Complexity and Democracy: Pathologies of Modern Governance Paul Babbitt Southern Arkansas University Prepared for delivery at the 2012 Annual Meeting of the American Political Science Association, August 30-September 2, 2012 ©Copyright by the American Political Science Association Draft: Please do not cite or circulate without author’s permission. It is unlikely the assertion the world is becoming more complex will raise eyebrows. Rapid changes in technology, potential for catastrophic environmental changes and increased economic interdependence are among the challenges confronting policymakers. It is commonly asserted that in the face of such challenges, democracy needs to step aside and allow experts and bureaucrats increased discretional power—technocracy is viewed as the only solution to governance in times of complexity. Alas, a casual observation of how things are going in the world reveals that technocrats are not capable of dealing with those challenges. The economic collapse of 2008, one caused almost entirely by technocratic decision makers in and out of government, and the subsequent failure to do much to rectify the situation is but one recent example of the failure of elites.1 Under such circumstances, it might be reasonable to conclude that the world has become ungovernable. However, 1There is much more to be said about the role of elites in the economic collapse. See, for example, Shapiro 2011, 18-19. It does not seem to be a coincidence that the problems with the Euro are so intractable and that the European Union often seems to be nothing more than an attempt to implement a technocratic government not directly accountable to voters. 2 properly understood, democracy can address many of the problems arising from governing a world increasing in complexity. In some respects, this paper is offered as a rejoinder to common criticisms of democracy and public opinion—the ignorance of the people, the instability of public opinion, and an irrational public, but one that is also rationally ignorant. Despite the well- documented shortcomings of the public, the decision to move away from democracy demands a comparison with the proposed alternatives. It is important to compare the performance of democracy with the performance of actual experts, not abstract experts endowed with perfection knowledge. Of course democratic governments make poor decisions. But so do non-democratic governments. The historical record is clear— democratic governments tend to outperform non-democratic governments on almost any reasonable measure. Thus, advocating democracy may not require a tradeoff between a less efficient but more just government. Democracy works. Not perfectly, of course, and not all the time everywhere, but most of the time. Moreover, the expansion of democracy in the last two decades also suggests that the conditions necessary for democracy are less stringent than many have thought. Again, the important point is not that the nascent democracies in Eastern Europe, Latin America, Africa, and now parts of the Arab world are perfect or even admirable. The point is that installing democracy has not made any of these nations less well-off. Technocracy is government by modern day guardians. These modern guardians are not just making choices about the means to reach democratically determined ends, but have substantial influence over ends as well. Sometimes, in the case of modern theocracies 3 or on the other end of the spectrum Marxist inspired governments, the moral leadership is explicit. In modern liberal democracies, the moral claims are perhaps less explicit but still exist. Generally, the technocrats follow some form of utilitarianism and make decisions to maximize some good or set of goods. At base, choosing among goods is a moral choice. Tradeoffs among collective goods are complex decisions. Pluralism is another factor that adds to moral political complexities. Thus, even if technocrats make claims about their moral neutrality, in the end, they make moral choices. There are good reasons to suspect the superiority of public opinion over technocracy in making moral choices: “In spite of the limited information at their disposal, many simple men are often wiser than their governments; and if not wiser, then inspired by better or more generous intentions.”2 No doubt, public opinion has from time to time favored horrific injustices. But no elite has ever been immune from committing injustice. In addition, even if there is a conceptual distinction between means and ends, maintaining such a distinction in practice is impossible. In any event, unraveling all this exposes a profoundly complex moral universe, even when there is widespread agreement regarding fundamental values. The defense of democracy offered here is of democracy as currently practiced, full of short sighted rationally ignorant self-interested individuals. Thus, this paper presents an account of democratic theory at odds with much that underlies the “deliberative” or other demanding models of democratic theory. The advantages of democracy for dealing with complexity are consistent with representative pluralist democracy, provided certain basic conditions regarding free exchange of information and public influence on policy are met. Abandoning unreachable standards of citizen competence and civic virtue does require 2 Popper, 468. 4 lowering expectations for democratic outcomes. Under current conditions, public opinion may be at times wrongheaded and foolish. Second, such democracy will often entail little more than the aggregation of individual or local interests rather than the communal pursuit of some common good. However, a democratic government has greater capacity for self- correction than ideologically constrained experts. Thus, rather than being antagonistic or in tension with one another, democracy is well suited to provide governance under conditions of increasing complexity and interdependence. To make appropriate adjustments, it is important to understand the specific elements that are conducive to effective democracy under conditions of complexity. Conventionally, much democratic theory holds that simplicity is more conducive to democracy than complexity. Rousseau and Montesquieu saw small, homogeneous communities as the only place where democracy was likely to thrive. However, by connecting democracy to the “general will” what small-state theories of democracy are in fact doing is creating the possibility where the “public good” would be obvious, and dissent would simply be irrational. In short, the account of democracy is one that would produce much the same kind of decisions as a monarchy or aristocracy. In conditions of complexity, by contrast, it is precisely the absence of any knowable objective public good that makes democracy necessary. James Madison turned this account on its head, and argued that in a large republic, majority factions that threatened liberty would be less likely to form. Thus, Madison saw that in large republic, popular government was less a threat to liberty than in small communities. In any event, the potential for democracy does depend in part on scale. But if democracy is seen as an instrumental good, rather than a good in itself, then participatory limitations of large complex democracies are not necessarily a problem as 5 long as the opportunities for participation are sufficient to preserve the advantages of democracy for dealing with complexity. Empirical studies of democracy complement this idea, presenting a public ill- informed and thus not competent to handle the business of governing. Much empirical work in political science has documented piles of evidence of the incompetence. Charges of incompetence have also penetrated popular discourse, though often the charge of ignorance has more to do with disagreement than knowledge levels. These studies have identified voters seriously deficient in even the most basic knowledge of politics and public affairs. Their opinions on policy issues often appeared arbitrary and unstable, so much so they were often considered “non-opinions,” statements made only because the citizen was asked a question in a survey. The research faulted most Americans not only for their ignorance of “basic political facts” but also for their lack of “ideological constraint” that is they failed to follow the ideological patterns widely used by political elites. Indeed, it seemed to the researchers that their opinions failed to follow any discernible pattern at all. Many theorists are unsurprised by these findings. Joseph Schumpeter and Walter Lippmann argued that democracy in the sense of a government of the people is impossible and the best that can be hoped for is a competition among elites for popular support.3 Public opinion could in no way be used to guide public policy. Other theorists imagine popular government posing a threat to individual liberty and private property.4 On the other side of the political spectrum, other theorists posited that democracy failed to serve 3 See Schumpeter, 250-283 and Lippmann, 51-53. 4 The list of theorists here is substantial, but prominent examples would include F.A. Hayek, J.K. Rowley, James Buchanan, and Gordon Tullock. A recent sustained argument along these lines can be found in Caplan, 2007. 6 the interests of the people because the people were duped by some kind of hegemony.5 Despite their ideological variations, what these criticisms share is that they begin from a certain preferred outcome—a more aggressive foreign policy, a socialist revolution, or an acquiescence to free market capitalism and note ways in which democratic governments fail to provide those outcomes. The fault lies with democracy, not the preferred policy of the author. In the case of civil liberties, there is evidence that public opinion sometimes falls short of protecting groups particularly disliked is often cited as evidence of the necessity of a something along the lines of judicial review, to protect the masses from themselves. In the United States at least, judicial review’s record of protecting civil liberties is decidedly mixed.6 Moreover, the distinction between elites and the public in support of civil liberties is smaller than the distinction among elites competing in a democratic system.7 Theorists more sympathetic to democracy have also noted the paradox of the growth of democracy that coincides with the impotence of democratic governance. According to Norberto Bobbio’s paradoxes of democracy, democracy appears ill-suited to govern a modern world of large, complex, states and yet at the same time it is precisely in the context of such states that demands for democracy are most insistent. Indeed, historically, bureaucracy and democracy developed and grew simultaneously.8 Looking at the same conditions, John Dunn argued that in fact the growth of democracy was largely a 5 I have in mind here Gramsci. But any argument premised on the notion of false consciousness would apply here. 6 Dahl 1956, 105-112. To be sure, Dahl’s account is old. But the basic point remains—for most of its history, the US Supreme Court has not been especially counter-majoritarian when it comes to protecting civil liberties and civil rights. 7 Sniderman, et al 353. 8 Bobbio, 68-70. 7 fiction, standing not for government by the people, but instead, for the ruler’s “good intentions”. Despite these good intentions, modern democracy is an oxymoron and in fact modern states concentrate more power at their center than ancient or medieval states could even imagine.9 As the world becomes increasingly complex, the capacity of the public to manage with rapidly changing conditions is strained. The only possible form of democracy therefore is some version of elite democracy where the people cede the power of government to one or another of competing elites. In addition to the problem of technocratic failures, reconciling a system where the technocratic elite—a modern “guardian” class—threatens democratic governance.10 Despite this, increasing numbers of people throughout the world are living in democratic nations. Yet as non-democratic forms of government receded in respectability, elite democracy emerged in contrast to what is unflatteringly called mass democracy. The idea of mass democracy was perhaps best articulated by Col Rainborough at Putney in 1647: “The poorest he that is in England hath a life to live, as the greatest he; and therefore truly, Sir, I think it's clear, that every man that is to live under a government ought first by his own consent to put himself under that government; and I do think that the poorest man in England is not bound in a strict sense to that government that he hath not had a voice to put himself under.”11 Elite democracy gives the people some say in which elites govern. But in so doing, the role of the public is thus reduced to that of a rubber stamp, lacking even real veto power. Thus, elite democracy is hardly to be considered democracy at all, and has more in common with rule of the one or the few, rather than the rule of the many. Even if 9 Dunn, 12. 10 Dahl, 335. 11 Puritanism and Liberty, 53. 8 elite democracy is still considered a form of democracy, its pathologies are different from the pathologies of mass democracy, and have very different origins. The pathologies of elite democracy have grown acute over the last several decades as elites struggle to cope with increases in complexity and the technical demands of governance. I want to suggest the inverse of this common account. Rather than being unsuited to modern complexity, mass democracy is the only form of government capable of coping with the challenges of complex policy issues. Characteristics of the politically sophisticated, particularly the ability to think in increasingly abstract or theoretical categories, are especially ill-equipped to deal with increasing levels of complexity. In a complex world, “ideological constraint” or dogmatism is especially ill suited to even acknowledge rapidly changing conditions and can be recognized as one of the causes of our current political malaise. It turns out that in the United States at least, the public is much less ideological extreme than the elites, and that the increased polarization of the elites has led to a “disconnect” between public opinion and policy.12 Thus, ideological constraint has become antagonistic to democracy. There is at least some evidence that ordinary people think about politics in nuanced and sophisticated ways not easily uncovered by public opinion research. The link between the structure of public opinion and the success of democracy is obscure to say the least and depending on how the researcher approaches the question, can lead to contradictory results.13 I argue that the absence of ideological constraint is not a weakness of mass democracy, but rather a strength. Ideology is but one 12 Fiorina, xix. 13 Hochschild 188. 9 heuristic among many human beings can use to form consistent positions on many issues.14 The insights from these accounts of public opinion have direct bearing on evaluating the value of various normative accounts of democratic theory. Central to this relevance is the way in which ambivalence and uncertainty in the face of complexity and unpredictability is a most reasonable response. Furthermore, it is a response that provides support for political liberty. Liberal democracy only makes sense in a complex uncertain world, otherwise, we would do better to follow the one true path to paradise. Rather than analyze the strengths and weaknesses of the individuals who make up a democratic system, it may be better to account for its success (and its limitations) by looking at the system as a whole. That is, public opinion, democratic competence, and civic virtue cannot be considered in isolation. Rather, the system as a whole is what matters. People, at least compared with technocrats, are generalists rather than specialists. This, however produces virtues essential to a democracy: “tolerance, capacity to cope with ambivalence, openness, and a willingness to learn.”15 Democracy is a “self-organizing system.” Since democracy involves more than government, but both formal and informal organizations of every imaginable type, some deeply engaged in politics and some not at all engaged, it reflect the world in which it exists. The self-organizing nature of democracy in fact suggests democracy is most suitable for complex. societies, and complex societies are most suitable for democratic governance. 14 Sniderman, et al 428. 15 Moon, 221 10 What is Complexity? In science, complexity is related to chaos theory. A formal description of chaos is any set of interactions where small differences in initial conditions yield large differences in outcomes.16 Chaotic systems can be entirely deterministic but are difficult or impossible to predict.17 Complexity, however, resists formal definition. In terms of politics a working account of complexity involves multiple interactions among known and unknown agents. Furthermore, complexity is also involves the types of interactions and agents. As the number of interactions and agents increase along with their types, complexity increases. Given that politics may or may not be deterministic, and given its complexity, reliable predictive models are likely to be elusive. Integrating insights from complexity theory is not unknown to political science, but seems to be at odds with both behaviorism and rational choice approaches to politics. In public policy, complexity increases epistemological uncertainty, which raises serious problems for linear approaches to public policy.18 Herbert Simon was a pioneer in describing complexity in general and made contributions in this regard in political science. He describes complex systems as one “made up of a large number of parts that interact in non-simple ways.”19 Though somewhat tautological (for what else is non-simple but a synonym of complex?) this description suggests understanding complex systems as any system where reductionist accounts fail. The practical advantage of this account is that it describes complexity in terms of the capacity of various models used by policy makers, 16 Mitchell 20. 17 Mitchell 33. 18 See Sanderson 2009 and for a response, Little 2012. 19 Simon, 1962 468. 11 social scientists, and other experts. The disadvantage is that the description is not objective or easily operationalized. Complexity is understood in terms of the problem at hand and the tools available to deal with that problem. Some political theorists have considered the impact of complexity on politics as well. Danilo Zolo identifies four characteristics of complexity relevant to political theory, specifically, democratic theory. First, complexity increases as the number of options increase and the number of variables that must be considered increase. Second, complexity increases along with the interdependence of those variables. Third, complexity results in instability, or rapid change. Finally, people in a complex environment recognize that and realize that the environment will resist prediction.20 Given that over history, society has grown more complex in the sense that there are more entities interacting with each other at an ever higher velocity, it is easy to see how chaos has increased from the time when people largely lived in isolated villages, when travel and communications were slow, and much of the world could be safely ignored most of the time. Complexity is often conflated with advanced technology. However, they are two distinct issues. The problems that advanced technology poses for modern political societies are different in important ways from complexity. To determine the safety of a nuclear power plant is safe or not is a relatively straightforward problem for nuclear engineers. The problem of how to incorporate that kind of expertise into a democratic system is well known and has attracted considerable commentary. The confusion between complexity and technological sophistication generates the idea that a technocratic elite, necessary for 20 Zolo, 3-4. 12 developing and administering advanced technology, are also best suited for dealing with complexity. But some of the more challenging problems for public policy do not really involve technology in the conventional sense. Rather, the problems are better understood as problems of complexity—problems where multiple entities interact over time in numerous ways. Even if the behavior of each entity is well understood, the ways in which multiple simple entities interact are often chaotic. To be sure, technological advances in communication, information storage, retrieval and processing, and transportation are causes of increased complexity. But in isolation, their effects are not really complex in the formal sense. Whether I use the internet or a cell phone, a text message, or a regular phone to communicate with a colleague, the result is more or less the same even if the volume has increased. That is, the initial conditions do not make much difference in the outcome. Rather, complexity emerges because lots of people are able to communicate with lots of other people rather rapidly, and no one can reliably anticipate the results of such communication. In politics, this has raised a number of interesting questions regarding the impact of social media on politics, the way in which it is on the one hand more difficult for governments to keep secrets but on the other hand it appears to make government monitoring easier, the decline of professional journalism and the rise of amateur blogs, among many others. Though there is an obvious rise in complexity, it does not stem from the growth of sophisticated technology per se but rather the increased complexity of communication networks the technology makes possible. One premise behind this work is that government and corporate bureaucracies and technocracies are increasing ill-equipped to cope with complexity. To an extent, the 13 empirical evidence this of seems obvious. The reasons why are less so. Again, consider the continuing worldwide economic malaise. Economists have used a number of abstract models to recommend policy changes that they predict will lead to economic growth. But while economics may be up to the task of explanation, it is not clear that it is able to predict. In short, the public may be correct to disagree with the widely held views of economists.21 Even after the fact, causality is impossible to demonstrate given the epistemological problem of counterfactuals. The bail out of investment banks in 2008 may have prevented all out economic catastrophe, but no one knows for certain. Or, the economic stimulus of 2009 may have saved a number of jobs, but maybe it did not. It seemed to be disappointing to those who promoted it. Technocratic and bureaucratic organizations deal with complexity with hierarchy and specialization. Democracy, by contrast, remains a system without formal hierarchy or specialization. Even the informal hierarchies described in the empirical literature featuring opinion leaders and the like are flat relative to the hierarchies of bureaucracy. And democracies provide a kind of specialization as well—call it geographical specialization, where individuals are intimately familiar with people and events in their immediate environment. Hierarchies based on expertise may be an appropriate way to deal with technological sophistication. Doctors, engineers, and the like are well qualified to deal with the specific problems for which they have been trained. They have an important role to play in the policy making and evaluation process as well. But they have no authority to govern. Even if some have greater expertise in an area relevant to public policy than others, 21 Gaus, 397. 14 it does not logically follow that they should govern.22 Like the craftsmen in Plato’s “Apology,” their expertise in those areas does not qualify them for ruling others. Democracy and Self-Organizing Systems The study of complexity examines how self-organizing systems emerge. The idea of a self-organizing system is simple enough—a system that functions with no direction from above. The individual units organize themselves into patterns. Examples include the way neurons in brain result in a unified consciousness and how evolution has produced multi- celled organisms, and the way the immune system works. In the social sciences, the idea of self-organizing systems is most familiar in economics. For instance, in the work of F.A. Hayek, markets organize themselves without (or in spite of) policy planning. These processes have certain common elements—for instance, trial and error is an important element in how these systems work. However, there are important differences as well. For instance, the organizations in question solve difficult problems—the distribution of resources or the defending an organism from dangerous infections. But other examples, such as consciousness or the evolution of multi-celled organisms, appear to be an epiphenomenal by-product of a complex process. In economics, the basic idea is that the market produces allocations of resources more efficiently than could be accomplished by bureaucratic planning. The nature of this process is not mysterious. Economic choices are made by human beings who are pursuing their self-interest. In doing so, as Adam Smith noted, they serve the public interest: “Every 22 Estlund 3. 15 individual is continually exerting himself to find out the most advantageous employment for whatever capital he can command. It is his own advantage, and not that of society, which he has in view. But the study of his own advantage naturally, or rather necessarily leads him to prefer that employment which is most advantageous to the society.”23 Markets are simply the result of a sum of individual decisions and economic transactions. From these uncoordinated individual acts, a market emerges, and in general, the more these decisions are made by individuals themselves, the more productive the economy. Part of the reason for this is epistemology: “Far from being appropriate only to comparatively simple conditions, it is the very complexity of the division of labor under modern conditions which makes competition the only method by which such coordination can be adequately brought about.”24 A minimal condition for democracy is political competition.25 An opposition has an incentive to expose corruption as well as policy failures.26 One need not accept the free market implications of the notion of the market as a spontaneous organization to recognize the application of this idea to democracy.27 In like manner, citizens pursuing their self-interest can produce collective choices for governing themselves that work better than choices imposed on them by the technocratic elite. 23 Adam Smith, 454. 24 Hayek, 95. 25 I consider contested elections a minimal condition for democracy. Thus, where there are elections but only one viable candidate, or where a single party tends to dominate electoral results and competition for the party’s nomination is minimal as undemocratic. It should not be surprising that under such conditions we would find corruption and other indications of poor policy performance. 26 Shapiro 2003, 63. 27 Hayek himself, like many free market economics, disavows the democratic implications of his free market economics when democracies vote to regulate markets. For instance, see The Road to Serfdom 109-111 and The Constitution of Liberty 176. 16 In seeing democracy as a self-organizing system, it is important to see it as a political system, not just how government is organized. 28 That is, political actors under democracy need not be directly involved with government. They need not even vote. The distinction between democracy and authoritarianism is that in democracy, citizens act even as they are acted upon. Furthermore, boundaries between ruler and ruled in democracy are porous or non-existent; they are contingent. In some respects, then, democracy is spontaneously organized like science or a free market economy.29 Democracy however is distinctly cooperative, and benefits from a kind of division of labor in acquiring knowledge.30 Individuals who possess the knowledge can spontaneously share it and reap epistemological advantages.31 The self-organizing nature of democracy can also be recognized in ways in which parties and other organizations respond to various issues that emerge. Grass roots mobilization frequently begins with demands that a problem be recognized. In policy terms, agenda setting ought to be democratic so the most urgent problems become part of the political landscape. Parties and other political entrepreneurs similarly compete for voters and public support in part through proposing responses to these concerns. Organizations and ideas that fail to resonate in the public mind fall by the wayside. All of this happens without any central authority organizing it. That is, it happens spontaneously. 28 diZerega 756. 29 diZerega 764. 30 Bohman 603-4. 31 Estlund, 177. 17 Democratic Practice The potential of democratic problem solving is not restricted to abstract and speculative theory, but can be observed in political practice. The accounts that follow reflect diverse approaches to the study of politics and uncover different kinds of ways that people can govern themselves competently, if not wisely. In evaluating democracy, it is important to keep in mind the standard against which it should be judged is not perfection, and no one reasonably expects citizens to be omnicompetent in all, or even a few areas of policy importance. Furthermore, evaluating actual democracy against one composed of a small number of homogenous and deeply committed citizens fails to recognize the advantages of real democracy. In The Rational Public, Page and Shapiro describe how despite the many shortcomings of individual knowledge, collective public opinion is stable, presents real policy preferences, and responds to changes in circumstances in predictable ways. Moreover, given good information, the public’s preferences will in fact result in good public policy.32 It does so in part because the collective cancels out random errors in individual policy preferences. If say 1 percent of the public holds the correct position on a given policy, and the rest of the public chooses their position randomly, then 50.5% of the people will hold the correct opinion. However, in practice, there are all kinds of ways errors could be systematic and biased. Thus, the wisdom of the crowd will also be systematically biased. Real world democracies can overcome systematic error over time, because as unfavorable results come in, public opinion shifts away from the systematic error and the public 32 Page and Shapiro 1992, 17. 18 corrects itself. This does require a functioning democracy—Page and Shapiro lay the blame for the dysfunction of democratic governance not on the defects of the people but on the elites.33 Manipulation and propaganda can influence public opinion in ways which diminish its capacity for government. But there is no reason to believe that elites would fare any better. There is no doubt that information needs to be widely dispersed and circulated in order for democracy to result in wise policy. However, there are limits to the effectiveness of propaganda and manipulation, at least in an open society. A functioning democracy presents a further challenge to propaganda and manipulation, because in addition to the multiple sources of information available given freedom of expression, and opposition party has an incentive to provide information to counter the information presented by the party in power. Of course, there are no guarantees—both the party in power and the opposition may be equally deceptive. That limits the effectiveness of democracy.34 But ultimately, the results of the policy alternatives will limit the effectiveness of propaganda. An example of this might be US public opinion towards the Iraq war. Public opinion was initially favorable to the Iraq war, arguably because it had been systematically misled by propaganda (regarding both the reason for the war and the promise that the war would be brief and inexpensive like the first Gulf War). In addition, elites widely believed Saddam Hussein was stockpiling weapons of mass destruction, and Hussein himself seemed to act as if he had them. Nevertheless, it soon became clear that Iraq had no such weapons. Opinion toward the war turned negative as it became clear the war would not deliver the 33 Page and Shapiro 1992, 387. 34 Page and Shapiro 1993 62. 19 promised outcome and the cost was far higher than the Bush administration promised. Had democracy prevailed, the US would have withdrawn from Iraq earlier than it did. Had the public had better access to information, the US may never have started the war in the first place. A similar pattern can be found with public support for the war in Viet Nam. Public opinion shifted decisively against the war with the Tet Offensive in 1968.35 Of course, it is precisely this tendency of public support for military activity to wane as war drags on that many people cite as a reason for curtailing the influence of public opinion on policy. This was certainly a central part of Walter Lippmann’s criticism of democracy in The Public Philosophy. Lippmann asserted that the wars of the first half of the twentieth century illustrate the failure of democracy.36 However, even if Lippmann’s analysis is correct (and it seems to me that most of the responsibility for the horrors of that era can be laid at the feet of the elites who marched Europe towards disaster in the First and Second World Wars) the evaluation of the performance of democracy cannot be made in a vacuum. It has to be compared to something. Perhaps one could argue that it was right for the Union to persist in fighting to preserve the Union during the US Civil War. But history is unlikely to look so favorably on elite persistence in defiance of opposition in the Viet Nam war or the recent wars in Iraq and Afghanistan. Elites in support of those actions insist that continuing to fight is justified by the fact that so much has already been sacrificed—more than a sunk costs fallacy, but a tragedy. Indeed, it seems that democracies are far better than elites in avoiding sunk cost fallacies at least when it comes to war. Sometimes the rational thing to do is abandon a project. Finally, how well public opinion towards the Vietnam War or the Iraq War demonstrates democratic competence depends a great deal on perspective. At 35 Page and Shapiro 1993, 50. 36 Lippmann, 8-13. 20 least in the case of Iraq, public opinion at first embraced a dubious project and only turned against the war after much blood was shed. On what basis should we judge the performance of public opinion? Did the public learn fast enough? And compared to whom? The answer to these questions are in the end moral judgments. One key to the effectiveness of democratic governance stems from the unique information available to ordinary people. In addition, democracy provides incentives to the government to provide effective policy. These two principles lie at the heart of Amartya Sen’s finding that famine does not occur in democracies.37 The role of democracy in Sen’s analysis is limited to providing information to bureaucrats. Sen does not propose that democratic processes could replace technocrats. But he does argue persuasively that democracy can be essential to prevent disasters that emerge from misguided policies. That is, political incentives are at least as important as economic incentives in delivering effective government policy. It seems appropriate given that we are in New Orleans to note that a similar kind of analysis could be applied to understand what went wrong (and perhaps what went right) with the government response during Katrina. Famines need not be complex, and the ways in which democracy provides some protection from famine itself does not suggest that democracy is most suited to complex societies. The central point from the perspective of this paper is the way in which democracies provide essential information to policymakers. This assumes, as Sen notes, that democracy includes a free press. In addition, Sen does not go into specifics regarding democratic procedure. All that is necessary is that information moves freely and that 37 Sen, 180-181. 21 government officials have incentives to pay attention to public opinion. As complexity increases, the role of information on the ground and its transmission to those in a position to do something becomes even more important. Since complexity makes predicting policy outcomes more difficult, it is necessary for policy makers to pay attention to the information provided by democratic mechanisms. Even if the recent financial crisis stems from an economic system so complex that even experts do not understand it, and ordinary people are in the dark about the complexities of the financial transactions that swirl around there head, there is still a parallel with basic information concerning mortgages and inflated real estate prices and conditions on the ground during a famine. The real problem with the foreclosure crisis was the ineffectiveness of incentives in preventing bad loans. Thus, it may make sense to implement procedures specifically designed to inform the policy making process. Such procedures would include active and ongoing consultation with affected populations as well as incentives for policy makers to use that information. At a minimum, these policies would involve consulting with people affected by the policy at each step in the policy making process. Something even more profound lies at the heart of Elinor Ostrom’s findings regarding self-government’s potential in addressing on of the most elusive problems in public policy, governing the commons. Ostrom’s work shows that self-government can sometimes solve challenging policy problems, specifically common pool resources problems. Resources held in common, that is not owned by any particular individual, are often subject to depletion because it is rational for each person to take as much of the resource as possible. This is widely known as the tragedy of the commons. In many cases, assigning individual ownership in such situations is difficult or impractical. Examples 22 include fisheries, grazing areas, forests, and irrigation systems. The common solution to these problems involve privatization of the resource, which is intended to give owners an incentive to preserve the resource, and regulation, which uses various coercive measures to limit exploitation of the resource. Democratic management presents an alternative, and Ostrom cites examples of how this works. Similarly to Sen, local knowledge is essential, and accounts for superiority of community management over central government regulation.38 Also, similarly to Sen, the way in which incentives are arranged is critical. One critical element in Ostrom’s cases is that the institutions in question have evolved over time and adapted to changing circumstances.39 That is, they are self- organizing. Ostrom herself is rather cautious about applying insights from her examples to other situations. Her cases seem to depend on the existence of particular circumstances, and attempts to implement similar models elsewhere would need to proceed cautiously. Most specifically, Ostrom predicts that self-government solutions to common pool resource problems will most likely be successful if the community shares common judgments regarding the unsustainability of current practice, changes in rules will affect members similarly with no one gaining or losing a great deal, exit is not viable strategy, and the group is small and socially cohesive.40 It goes without saying that these kinds of solutions should not be imposed in a top down fashion. However, it does provide a powerful argument that central regulators and other outsiders ought to leave functioning systems alone. More importantly, her work shows that self-governing institutions can arise spontaneously and handle complex resource management tasks. 38 Ostrom 20. 39 Ostrom 58. 40 Ostrom 211. 23 Both Sen and Ostrom observe people who are more or less motivated by self- interest rather than civic virtue or some other high ideal. The other commonality is the way in which these structures take advantage of local knowledge. These conditions for citizen competence are not rare, nor do they require an extraordinary degree of education or even commitment to an abstract notion of a common good. That is not to say that they will work at any time under any set of circumstances. What it does suggest though that the conditions for successful self-government are less particular than many political scientists think.41 Furthermore, they both provide reasons why self-government works when other systems of government will not. In the cases described above, it is important to recognize that democracy or self-government is not a panacea. However, the empirical evidence demonstrates that democracies can perform quite competently in a range of different environments and under many different circumstances. Understanding the importance of how incentives work--something all too often missing from much democratic theory—is also central to recognizing why democracy results in effective government. Linking these findings to those of public opinion discussed at the beginning of this section suggests that tying public policy more closely to public opinion may, at least under certain circumstances, have beneficial effects. Real Limits to Democratic Competence Though much of literature concerning democratic competence misses the mark by either overstating the ability of experts or evaluating the citizen competence in 41 South Africa presents another interesting case. See Shapiro 2011, 80-81. 24 inappropriate ways, it would be silly to pretend there are not real limits in the ability of the people themselves to address policy challenge that are especially relevant in a complex world. In addition to practical problems of implementation, two particular challenges are posed by the cost of good information and the limits of policy experimentation. In addition, delegation seems to be inevitable, and thus how democracies are to delegate authority to experts and yet remain democratic also needs to be addressed. What follows in this section is far from a complete account of all the limits to democratic competence and is little more than a sketch that suggests other possible responses to real limits. It remains important to distinguish between real limits to democratic governance and irrelevant limitations. Information costs Even those theories of democracy that hold people are capable of forming reasonable opinions about policy recognize that those reasonable views depend on access to good information. Critics have no problem demonstrating that such information is not readily available and even where it is, it competes with all manner of propaganda for attention. There is also no reason to believe that good information is significantly more “believable” than bad information. Also, there is substantial evidence that people are not only more willing to believe claims that confirm their prejudices than true claims; they are also motivated to seek out claims that confirm their prejudices. Research on cognition may present further challenges to democratic competence. Again, though it is important to note that experts can suffer from their own cognitive shortcomings.42 42 See Tetlock. 25 A frequent criticism of the potential of democratic processes yielding effective policy outcomes stems from the idea of rational ignorance. The basic idea is that the marginal utility of being well-informed about public policy is effectively zero. However, in some respects, staying ignorant is impossible—some information is unavoidable. Other information is available at very low cost. Information technology brings down the cost even further in some ways. The strength of a self-organizing system is that people will not need to seek political knowledge. They will encounter it as a matter of course. Indeed, they will have access to information unavailable to anyone else. Conditions of complexity mean that at least as far as their own condition is concerned, people have the best information of their individual specific circumstances. There is no other source that comes close to it. Such an observation is beyond obvious perhaps, except that it is precisely in their individual conditions that complexity reigns. That is, democracy overcomes the problem of complexity of individual conditions through aggregation. The aggregation of individual assessments of their own condition taken from the ground is bound to be more reliable than elite assessments taken from the mountaintop. This same mechanism functions at the policy evaluation stage. As a policy is implemented, the only test of its effectiveness is how it improves the lives of individuals. One of the cornerstones of liberalism has been an the individual’s exclusive ability understand his own interest: “If a person possesses any tolerable amount of common sense, his own mode of laying out his existence is best, not because it is best in itself, but because it is his own mode.”43 The idea that individual are best at assessing their own 43 Mill  1988, 135 26 condition is a necessary condition for this account of liberty to make sense. However, the idea democracy also rests on the principle when applied to groups of people. Much is made of the difference between individual perceptions and collective reality. There is a difference between the aggregation of individual evaluations of circumstances and what is actually happening. Individual evaluations are based on mediated impressions. The public’s over-estimation of crime is a well-known example. People tend to perceive a more dangerous world not so much because of their experience of crime but rather because they have been exposed to more news accounts of ever more horrific crimes. Such accounts may be better understood as limitations of individual assessment of global conditions. Such errors are less likely at the individual level. Information sharing among individuals can help lift the fog of war. More importantly, aggregating individual assessments may be limited at identifying critical needs that exist among a minority of the population. It may be necessary to frame assessments properly. Most people in the United States have health insurance, and are not facing a catastrophic illness. Thus, from their perspective, there really is no crisis that they are experiencing at an individual level. However, if the issue is framed as individual vulnerability, the numbers will surely become rather different. Most people stand to lose health insurance if they lose employment, and anyone may suffer a catastrophic illness. Recognizing vulnerability then seems well within the capability of most citizens. Trial and Error Another limitation to the self-organizing concept of democracy is that it works largely on the basis of trial and error. Strictly speaking, a self-organized democracy would 27 implement policies which would then be abandoned or modified randomly until a satisfactory solution is found. The scale of time in evolution is billions of years, a time frame that is politically unacceptable. While self-organizing systems in biology occur through random changes, self-organizing social systems do not. Rather, individuals who make decisions are guided by their own intelligence. A near infinite number of approaches to public problems will be excluded because they are ridiculous or because no one imagines them. Thus, trial and error will involve a limited number of plausible solutions. Even so, it may be far too slow a process to deal with any public policy. However, not all policy challenges are amenable to this solution. In some cases, experts really do know better than the public what to do in a given situation. It would be useful if a reliable way of determining when experts really know best could be determined. There has been some provocative research in this direction. For instance it appears that environments that are regular and one where experts can develop a great deal of experience will lend itself well to expert analysis.44 On the opposite end of the spectrum, crises that demand an immediate response seem to require a non-democratic response. Even in the crisis situation, democratic responsiveness can vastly improve the effectiveness of the intervention. In novel, non-crisis situations, however, incremental approaches may take advantage of some of the elements of trial and error. Second, policy experimentation can often occur simultaneously. One alleged strength of federalism in the United States is that it provides “laboratories of democracy” where different localities can use different approaches to solve policy problems and other localities can adopt those that appear successful. An emphasis on local problem solving where possible can take advantage of the 44 Kahneman, 240. 28 strengths of democratic problem solving. When the problem at hand was a global problem such as global warming or climate change any policy choice will be irrevocable. What is done cannot be undone. Such situations encourage a bias towards incrementalism, which would allow democrat feedback that might limit the damage stemming from inappropriate policies. Delegation Almost any imaginable political system in the contemporary world will require delegation. Even if a public is committed and energetic, and the community simple and small enough for a town meeting system to handle legislative tasks, the executive functions will still need delegation. Both representatives and technocrats are delegates of a sort. Though the problems of representation for democracy have substantial relevance for the subject of this paper, most of my emphasis here has been on the problem of technocrats. As long as elected representatives are responsive to voters, representative democracy can perform as a self-organizing system. Current political trends in the US and Europe indicate a decline in responsiveness.45 Though no certain conclusion can be drawn, this suggests that the decline in democratic responsiveness has been a cause of recent political dysfunction in the US and Europe. A pluralist system may provide adequate avenues for democratic participation. In short, representative democracy is democracy. In addition, there is no necessary asymmetry between representatives and those who elect them—they are ordinary people who represent other ordinary people—in theory at least, nearly any citizen has the basic qualifications to serve as a representative. Technocrats, however, 45 Fiorina, 24-48. 29 because of their expertise, are different. But some authority will have to be delegated to them. Given the asymmetry between the knowledge of experts and the knowledge of ordinary people, technocrats may be able to exercise power of those they are supposed to serve. The limitations strongly suggest that there are important arenas of policy where it is appropriate for a democracy to delegate its authority to experts. Delegation can be done in a way that preserves democracy however, as long as decisions regarding delegation are made democratically and the people reserve the right to revoke or overrule the delegates.46 The argument of this paper suggest that the ability of ordinary people to overrule the technocrats should be extensive in order to preserve democracy, and that such an extensive ability does not threaten the effectiveness or efficiency of government. The challenge remains of how ordinary people are to evaluate the competence of the experts who serve them. Results are imperfect, as one needs to measure results against reasonable expectations. However, in order to serve the people, the onus is on the experts to persuade the people of their competence. Conclusion: Evolution, Markets, and Justice Problems of complexity are not unique to politics and policy. More often than not, the solution to these problems is non-hierarchical. Evolution, for one, has addressed (“solved” would imply a degree of stasis inapplicable to the kinds of problems described here) the problem of survival through undirected or non-hierarchical methods. The link 46 Robert Dahl, 114. 30 between complexity and evolution is to understand evolutionary processes as computational. That is, the process yields a solution to a problem. This is easy to see in economics—markets solve the problem of resource allocation. Democracy solves the problem of allocation of authoritative values or more simply who gets what. Its advantages in complexity make it possible to incorporate and “process” a great deal of information from a wide number of sources. The possibility this paper explores is that of democracy is a computational process that aggregates not just individual preferences, but processes information in order to develop an effective public response to public problems. The main reason for looking at this possibility is that democracies function at least as well as any system of government, despite the apparent ignorance and foolishness of individuals that make up this system. A problem with all this is that political behavior is deliberate rather than automatic. Individual behavior is not computational. People act politically with a kind of intention that differs from evolutionary processes in nature. There is also a difference between how human beings approach their role as an economic actor and their role as a political actor. Specifically, in nature and in markets, individuals are pursuing individual goals—survival or the accumulation of resources—and some kind of spontaneous order that was not anyone’s intention results. In politics, by contrast, people are often pursuing collective goals—the order that results is often precisely the point of political activity. Furthermore, the psychology of political behavior (and perhaps its biological and neurological foundations) is incompletely understood. Nevertheless, there still seem to be reasons to consider democracy at least as a related process. In the first case, it is still a process that involves the transmission of complex information. Doing so of course means going beyond 31 voting as a form of political participation. Indeed, in this respect, voting may not be considered all that democratic—choosing among a very limited set of options (yes or no; party a, b, or c) seems rather limited. However, the competition for votes is another matter altogether. From this perspective, there is a resemblance to formal models of democracy. Care must be taken to avoid the reductionism common in formal modeling to maintain the advantage of understanding democracy as a process of managing and processing complex information. Most importantly, in political activity, there is an exercise of authority and coercion. Political decisions are not always, perhaps not even usually, about self-interest. They are about many things—coercion, moral judgment, and identity to name but a few. Thus, politics does not just allocate goods, it also determines what is good. The history of political philosophy shows that there is no agreement about what is good, what is just, among elites. Liberal democracy is a solution to this problem that has distinct advantages. It is not purely procedural, because it entails a commitment to liberal goods. But liberal goods, as Mill points out, allow for a wide range of individual choices in how they pursue their lives. The account of democracy presented here has some important implications for democratic practice. As Adam Smith and F.A. Hayek among others argued that markets require their participants to embrace important mores to function well, so it is for democracy as well. Constitutionalism is one of many ways democracies regulate themselves, though a constitution may be rigid enough, or implemented in such a way that it itself threatens 32 democracy.47 But commitment to a constitution implies a commitment to certain normative principles. More specifically to the issues presented here, citizens need to be able to access quality information and citizens also need to be able to communicate their preferences in meaningful ways through decision making processes. There is at least one other commitment democratic citizens need to make. Conventionally, democratic theory proceeds from a set of normative assumptions. That is, democracy is conceived as having something to do with legitimacy—principles like consent of the governed, political equality, and non-dominance are among the justifications of democracy. But its legitimacy also depends on its ability to generate effective policy.48 In addition, in many normative accounts of politics, democracy itself must be under the law in order to protect minority rights or prevent the abuse of power. In short, citizens under democracy need to be democrats. The account of democracy presented here is a procedural one—an evolutionary process is by its nature anti-teleological. But any account of democracy that preserves freedom is going to be procedural. Citizens participating in democracy must be free to decide for themselves what should be done. Creating an ideal citizen who puts the good of the public ahead of her own assumes in advance the direction politics should take and as such is not really democratic. In this regard, while there may be a tension between the individual liberty of liberalism and democracy, they are not incompatible. Constitutions that enshrine individual liberty may limit what democracies can do, but their force requires widespread acceptance of the principles they articulate. Popular acceptance and perhaps 47 The Citizens United decision by the Supreme Court is an example. The problem with the decision is not that a democratic decision was overturned, which may be consistent with constitutional democracy, but that the implications of the decision itself threaten democratic practice. 48 Estlund, 8. 33 even enthusiasm for the principles of liberal democracy may seem to be a sandy foundation for liberty and democracy. However, Adam Smith and F.A. Hayek recognized that functioning free markets ultimately rested on similar foundations. Corruption then remains an ever present danger. No democracy is immune from serious error, and it may be that certain cultural or institutional arrangements make a democracy more or less susceptible to error. Over time, as a population gains experience in governing itself, the results from its democracy may improve. Linking democracy to epistemology may seem to risk cutting the normative legs out from under what is widely perceived as one of the great moral achievements of recent times—the idea that all governments that aspire to justice also aspire to democracy. The question of whether citizens are competent enough for self-rule seems to be a different question from whether people have the right to govern themselves. Further, basing democracy on citizen competence rather than rights subverts the moral case for democracy.49 Historically, however, the embrace of democracy in part has its roots in the success of democracy. And that success requires some explanation. Furthermore, if democracies regularly produced undesirable results, the moral claim would have no standing no matter how elegant and persuasive the claim might be. Democracy would cease to be a political theory and would become little more than a utopian ideal. Moreover, appeals to justice do little or nothing to address the paradoxes of democracy and the attempt to exclude democratic procedures from ever wider areas of policy. 49 Dahl, 126. 34 Knowing why democracy works is important in order to understand its limits and its potential as well as how to address its deficiencies and failures. If a theory is to be political, it must have political relevance. Some of the limits are discussed above; there are certainly others. Understanding those limits may provide insight in how to address those. Information flows and belief formation are essential elements in the functioning of any real democracy. It will be interesting to see how innovations in information technology and social media influence democratic behavior. In any event, those who are committed to democracy ought to seek to understand how it really works before they consider how it might be made to work. Works Cited Bobbio, Norberto. Which Socialism. Edited by Richard Bellamy. Translated by Roger Griffin. Minneapolis, MN: University of Minnesota Press, 1987. Bohman, James. "Democracy as Inquiry, Inquiry as Democratic: Pragmatism, Social Sciences, and the Cognitive Division of Labor." American Journal of Political Science, 1999: 590-607. Dahl, Robert. A Preface to Democratic Theory. Chicago: University of Chicago Press, 1956. —. Democracy and its Critics. New Haven, CT: Yale University Press, 1989. diZerega, Gus. "Liberalism, Democracy, and the State: Reclaiming the Unity of Liberal Politics." The Review of Politics, 2001: 755-782. Dunn, John. Western Political Theory in the Face of the Future. New York: Cambridge University Press, 1979. Estlund, David M. Democratic Authority: A Philosophical Framework. Princeton: Princeton University Press, 2008. Fiorina, Morris. Disconnect. Normon, OK: University of Oklahoma Press, 2009. Gaus, Gerald. "Is the Public Incompetent? Compared to Whom? About What? ." Critical Review, 2008: 291-311. Hayek, F.A. The Constitution of Liberty. Edited by Ronald Hamowy. Chicago: University of Chicago Press, 2011 . 35 —. The Road to Serfdom. Edited by Bruce Caldwell. Chicago: University of Chicago Press, 2007 . Hochschild, Jennifer. "Disjunction and Ambivalence in Citizens' Political Outlooks." In Reconsidering the Democratic Public, by George E. Marcus and Russell L. Hanson, 187-210. University Park, PA: Penn State University Press, 1993. Kahneman, Daniel. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux, 2011. Little, Adrian. "Political Action, Error and Fialure: The Epistemological Limits of Complexity." Political Studies, 2012: 3-19. Mill, J.S. "On Liberty." In Utilitarianism, On Liberty, and Considerations on Representative Government, by J.S. Mill, edited by H. B. Acton, 69-185. London: Everyman's Library, [1859 ]1988. Mitchell, Melanie. Complexity: A Guided Tour. New York: Oxford University Press, 2009. Moon, J. Donald. "Theory, Citizenship, and Democracy." In Reconsidering the Democratic Public, by George E. Marcus and Russell L Hanson, 211-222. University Park, PA: Penn State University Press, 1993. Ostrom, Elinor. Governing the Commons: The Evolution of Instituions for Collective Action. New York: Cambridge University Press, 2008. Popper, Karl. "Public Opinion and Liberal Principles." In Conjectures and Refutations, by Karl Popper, 467-476. New York: Routledge, 2002. Sanderson, Ian. "Intelligent Policy Making for a Complex World: Pragmatism, Evidence, and Learning." Political Studies, 2009: 699-719. Sen, Amartya. Development as Freedom. New York: Alfred A. Knopf, 1999. Shapiro, Benjamin I. Page and Robert Y. The Rational Public: Fifty Years of Trends in Americans' Policy Preferences. Chicago: University of Chicago Press, 1992. Shapiro, Ian. The Real World of Democratic Theory. Princeton: Princeton University Press, 2011 . —. The State of Democratic Theory. Princeton: Princeton University Press, 2003. Smith, Adam. An Inquiry into the Nature and the Causes of the Wealth of Nations. Edited by R.H. Campbell A.S. Skinner and W.B.Todd. Vol. I. II vols. Indianapolis, IN: Liberty Fund, 1981. Sniderman, Paul M., Joseph F. Flethcer, Peter H. Russell, Philip E. Tetlock, and Brian J. Gaines. n.d. Sniderman, Paul M., Michael G. Hagen, Philip E. Tetlock, and Henry E. Brady. "Reasoning Chains: Causal Models of Policy Reasoning in Mass Publics." British Journal of Political Science, 1986: 405-430. 36 Tetlock, Philip E. Expert Political Judgment: How Good Is It? How Can We Know? Princeton: Princeton University Press, 2005. Woodhouse, A.S.P., ed. Puritanism and Liberty. London: J.M. Dent and Sons Ltd, 1992. Zolo, Danilo. Democracy and Complexity: A Realist Approach. Translated by David McKie. University Park, PA: Penn State University Press, 1992.