwined empirical and normative attention. In the future I hope that comparativists will read more democratic theory and theorists more comparative work, to the point where each can, with the help of their colleagues, contribute productively to the development of both fields. In particular, I am urging here that political theorists can profitably ally with comparativists and other empirical political scientists to investigate the sources of democratic legitimacy in countries other than our own and Great Britain. In Denmark, we might concentrate on the strengths and weaknesses of their forms of negotiation. These forms of negotiation, developed historically not only by Denmark but also by other relatively neocorporatist states in Europe, have heavily influenced the relatively successful processes of the EU bureaucracies, which unlike Denmark have highly heterogeneous constituencies.

Just as one size does not fit all in economic development, so too one size does not fit all in the building of legitimate democratic action. The new field of comparative political theory is investigating, among other things, the sources for democratically legitimate action in the cultures and philosophies of nonwestern countries. My point is that as this work goes forward, the focus should be as much on the sources of coordinated, intelligent action—and relatively legitimate coercion—as on resistance.

If we think about problems of global scale, like climate change and weapons of mass destruction, the focus on action becomes even more necessary. Decisions at the global level cannot be as democratically legitimate as those at a national scale. In the foreseeable future, decisions at the global level will be even less likely than those at the national level to be discussed, much less resolved, in an arena governed only by the “forceless force of the better argument.” Nor will decisions be made in a way that even approximates the equal power of each individual or the proportionate power of those affected. To achieve action capable of addressing collective action problems on a global level, we will have to accept ongoing coercion that is far less democratically legitimate than the coercion we accept at the level of the nation state. Yet we must take action, as soon as is humanly possible, for the sake of unborn generations.

Elections are only the starting point of democracy at the global level. Elections produce representatives, who make appointments, and those appointees make appointments and those appointees make further appointments. As the lines of delegation get longer, they get farther and farther from their source of legitimacy in the will of the people. But the lines themselves can also be more or less legitimate. One major requirement for their legitimacy is being relatively clean—as free as possible from both illegal corruption and legal, institutional corruption.Footnote ²¹ In addition, we need to think harder about the deliberative criteria for democratic legitimacy when delegated appointees make decisions.

If, as I believe, democratic legitimacy inheres not only in aggregative forms such as elections but also in deliberative forms, greater democratic legitimacy could derive from structures that embed the delegated decision makers in networks of public-spirited peers and epistemic communities, surface relevant criteria for decision, and thus improve the quality of consideration of available options. Greater legitimacy would certainly inhere in structures that encourage deliberation in a context of mutual respect and attention to the public interest, as observers report seems to be the case in at least some parts of the EU bureaucracy (Naurin Reference Naurin2007). Greater deliberative legitimacy could also derive, when time permits, from supplementing long lines of delegation with mechanisms that generate citizen input in the decisions through deliberative initiatives, stakeholder groups, or random selection. Such groups must be consultative, not empowered, until such time as empowered groups, chosen by normatively legitimate processes of lot or election, can command sociological legitimacy. Inserting empowered citizen groups into the end of the process now would short-circuit legitimate lines of delegation.

In this dual vision of delegated legitimacy, clean lines of delegation provide the formal legitimacy, while a variety of other mechanisms enhance the legitimacy that derives from taking into consideration the interests of all affected. Judgments about the legitimacy of decisions that emerge from these long lines of delegation should be based not only on the formal legitimacy of their mandate and the legitimacy of the specific forums in which the decisions are made, but also on the legitimating elements of the deliberative and representative systems that feed into those forums.Footnote ²²

At the global level, a commitment to only modest interference with capital also poses a threat to humankind. An unregulated market of corporate capitalists, pursuing the best deal and responding to (or creating) demand, will inevitably generate global financial busts, global warming, arms escalation, and nuclear proliferation. The market, with its many virtues, requires regulation, as does the power of capital to intervene in and illegitimately skew political decisions. At present we are almost unimaginably far from getting to Denmark on a global level. But if we are to approach that goal, democratic governments need to find a way to interfere more than modestly with capital.

A THEORY FOR OUR TIMES

Dahl's two impediments to democracy in the United States are thus impediments to progress not only at the national level but at the global level as well. On both levels we must innovate both institutionally and in the realm of democratic theory. We need theory that can focus on action—and coercion—as well as resistance, on negotiation as well as deliberation, on systems as well as specific forums, on delegated power as well as direct democracy or representative elections, on new forms of representation as well as old, and on new ways of engaging citizens in conceiving of and acting for the good of unborn generations as well as our own.

In short, we need a democratic theory for our times, both at the state and the global level. Madison's theory of democracy will no longer work. The plethora of veto points that the framers recommended and that the state legislatures accepted will not allow enough positive action. At the global level, we need delegates of nations who can negotiate and coordinate with one another to produce action. That action will require coercion, which should be as legitimate as possible. Ironically, collective action at the global level may need to look more like the negotiations leading to collective action in the tiny state of Denmark than like the negotiations leading to collective inaction in the huge two-party United States. As we try to bring the actions of both our own nation and the globe closer to the ideals of democracy, I assume that the regulative ideals will remain, at base, equality, freedom, the common good, and the protection of basic interests, particularly those of the most vulnerable. But how we interpret the normative and institutional implications of those ideals will have to evolve as we think harder about our ideals and learn from experience.