Preventing Extremism

Excerpt from Going to Extremes - How Like Minds Unite and Divide by Cass R. Sunstein

Preventing Extremism

Suppose that a nation seeks to prevent (unjustified) extremism. What might it do? I consider three possible answers here. The first, favored by many conservatives, is traditionalism. The second involves consequentialism. The third, prominent in the American founding, involves checks and balances; a system of freedom of expression and a diversity of views are part and parcel of that idea. We can associate these three responses with three people: Edmund Burke, Jeremy Bentham, and James Madison. I devote the most attention here to checks and balances, because that approach seems most interesting and most productive both in ordinary life and in the political domain.

TRADITIONALISM

For those who are concerned about extremism, it is natural to focus on Burke’s great essay on the French Revolution, which can be understood as a sustained warning about the very processes I am exploring here. With his emphasis on the need to attend to traditions, Burke displayed a keen interest in popular passions and the risk that a group of people, stirred up by each other and one or another idea, can go in extreme directions. 

Burke’s central claim is that the “science of constructing a commonwealth, or reforming it, is, like every other experimental science, not to be taught a priori.”1 To make this argument, Burke opposes theories and abstractions, developed by individual minds, to traditions, built up by many minds over long periods. In his most vivid passage, Burke writes:

We wished at the period of the Revolution, and do now wish, to derive all we possess as an inheritance from our forefathers. . . . The science of government being therefore so practical in itself, and intended for such practical purposes, a matter which requires experience, and even more experience than any person can gain in his whole life, however sagacious and observing he may be, it is with infinite caution than any man ought to venture upon pulling down an edifice which has answered in any tolerable degree, for ages the common purposes of society, or on building it up again, without having models and patterns of approved utility before his eyes.2

Thus Burke stresses the need to rely on experience and, in particular, the experience of generations. He objects to “pulling down an edifice,” a metaphor capturing the objection to passionate movements that start social or political life from the ground up. It is for this reason that Burke describes the “spirit of innovation” as “the result of a selfish temper and confined views.”3 It is for the same reason that Burke offers the term “prejudice” as one of enthusiastic approval, noting that “instead of casting away all our old prejudices, we cherish them to a very considerable degree.”4

Why, exactly, would prejudices appeal to Burke? The word itself supplies an answer. Prejudices operate before judgment they supply answers that antedate reflection, including those forms of reflection that can be found and spread among group members. If prejudices are rooted in long-standing practices, it should not be surprising to find that Burke trusts them. Emphasizing the critical importance of stability, Burke adds a reference to “the evils of inconstancy and versatility, ten thousand times worse than those of obstinacy and the blindest prejudice.”5

Burke’s basic claim, then, is that reasonable citizens, aware of their own limitations, will effectively delegate a good deal of authority to their own traditions. “We are afraid to put men to live and trade each on his own private stock of reason” because of the concern that any one person’s stock “is small, and that the individuals would do better to avail themselves of the general bank and capital of nations, and of ages. Many of our men of speculation, instead of exploding general prejudices, employ their sagacity to discover the latent wisdom which prevails in them.”6

Traditionalism can certainly operate as a check on extreme movements. If group members have respect for what has been done before, they are unlikely to reject it in favor of what emerges from internal discussions. Those who respect traditions are most unlikely to polarize. But as a full response to the problems I have sketched here, traditionalism runs into serious difficulties. Traditions may persist not because they are good, but because of the same sorts of social influences that produce group polarization. After all, traditions convey information, and people may follow them not because they believe,  independently, that they are good, but because other people (appear to have) believed that they are good. Thus people may follow traditions even though they have independent reasons to believe that those traditions are bad. In experimental settings, there is good evidence that people will follow even arbitrary traditions, failing to bring their own judgments to bear.7 The basic problem is that traditions convey information, and people may think that the information is more helpful, or more revealing, than it actually is.

In a separate phenomenon, people may know that traditions are bad and have little confidence in the judgments of the past, but nonetheless may adhere to past practices as a result of reputational pressures ranging from disapproval to ostracism to punishments both formal and informal. Those who deviate from traditions often face serious social sanctions. 8 In any event, there is no reason, in the abstract, to think that long-standing practices are always better than the alternatives proposed by reformers or even  extremists.9

After all, the American Revolution was not a traditionalist one. Federalist Paper No. 1 offers a direct challenge to Burkeanism with the suggestion: “It has been frequently remarked that it seems to have been reserved to the people of this country, by their conduct and example, to decide the important question, whether societies of men are really capable or not of establishing good government from reflection and choice, or whether they are forever destined to depend for their political constitutions on accident and force.” The opposition between “reflection and choice,” on the one hand, and “accident and force,” on the other, suggests a sharp critique of those who value traditions as repositories of wisdom.

Consider, too, the words of James Madison, writing in a very young America:

Is it not the glory of the people of America that, whilst they have paid a decent regard to the opinions of former times and other nations, they have not suffered a blind veneration for antiquity, for custom, or for names, to overrule the suggestions of their own good sense, the knowledge of their own situation, and the lessons of their own experience?10

In Madison’s unmistakably antitraditionalist account, Americans “accomplished a revolution which has no parallel in the annals of human society. They reared the fabrics of governments which have no model on the face of the globe.”11

These are largely rhetorical passages, but there is actually an argument in the background, one that turns Burkeanism on its head. Thomas Jefferson himself captured that argument with his objection that some people “ascribe to the men of the preceding age a wisdom more than human.” His response, a kind of rebuke, was that the age of the founders “was very like the present, but without the experience of the present; and forty years of experience in government is worth a century of book-reading.”12 Burkeans tend to cherish the wisdom of those long dead, but in Jefferson’s view, their stock of wisdom was far more limited than ours. In the same vein, Pascal contended that we are the ancients: “Those whom we call ancient were really new in all things, and properly constituted the infancy of mankind; and as we have joined to their knowledge the experience of the centuries which have followed them, it is in ourselves that we should find this antiquity that we revere in others.”13

Jeremy Bentham attacked ancient wisdom in identical terms, contending that those who are ancient are, in the relevant sense, very young.14 Bentham acknowledged that old people have more experience than young people, but he insisted that “as between generation and generation, the reverse of this is true.”15 In fact, “the wisdom of the times called old” is “the wisdom of the cradle.”16 Bentham deplored the “reigning prejudice in favor of the dead” and also the tendency to disparage the present generation, which has a greater stock of knowledge than “untaught, inexperienced generations.” Consider in this regard attacks on established practices in such areas as discrimination on the basis of race, sex, and disability. As we have seen, the movements that led to these attacks benefited greatly from group polarization. In many domains, traditionalism is an important constraint on unjustified movements, but too often, traditions are themselves a product of the same sorts of forces that lead people in bad directions.

CONSEQUENCES

A separate approach to the risks associated with group polarization is to call for careful investigation of the consequences of one or another course of action. It is important to see that this approach, associated with Bentham, may or may not check extremism. In the end, the investigation of consequences might support rather than undermine the argument for dramatic steps. But there is nonetheless an important constraint here a kind of reality check. Consequentialism promises to discipline extreme movements or, for that matter, enthusiasm for the status quo by subjecting them to certain tests. The consequentialist hope is that the necessary investigation can dampen social controversy and the risks associated with group polarization by asking people to engage with the facts.17

Suppose, for example, that people in a certain neighborhood are becoming frightened about a particular risk, such as crime, pesticides, or electromagnetic fields. If they are stoking one another’s fears, it might be useful to ask: Are people really at serious risk? What, exactly, do we know about the size of the risk, and what do we know about the burdens of trying to reduce it? Questions of this kind can prevent a situation in which echo chambers serve to produce unjustified fear of risks that really do not deserve concern.

In the context of government action, an important example involves debates over the role of deterrence in criminal law. Many people think that we should be asking serious questions about the deterrent effect of certain punishments the death penalty, restrictions on gun ownership, life imprisonment for those who have committed three felonies. Their hope is that if we ask about such issues, we will be able to reduce the intensity of certain conflicts by engaging closely with the facts.18 If we can agree that the argument for gun control must stand or fall on whether it is effective in reducing crime, then perhaps we can quiet some of our disagreements on that issue. If the death penalty does not deter crime, perhaps we can agree to abolish the death penalty. At the very least, that is the hope of those who seek to put a spotlight on the question of deterrence.

In the regulatory context, where the democratic process is often badly polarized, we can find many examples. Consider, for example, the regulation of arsenic in drinking water. Some people are not concerned by very small levels of arsenic exposure say, 50 parts per billion. They believe that such small levels cannot possibly deserve social concern or serious sums of money. They think that the government should pay attention to more serious problems. Other people think that arsenic is a poison that is toxic even at low doses, and they believe that the permissible level should be very low say, 3 parts per billion. We can easily imagine a situation in which the disagreement becomes very intense, as like-minded people talk mostly to one another, stirring themselves up into real antagonism. And in fact, a battle of just this kind occurred over arsenic regulation in 2000.19

To those who care about consequences, the best way to proceed is to assess the actual effects of low levels of arsenic exposure and to see how much it would cost to reduce those effects. What is the risk of cancer at 50 parts per billion? Is it one in a million? One in 10,000? If the permissible level is reduced to 3 parts per million, how many lives will be saved per year? And how much will that reduction cost? Will people’s water bills increase significantly or, instead, trivially? Answers to these questions might also enable us to find some common ground. In the process, such answers seem to offer a kind of check on various dangers, including interest group power, confusion, and group polarization. If some groups are calling for no regulation, and others for aggressive regulation, an analysis of consequences might provide helpful, at least if it can restrict the range of reasonable disagreement. 20 If Boulder types insist that certain substances should be banned, and if Colorado Springs types insist that no regulation is the right approach, an investigation of actual effects might serve as an arbiter, or as a demonstration that neither side has it quite right. Consider in this light the highly polarized debate over climate change. Perhaps a careful investigation of the consequences both of greenhouse gas emissions and of efforts to reduce them can help a great deal.

In some domains, of course, disputes over consequences can reflect rather than dampen group polarization. Some groups might find, and argue, that the benefits of certain steps are low and that the costs are high; other groups might press the opposite conclusions. In the history of federal regulation, it is hardly unusual to see different sides invoking their view of consequences in favor of radically different conclusions. Consisting solely or mostly of like-minded types, some groups might conclude that the analysis clearly justifies some kind of regulation, whereas other groups might conclude that the same analysis justifies nothing at all. Disputes over climate change certainly reflect conflicts of exactly this sort.21 Group polarization is undoubtedly a contributor to those conflicts. But even in the domain of climate change, where so many questions remain open, an analysis of the consequences can impose a valuable discipline. It suggests, for example, that global inaction is quite indefensible, and also that there are limits to how aggressively we should cut emissions in the near future.22

Bitter debates over occupational safety, disability rights, national security, and affirmative action to say the least, polarizing issues might well be rendered somewhat less hot with a better understanding of the consequences of one approach or another. Of course, it is especially likely, in those domains, that polarization will be replicated, not diminished, as different groups attempt to assess consequences. But the great promise of the inquiry is that the range of reasonable disagreement can be narrowed and that when people take strong stands, it will be after engagement with actual effects, rather than solely a product of social interactions.

CHECKS AND BALANCES

A third answer, and perhaps the most valuable, emerges from an investigation of the American founding. To understand that answer, it is important to see that the founding period witnessed an extraordinary debate over the nature of republican institutions. What kinds of institutions would best suit the young nation? That question was debated in part by reference to the views of the political theorist Montesquieu, who was a revered source for all sides and a central figure in the development of the idea of separation of powers.

The antifederalists, eloquent opponents of the proposed constitution, complained that the document’s framers had betrayed Montesquieu by attempting to create a powerful central government, one that could not possibly manage a diverse nation. Brutus, an especially articulate antifederalist, spoke for the republican tradition when he urged: “In a republic, the manners, sentiments, and interests of the people should be similar. If this be not the case, there will be constant clashing of opinions; and the representatives of one part will be continually striving against those of the other.”23

Those who favored the constitution believed that Brutus had it exactly backward. Skeptical about uniformity of opinion, they affirmatively welcomed diversity, disagreement, and the “constant clashing of opinions.” They sought a situation in which “the representatives of one part will be continually striving against those of the other.” Alexander Hamilton spoke most clearly on the point, urging that the “differences of opinion, and the jarring of parties in [the legislative] department of the government . . . often promote deliberation and circumspection; and serve to check the excesses of the majority.”24 As the framers stressed, widespread error is likely to result when like-minded people, insulated from others, deliberate on their own. In their view, heterogeneity of view can be a protective force. A constitution that ensures the “jarring of parties” and “differences of opinion” can provide safeguards against unjustified movements of view.

More particularly, the institutions of our Constitution reflect an implicit fear of polarization, creating a range of checks on potentially ill-considered judgments. The most obvious example is bicameralism. The idea of a bicameral legislature was designed as a safeguard against a situation in which one house in the framers’ view, most likely the House of Representatives would be overcome by short term passions and even group polarization. In the founding period, the Senate was thought to be especially important in this regard. Consider the widely reported story that on his return from France, Thomas Jefferson called George Washington to account at the breakfast table for having agreed to a second chamber. “Why,” asked Washington, “did you pour that coffee into your saucer?” “To cool it,” quoth Jefferson. “Even so,” said Washington, “we pour legislation into the senatorial saucer to cool it.”25 James Wilson’s great lectures on law spoke of bicameralism very much in these terms, referring to “instances, in which the people have become the miserable victims of passions, operating on their government without restraint” and seeing a “single legislature” as prone to “sudden and violent fits of despotism, injustice, and cruelty.”26

We can understand many aspects of the system of checks and balances in the same general terms. The duty to present legislation to the president protects against polarization effects within the legislative branch.27 The opportunity for presidential veto supplements the system of bicameralism, further reducing the risk of hasty or ill-considered legislation. The very fact that the president cannot make law on his own, and must rely on Congress for authorization, creates a crucial safeguard against the potentially disastrous effects of group polarization within the executive branch. Compare dictatorships and tyrannies, which concentrate political power within a single branch, prone to grotesque error, in part because of polarization. And because law cannot operate against citizens without the concurrence of the legislative and executive branches, enacting and then enforcing the law, there is a further safeguard. 

Federalism itself was, and remains, an engine of diversity, creating “circuit breakers” in the form of a variety of sovereigns with separate cultures. In the federal system, social influences may produce error in some states, and states can certainly fall into cascades. But the existence of separate systems creates a check on the diffusion of error. In this respect, federalism permits states to restrain one another. A particularly important part of this process involves the right to exit. If one state oppresses its citizens, there is always the freedom to leave. That very freedom creates a before-the fact deterrent to oppressive legislation. It also creates an after-the-fact safeguard. In this sense, the right to travel, from one sovereign to another, is first and foremost a political right, akin to the right to vote itself. And if a form of group polarization occurs in one state, the federal system ensures that other states will come to different views. Here, too, we can find a safeguard of liberty.

An understanding of group polarization also casts fresh light on one of the most important and controversial provisions in the American Constitution: the grant of power to Congress, and not the president, to “declare war.”28 The debates in the framing period suggest a fear of two risks: The president might make war without sufficient authorization from the citizenry, and he might do so without sufficient deliberation and debate among diverse people. Thus Pinckney urged that the Senate “would be the best depository, being more acquainted with foreign affairs, and most capable of proper resolutions.”29 Butler, by contrast, sought to vest the power of war in the president, urging that he “will have all the requisite qualities, and will not make war but when the Nation will support it.”30 Madison and Gerry made the key compromise, suggesting that Congress should have the power to “declare” war. This provision was understood to permit the president “to repel sudden attacks.”31 But otherwise, the president would be required to seek congressional approval, in part on the theory that (in Mason’s words) this would amount “to clogging rather than facilitating war” and to “facilitating peace.”32

If warmaking is seen to be an especially grave act, we might be especially troubled about permitting the president to make war on his own. This is not at all because the president is immune from political checks. It is because group dynamics within the executive branch create a risk of polarization, as like-minded people push one another to indefensible extremes. We have seen that the executive branch is distinctly at risk here, at least to the extent that presidents create Teams of Unrivals who reinforce, and do not contest, the initial inclinations of their leaders. A requirement of congressional authorization ensures a check from another institution, with diverse voices and a degree of independence from the executive branch.

Of course, the system of free expression is fundamental to the process. Unjustified extremism, in government and outside it, can be met by opposing voices. The First Amendment has a checking function;33 it is part and parcel of checks and balances. Those who are tempted to adopt extreme positions, perhaps as a result of group polarization, can be countered by others who believe that their facts and their values are wrong, or at least are missing something important. So long as people are listening to one another, free speech should ensure that when extreme movements occur, it is not because dissenting voices have not been heard.

These points have significant lessons for how to combat risks of violence, including terrorism. If terrorism results from a process of radicalization in which like-minded people congregate, a response is to interject moderate or dissenting voices, preferably from trusted sources. In “Muslim enclaves,” the “battle is completely one-sided. The true believers, who populate the radical forums, are working themselves into a frenzy with no moderate voice present to calm them down. This leads to ever greater radicalization on the part of the participants, who slowly take on the views of their friends.”34 The countless “Muslims who reject violence need to enter this arena and participate in the discussions to influence and stop this slide toward ever greater radicalization.”35

DELIBERATIVE DEMOCRACY AND THE WISDOM OF CROWDS

In the last decades, a great deal of attention has been given to the idea of “deliberative democracy.”36 The central idea is that well-functioning democracies do not rely on snapshots of public opinion or on what most people think should be done. Instead, they attempt to combine deliberation and reason-giving with accountability. They insist that public action should follow a process in which people exchange information and ideas.

An understanding of group polarization raises some serious cautionary notes about deliberative democracy. It is foolish to celebrate deliberation as such or to believe that groups will arrive at truth or even sense. The exchange of information and ideas can and does breed unjustified extremism. Groups often blunder precisely because they put a high premium on deliberation.

Consider in this light the recent enthusiasm for the wisdom of crowds.37 The key claim here is that if you take the median or average view of a group of people, that view will often be stunningly accurate far more accurate than the view of the overwhelming majority of group members. The accuracy of judgments of statistical groups can often be explained by reference to the Condorcet Jury Theorem, which offers one of the most interesting results in modern social theory.38 To see how the Jury Theorem works, suppose that people are answering the same question with two possible answers, one false and one true. Assume, too, that the probability that each voter will answer correctly exceeds 50 percent. The Jury Theorem says that the probability of a correct answer, by a majority of the group, increases toward 100 percent as the size of the group increases. The key point is that groups will do better than individuals, and big groups better than little ones, so long as two conditions are met: Majority rule is used, and each person is more likely than not to be correct. 

The Jury Theorem is based on some fairly simple arithmetic. Suppose, for example, that there is a three-person group in which each member has a 67 percent probability of being right. The probability that a majority vote will produce the correct answer is 74 percent.39 As the size of the group increases, this probability increases, too. It should be clear that as the likelihood of a correct answer by individual members increases, the likelihood of a correct answer by the group increases, too, at least if majority rule is used. If group members are 80 percent likely to be right, and if the group contains ten or more people, the probability of a correct answer by the majority is overwhelmingly high very close to 100 percent.

The Jury Theorem suggests that large groups can often be wise indeed. But there is a dark side to the Jury Theorem, and it, too, has important implications. Suppose that each individual in a group is more likely to be wrong than right. If so, the likelihood that the group’s majority will decide correctly falls to zero as the size of the group increases!

Imagine that an organization consists of a number of people, each of whom is at least 51 percent likely to be mistaken. The organization might be a political party, a religious group, a university faculty, or a terrorist group. The probability that the organization will err approaches 100 percent as the size of the group expands. Condorcet explicitly signaled this possibility and its source: “In effect, when the probability of the truth of a voter’s opinion falls below 1/2, there must be a reason why he decides less well than one would at random. The reason can only be found in the prejudices to which this voter is subject.”40 The errors might stem not only from “prejudices” but also from confusion and incompetence. On tricky math problems, there is no reason to think that the average answer of a large group will be right. So, too, on complex issues involving politics. And even if people are competent, they might well be led astray, especially if they are subject to “prejudices” or if they are dealing with highly technical questions.

Now let us return to the topic of deliberation. Even if the average view of a large group is likely to be right, a process of deliberation might skew people’s judgments, leading to major errors. As people influence one another, a bias might creep into the group’s thinking, leading it astray. We have encountered many examples. In some circumstances, deliberation actually produces a serious distortion. Groups often do pretty well if they take the average answer, without deliberation; under bad conditions, the average answer will get worse after they have spoken to one another.

But I do not mean to challenge deliberative conceptions of democracy here or to suggest that businesses, families, religious organizations, and unions would do better simply to take polls. An understanding of group polarization does not justify that radical conclusion. It suggests only that we need to specify the idea of deliberation, rather than to celebrate it as such. A system of deliberation is likely to work well if it includes diverse people that is, if it has a degree of diversity in terms of approaches, information, and positions.41 Cognitive diversity is crucial to the success of deliberative democracy and its analogues in the private sector.

A NOTE ON GROUP POLARIZATION AND WAR

Luther Gulick was a high-level official in the Roosevelt administration during World War II. In 1948, a few years after the Allied victory, Gulick delivered a series of lectures, unimaginatively titled Administrative Reflections from World War II, which offered, in some (tedious) detail, a set of observations about bureaucratic structure and administrative reform.42 In a brief and far from tedious epilogue, Gulick set out to compare the warmaking capacities of democracies with those of their Fascist adversaries.

Gulick began by noting that the initial evaluation of the United States, among the leaders of Germany and Japan, was “not flattering.”43 We were, in their view, “incapable of quick or effective national action even in our own defense because under democracy we were divided by our polyglot society and under capitalism deadlocked by our conflicting private interests.”44 Our adversaries said that we could not fight. They believed what they said. And dictatorships did seem to have real advantages. They were free of delays, inertia, and sharp internal divisions. They did not have to reckon, on a continuing basis, with the opinions of a mass of citizens, some with little education and little intelligence. Dictatorships could also rely on a single leader and an integrated hierarchy, making it easier to develop national unity and enthusiasm, to overcome surprise, and to act vigorously and with dispatch. But these claims about the advantages of totalitarian regimes turned out to be bogus.

The United States and its allies performed far better than Germany, Italy, and Japan. Gulick linked their superiority directly to democracy itself. In particular, he emphasized  the kind of review and criticism which democracy alone affords.”45 With a totalitarian regime, plans “are hatched in secret by a small group of partially informed men and then enforced through dictatorial authority.”46 Such plans are likely to contain fatal weaknesses. By contrast, a democracy allows wide criticism and debate, thus avoiding “many a disaster.” In a totalitarian system, criticisms and suggestions are neither wanted nor heeded. “Even the leaders tend to believe their own propaganda. All of the stream of authority and information is from the top down,” so that when change is needed, the high command never learns of that need. But in a democracy, “the public and the press have no hesitation in observing and criticizing the first evidence of failure once a program has been put into operation.” 47 In a democracy, information flows within the government between the lowest and highest ranks and via public opinion.

With a combination of melancholy and surprise, Gulick noted that the United States and its allies did not show more unity than Germany, Japan, and Italy. “The gregarious social impulses of men around the world are apparently much the same, giving rise to the same reactions of group loyalty when men are subjected to the same true or imagined group threats.”48 Top-down management of mass morale, by the German and Japanese leaders, actually worked. Dictatorships are not less successful in war because of less loyalty or more distrust from the public. They are less successful because their leaders do not receive the checks and corrections that come from democratic processes.

Gulick is offering a claim here about how institutions perform better when challenges are frequent, when people do not stifle themselves, and when information flows freely. Of course, Gulick is providing his personal account of a particular set of events, and he does not really demonstrate that success in war is a product of democratic institutions. The Soviet Union, for example, fought valiantly and well, even under the tyranny of Stalin. But Gulick’s general theme contains a great deal of truth. Careful studies show that democracies do especially well in fighting wars.49 One reason is that democracies do not start wars if they are not likely to win. A more interesting reason, fitting with Gulick’s theme, is that because of individual liberty, democracies will have better access to accurate information, so that they are able to correct mistaken courses of action.

There is a more general point here. Institutions are far more likely to succeed if they subject leaders to critical scrutiny and if they ensure that courses of action will face continuing monitoring and review from outsiders if, in short, they use diversity and dissent to reduce the risks of error that come from group polarization.

DIVERSITY AND BALANCE

The idea of checks and balances is an old one, but it can be used innovatively. In the United States, a great deal of national policy is established by the so-called independent regulatory commissions, such as the Federal Communications Commission (FCC), the Federal Trade Commission, the Securities and Exchange Commission, and the National Labor Relations Board. These agencies typically consist of five members, who are appointed by the president (with the advice and consent of the Senate), serve for specified terms (usually seven years), and make decisions by majority vote. Because of the immense importance of their decisions, any Democratic president would much like to be able to ensure that the commissions consist entirely or almost entirely of Democratic appointees; Republican presidents would certainly like to shift policy in their preferred directions by ensuring domination by Republican appointees.

Under existing law, however, presidential control of the commissions is sharply constrained, for no more than a bare majority can be from a single political party. Congress has explicitly so required, and indeed, this has become the standard pattern for the independent agencies. Hence, for example, the National Labor Relations Board (NLRB) and the Federal Communications Commission must consist either of three Republicans and two Democrats or of three Democrats and two Republicans. From the standpoint of the president, a particular problem arises in a time of transition from one administration to the next. A Democratic president, for example, is often disturbed to learn that agencies entrusted with implementing legislation policy will have at least two Republicans (appointed by his predecessor).

It should be clear that the requirement of bipartisan composition operates as a constraint on group polarization and extreme movements. Five Democratic appointees to the NLRB, for example, might well lead national labor law in dramatic new directions. To this extent, bipartisan membership serves to limit unwarrantedly extreme changes in regulatory policy. We can understand the requirement as an effort to ensure against group polarization as an effort to reduce the risk that when labor law is moving in significant ways, it is simply because of social interactions among likeminded people.

Do similar considerations apply to the federal judiciary? At first glance, the judiciary is quite different, because many people believe that it is not supposed to make policy at all. But the evidence reveals a more complicated picture. We have seen that extreme movements are shown by DDD and RRR panels, in the sense that judges, on such panels, are especially likely to vote in line with ideological stereotypes.

A controversial implication is that in the most difficult and ideologically charged cases, those who seek to avoid the effects of group polarization should consider diverse judicial panels, as in the context of the NLRB and the FCC. This implication is controversial because the judiciary is not understood as a policy-making institution, because such an approach might cement judicial self-identification in political terms, and because efforts to ensure ideological diversity might well be taken as inconsistent with the commitment to judicial neutrality. But the discussion here suggests that judges are policy makers of a distinctive kind and that in principle, the argument for diversity, as a means of counteracting group polarization, is not significantly different from the argument in the context of the independent regulatory commissions. Recall that while the NLRB must be DDDRR or RRRDD, reviewing courts are not similarly constrained, and that the ultimate fate of NLRB decisions and hence national labor law, even in the most important domains, will often be radically different if the reviewing court is RRR or DDD. By contrast, appellate panels are far more moderate if they are RRD or RDD.

This example has general implications. In many domains, private and public institutions consist of like-minded people; they are akin to our experimental groups in Boulder, to RRR panels, to the White House under George W. Bush, even to social networks consisting entirely of angry Muslims. Well-functioning groups attempt to ensure a diversity of views, if only to protect themselves against blunders and confusion. If teams of doctors want to make accurate diagnoses, they will promote a norm of skepticism, even among younger and less experienced members.50 If corporations want to avoid disaster, they do best to create diverse boards that do not defer to the CEO. A close observer of corporate failures concludes that one remedy lies in “strong, high-functioning work groups whose members . . . challenge one another and engage directly with senior managers on critical issues facing corporations.”51 The problem is that when corporate directors, even intelligent and powerful ones, are placed “into a group that discourages dissent, they nearly always start to conform.” And “CEOs who don’t welcome dissent try to pack the court.” This is a serious problem for shareholders because the evidence suggests that “the highest performing  companies have extremely contentious boards that regard dissent as an obligation and that treat no subject as undiscussable.” Thus well-functioning boards contain “clashing viewpoints and challenging questions.”

These points are a tribute to the power of checks and balances to the value of creating Teams of Rivals, even in domains in which leaders usually seek team players, that is, those who go along with prevailing wisdom. If unjustified extremism is a problem, the old idea of checks and balances is likely to have a number of fresh uses. We have only started to realize its promise.