The Tyranny of the Majority: Fundamental Fairness in Representative Democracy
By Lani Guinier
(The Free Press, 324 pp., $24.95)
The distribution of political power is one of the most important issues facing any political system. And yet in our political system it is not at all easy to have a public discussion of voting rights, at least in the context of race. Sometimes it seems as if judgments about race are analogous to theological convictions. They are not movable. Many propositions in the debate are not really arguments or invitations to further discussion, but statements of identity: I am a principled person, the kind of person who thinks this; aren't you? The whole area is pervaded by accusations, mischaracterizations and strange dichotomies.
Consider, in this context, the idea that Lani Guinier is "the Quota Queen," a label from The Wall Street Journal that stuck. A recent computer search uncovered, astonishingly, no fewer than 327 separate articles repeating this phrase in discussions of Guinier, yet I have found no evidence that Guinier favors quotas. Or consider the influential editorial in this magazine last June opposing Guinier's nomination to become assistant attorney general for civil rights, because of her "published support for equal racial outcomes" and her "firm belief in the racial analysis of an irreducible, racial `us' and `them' in American society." Consider also Abigail Thernstrom's criticism of Guinier, in the same issue of this magazine, for having overlooked "an obvious solution to the set of problems racial gerrymandering creates: call the federal dogs off." Instead, according to Thernstrom, Guinier believes that "whites all think alike," which is frightening in a leader of the civil rights division, who will have power to attack "disciplinary measures in schools, ... classes for the academically gifted, `Eurocentric' curricular materials and competency tests" and will also have the authority "to insist on employment and promotion quotas where too few blacks--by the standard of proportionality--have been hired." Thernstrom continues: "She will be free to initiate suits to modify sentencing practices--labeled as `discriminatory'--that send a disproportionately high number of black men to jail."
It is conceivable, I suppose, that Guinier would have attacked "Eurocentric" curricular materials, insisted on racial quotas in employment and tried to attack the criminal sentences imposed on black men. But there seems to be no evidence in Guinier's writings to support these worries. (Guinier has written almost nothing about curricular issues, quotas in employment or sentencing.) We might well conclude that the debate over Guinier proves that Robert Bork's supporters were right. The attack on Bork, which was often distorted and unfair, appears to have grossly affected the confirmation process, helping to produce a system of "trial by sound bite," in which reasoned discussion is replaced by advertisements that treat nominees as bizarre cartoon characters with one of three traits: an "agenda," a weird personal life or "something to hide." Or, in other words, as not "in the mainstream."
What does Guinier believe? This book is a collection of her law review writings, along with a twenty-page introduction and a three-page epilogue. Much of the book is organized by a central narrative, which goes like this. The first generation of voting rights activism, represented by the Voting Rights Act of 1965, outlawed literacy tests, prohibited certain unfair registration procedures and ensured safe access to the polls; its goal was to protect the right to vote as a formal, technical matter. The second generation involved the effort to counteract sometimes subtle efforts to diminish black voting power--including "vote dilution" in at-large electoral districts, through which black voters would be submerged in a large white majority and the victory of white candidates would almost be ensured. And the third generation--well, Guinier thinks that the third generation is the contemporary one, which must now deal with a third generation of problems as well as those of the first two. Today, she writes, "it is sometimes not enough to ensure that minorities have a fair opportunity to elect someone to a legislative body. Under some unusual circumstances, it may be necessary to police the legislative voting rules whereby a majority consistently rigs the process to exclude a minority." For the third generation, the task is to make sure that minority votes really count.
It is fairly easy to describe what Guinier is against, and it is not so hard to describe what she is for. She is critical, for a start, of the performance of the Department of Justice under Ronald Reagan. Her first and least temperate essay declares (much too strongly, I think) that the Reagan administration "attempted to drive civil rights laws out of the marketplace for being more costly than they were worth." She is quite unhappy with its litigating positions, pointing out that it often argued for limited readings of civil rights laws, while at the same time siding with white and male plaintiffs in affirmative action cases. She is also unhappy with the post-Reagan federal judiciary, which she says is composed of people who meet ideological litmus tests, and which is, in her view, disproportionately white, male and conservative.
Readers will not be stunned to find that Guinier is critical of Reaganism. But they may be surprised to find that she is skeptical also about a prominent element of the movement for voting rights, which she calls "the theory of black electoral success." This theory holds that the goal of voting rights reforms is to elect "authentic" black representatives. Far from embracing this theory, Guinier thinks that it is inadequate. Her interest is, rather, in what the voters want, and she believes that black voters will have very little influence on policy if they elect black representatives to an office or two, especially if the representatives are consistently outvoted or marginalized on a multimember body. Guinier points to a number of reasons why the election of black representatives may have minimal effect on policy outcomes. Indeed, it is not at all clear whether it is better for a minority to elect one marginalized representative or to help elect several people who owe at least something to that minority.
It is by way of exploring the theory of black electoral success that Guinier discusses the "authenticity" issue that was so often raised during the debate about her confirmation. In fact, Guinier rejects the idea that only black candidates can represent black voters. She insists that "white candidates can legitimately represent nonwhite voters if those voters elected them." Throughout her book Guinier is interested in what voters want, not principally in what candidates want, and she indulges the "authenticity" assumption only to the extent that it focuses on what voters do and desire. It follows that a minority representative may be "inauthentic," but only if he or she does not have minority support. (Guinier does say that black representatives can sometimes add something by virtue of their history and experience, but this is not a major part of her argument, and in any case it does not seem like an especially alarming belief.)
Though doubting the theory of black electoral success, Guinier rejects the idea that voting is a purely individual right, and she is not opposed to race-conscious districting in principle. She says that groups, not individuals, elect candidates, and that candidates represent groups. She claims that ordinary territorial districting is actually a system for the representation of groups rather than individuals. For her, the problem with territorial districting is that the groups who are represented--a collection of diverse people who are geographically concentrated--may include systematic losers: that is, there may be groups within the geographical unit who cannot elect anyone at all. Since she thinks that candidates represent groups, the real question, for her, is which groups they represent. The problem with territorial districting is that it is not tied to any unifying characteristic other than geography. It therefore creates a large number of wasted votes, and Guinier is troubled by that.
So much for what Guinier is against. What does she favor? First, and most generally, she is for "taking turns" in politics, rather than ensuring that one group is permanently in power and another is permanently without power. And so she supports a set of policy reforms that are designed to increase the representation of minorities, usually in a race-neutral way. Guinier's proposals draw on all three generations of voting rights reform. Like the earliest reformers, she wishes to make voter registration easier, preferring automatic procedures to mechanisms that make it hard for poor people to register if, say, they do not own a car.
Guinier also argues in favor of cumulative voting, principally in state and local elections, as a remedy for proven discrimination, and within the legislature as well as the electorate. A system of cumulative voting involves the grant of multiple votes to each voter. Each voter can then allocate votes among a range of candidates as he sees fit. Thus, each of us might be given five votes for a certain seat, and we can choose what to do with those votes. We may give all five votes to one candidate, or distribute the five votes among several candidates. A system of cumulative voting exists in some localities and in many corporate elections. Though the point is often misunderstood, it is well-established that the system does not violate the one person, one vote rule because each person is given the same number of votes.
Guinier favors cumulative voting because it registers the intensity of preferences and thus makes it easier for minority groups to elect people whom they strongly favor. (Another advantage is that it is race-neutral.) One person may choose to allocate five votes among five candidates; another may give all of her votes to a single candidate. In this way a minority with intense convictions might be able to elect someone who could not be elected under a winner-take-all system. Cumulative voting thus addresses the problem of wasted votes and improves the prospects of those who are otherwise permanent losers.
Guinier writes more tentatively in favor of supermajority requirements, which would require more than 50 percent support within a governing body before action may be taken. In her most controversial claim, Guinier says that an "alternative remedy would be to require a supermajority vote on issues of importance to the majority or its equivalent, a minority veto on critical minority issues." The proposal is not attractive, as I will suggest; but in fairness it is worth observing that supermajority requirements are not unheard of in democratic processes. There are supermajority requirements for amending the Constitution and for overcoming a presidential veto; and the Reagan administration, as Guinier notes, approved a supermajority requirement for Mobile, Alabama, as a response to proven abuses. And such requirements can also be race-neutral.
Guinier's support for cumulative voting and supermajority requirements derives from her enthusiasm for "interest representation" in politics, a notion that seems to me intriguing but a bit obscure. Her goal is to ensure that a range of "interests" receives attention in the political process, so as to counteract what she sees as the serious problem of wasted votes, a problem that, in her view, arises whenever a group's preferred candidates are always bound to lose. Guinier prefers a system of "voluntary interest constituencies" to a system of territorial constituencies. But Guinier is not entirely clear about which "interests" deserve special consideration. She speaks in favor of race-neutral solutions and sometimes suggests that she is trying to protect all "minorities"--all groups, regardless of race, that are permanent losers. But she also suggests that her specific concern is racial and language minorities, since they have made "a strong, historically supported and congressionally mandated case for their claims that a homogenous, permanent majority has exercised disproportionate power consistently to degrade their influence on the political process as a whole."
There are many complications here. Consider Guinier's controlling assumption that "groups" are the subjects of voting rights. As she knows, there are hundreds or even thousands of "groups" in society, and most of them cannot elect representatives "of their own." We do not worry very much about ensuring representation for the innumerable social groups that, at least in some places, may elect disproportionately few people from their own ranks: people with greenish-gray eyes, German shepherd owners, smokers, optometrists, agnostics, Volvo drivers, Republicans in Chicago. In most cases we say that members of such minority groups do well enough if candidates must compete for their votes. Indeed, the competition for minority votes may even enhance minority power, as when a candidate must court a narrow minority that is the "swing vote" in an election.
To be sure, a well-functioning system of deliberative democracy requires a large degree of heterogeneity, which can provide new information and different perspectives on social problems. Homogeneity is no friend to democracy; diverse "interests" can be a creative force in the process of government by discussion. In this way the notion of interest representation in deliberative assemblies has much promise. But which groups deserve special protection in the ordinary competition for votes, and why? Guinier has not given complete answers to these questions or fully explained why and when the usual opportunity to participate in coalitions is an insufficient democratic safeguard for the innumerable "minority groups" whose members vote in elections.
Guinier sometimes seems to equate minority groups and "permanent losers," a term she does not clearly define, but that appears to refer to groups who cannot elect candidates at all, or who elect candidates who are themselves persistently marginalized or outvoted. But it is not enough to point to permanent losers, so defined. There are numerous permanent losers. And in a democracy many apparent losers, even permanent ones, are often winners, in the sense that they are able to obtain concessions from others, or even to persuade representatives of "the majority." This is so especially because the majority is itself not single-minded or unitary. The majority, too, is composed of an immense variety of minorities with conflicting interests. Representatives may rarely be pharmacists, boxers, post office employees or singers; but it is hoped that pharmacists, boxers, post office employees and singers all have enough political influence to participate in coalitions with others. Isn't the same often true of racial and language minorities?
We should recall here that every man and woman in society, no matter his or her race or ethnicity, is a member of many groups. One might be, at the same time, a Catholic, a man, a stockholder, a heterosexual, a science fiction fan, a Democrat, a computer programmer, a cat owner, a Caucasian, a baseball enthusiast, a stamp collector and much more. Each African American--each Hispanic, each Italian, each Catholic, Protestant or Jew--is a member of innumerable minority groups, not just one. In a liberal society, moreover, people like to define themselves by reference to multiple roles. In such circumstances the fact that some people are "permanent losers" in their capacity as members of one group may not be troublesome at all; and if it is troublesome, this is true only for some groups under some conditions.
All this suggests that it is necessary to explore why and when the capacity to participate in coalitions, for groups who do not elect people "from their ranks," is not sufficient, or is illegitimately thwarted. Sometimes it seems as if Guinier thinks that the opportunity to participate in coalitions is never sufficient, that every minority group is unfairly treated in winner-take-all systems with majority rule. But she does not back up this very broad conclusion. And if this conclusion is not supportable, then some minority groups are special after all, and it may be necessary to limit the analysis to particular minority groups and to say, as Guinier does, that the real concern is for "racial minority groups who have been, and continue to be, victimized as a result of their racial identity."
In support of this point Guinier argues that the lines between whites and African Americans are permanent and sharp. But the very notion of "group members" can be murky, even in this context, and while her position is not entirely clear, I think that Guinier overstates the extent to which whites and blacks are homogenous groups. After all, no one disputes the fact that whites disagree with each other on many things, including race-related issues. It is no less obvious that there are important disagreements within the black community as well, and that these disagreements extend to racially charged issues like crime protection, welfare reform and programs relating to reproduction.
What, then, might be said on behalf of Guinier's conclusions? I can imagine at least three reasons why African Americans may not be adequately protected by the rights to vote and to participate in coalitions that exist today.
One reason may be that African Americans face pervasive hostility and prejudice. Many groups do not elect candidates of their own, but they can band together with other groups to elect people that they prefer. Perhaps African Americans are systematically excluded from this opportunity. In some times this has been the case, and in some places it is still the case. There is still a lot of hostility and prejudice, and the omnipresence of white representatives may contribute to feelings of exclusion and second-class citizenship. But how pervasive is the phenomenon of exclusion from coalitions and coalition-building? Certainly multiracial coalitions have become a part of modern political life; certainly such coalitions help to elect candidates, as the most recent presidential election demonstrates.
Second, Guinier's remedial measures might be justified by a direct opposition of interests between blacks and whites. Perhaps blacks have political interests that are very different from those of whites. Perhaps their interests are systematically neglected in the political arena. (Some people think that this view provides a reasonable basis for questioning simple majority rule in post-apartheid South Africa. During a meeting with Senator Alan Simpson, Guinier said that her prescriptions would apply in South Africa, too, so as to protect the white minority. It is interesting that many people who were sharply critical of Guinier's prescriptions for the United States might like what she would prescribe for South Africa.) Sometimes Guinier does write as if blacks and whites have opposing interests, and undoubtedly this is true in some areas. But surely many of the interests of white Americans and black Americans overlap. Both groups have an interest in economic growth, in diminished crime, in higher wages and employment, in good public education, in a peaceful world. It is not even clear that the interests of blacks and whites diverge sharply on such race-related issues as welfare reform and anti-crime legislation.
Third, there is the narrower phenomenon of racial-bloc voting. If whites vote only for whites, and if African Americans vote only for African Americans, there may be a serious problem, certainly if the political interests of whites are rigidly opposed to the political interests of African Americans, and perhaps even if the interests are overlapping. Guinier effectively shows that racial-bloc voting persists in many places, with many whites refusing to vote for black candidates, and with many blacks strongly inclined to vote for a black candidate when this is an option. Racial-bloc voting would preclude biracial coalitions if white candidates could not garner black votes and if black candidates could not garner white votes. But sometimes whites do try hard to get black votes, even when a black candidate is in the race. And even where the African American vote is staunchly Democratic, white Democrats compete to attract black voters. Moreover, black candidates often try hard to get white votes, and sometimes they succeed (though racial prejudice can make this difficult).
There seem to be important holes in Guinier's theory. Her concerns, I think, might make sense under certain circumstances, but she has failed to specify what those circumstances are. In other words, she has failed to identify exactly why and when a "permanent loser" has a legitimate claim that the majority is a tyranny. But let us turn away from her theory and toward her practical reforms. Guinier offers a number of interesting suggestions; but again I do not believe that she has adequately justified them.
Take cumulative voting. It is a promising suggestion, and it has its virtues. It is race-neutral, an important benefit; and it might spur interracial coalitions. But would it be very much different in result from what Guinier criticizes, that is, the election of one minority representative in an assembly that is dominated by majorities? In a system of cumulative voting intense minorities will be able to get a representative where otherwise they could not, but that representative may well be consistently outvoted. Does such a system really represent much in the way of progress?
Guinier is also interested in procedural reforms within the legislative assembly itself, including cumulative voting. Perhaps this step would be helpful as a response to the marginalization of minority representatives. But cumulative voting by representatives would be quite cumbersome, and it might well increase legislative paralysis. In any case, it could give undue power to small but intensely interested groups. (Consider how cumulative voting might increase the power of tobacco companies or the gun lobby or the agricultural industry.)
Guiner does not spend much time on her most extreme suggestions, a supermajority requirement on issues of concern to a majority and a minority veto on issues of special concern to a minority. I doubt that these procedures make much sense, even as remedies. For one thing, it is very hard to know which issues count as those of special concern to a minority. Changes in welfare programs? In employment programs? In criminal penalties? It is even unclear what the minority's "interest" is with respect to such issues. Many people think that more stringent welfare programs would help African Americans, and many people think exactly the opposite. And a minority veto raises real concerns from the standpoint of democratic theory. To be sure, our constitutional system is far from majoritarian, and pure majority rule, as Guinier shows, need not be identified with democracy. Still, it would be unusual to give members of one racial or ethnic group a power to veto measures that concern them.
All this leaves much room for uncertainty. But there are three directions, I suggest, in which it would be wise to take our legal and political culture. (Two of them are suggested by Guinier herself.) The first is to seek race-neutral and race-blind voting reforms. Even if the underlying problem does have racial dimensions, including the disproportionately low social welfare of African Americans and the distinctive feelings of exclusion that stem from a history of barriers to electoral offices, race-neutral solutions to racial problems are still the best route to successful reform.
I do not believe that race-conscious programs should be taken to offend the Constitution, at least as a general rule, and I do not contend that the Supreme Court should invalidate all, most or even many of the numerous race-conscious policies adopted during the last decades. The question of affirmative action should be handled democratically, not judicially. But it also seems clear that race-conscious programs--especially in the area of voting rights, where the perception of impartiality, as well as the reality of it, is crucial--can stimulate racial antagonism, accomplish at best only some of their own goals and compromise some of the basic commitments that underlie the civil rights movement. No group should be, or be perceived to be, a special beneficiary of voting rights reform.
Guinier is also right to recommend a shift away from adjudication and toward politics as the principal place for reform, in the area of voting rights as elsewhere. Litigation is slow. It may well be unsuccessful. It tends to produce simple winners and losers, without allowing for principled compromises. We should remember that the voting rights acts were enacted politically, not judicially, and that most of the important work toward voting rights has occurred outside of the courtroom. A principal task for "the third generation" is to de-emphasize litigation and to place more faith in democratic arenas.
There is also another, more ambitious way to make progress on voting rights, though Guinier, focused on questions of procedure, does not discuss it. This would be to break the link between race and issues of wealth and poverty. Most whites are far from rich, and most poor people are white and most blacks live above the poverty line. Still, there is an undeniable association between skin color and social well-being, and it is hard to imagine that the issue of voting rights would seem quite so urgent if this association were diminished. Consider the fact that the United States ranks sixth among all nations in the United Nations Development Program's Human Development Report; that whites, considered alone, would rank first; and that blacks, considered alone, would rank thirty-first (next to Estonia). Consider, too, the fact that nearly one-third of black Americans live below the poverty level, as compared with about one-tenth of white Americans. The life expectancy of a white American is six years greater than the life expectancy of an African American. Worst of all, perhaps, 45 percent of African American children live below the poverty line, compared with 16 percent of white children.
If this divergence in the quality of life were not so great, might it not suffice to give everyone the right to vote, make registration easy and insist on a system of one person, one vote? If these numbers were not so alarming, might it not become even harder to know whether people's interests diverged systematically along racial lines, and easier to build interracial coalitions? These questions do not have clear answers; but I think that the questions themselves are enough to show that the problem of voting rights is enmeshed with the problem of racial inequality outside the explicitly political sphere. The problem of voting rights takes its present shape only because of the problem of huge racial disparities in general well-being.
It might be thought--and Guinier might well think--that such disparities justify the creation of special voting reforms, since without such reforms the disparities will not be redressed in ordinary politics. And it is true that the political process has often failed to respond to the second-class economic and social status of African Americans. But if our real concern is that second-class status, we should recall that reforms in the political process are often a crude and indirect way to respond to inequalities outside of politics. As long as the one person, one vote rule is respected, and as long as every member of every group is allowed to vote, it is hard to establish a satisfactory link between further voting reforms and specific measures to help people who are disadvantaged.
Sometimes the debate over voting rights for minority group members is mixed with the debate over inequality in general. And this clouds the discussion. Indeed, efforts to counteract existing social disparities need not be racially explicit. Like cumulative voting--and like child immunization programs, neonatal care, literacy programs, anti-crime policies, job training programs and educational reform--such efforts could well be racially neutral. The next stage of civil rights policy, in voting and elsewhere, should be directed self-consciously against concrete racial inequalities at the same time that it is self-consciously designed to avoid specificity about race. Until the social and economic disparities between black and white diminish, many of our disagreements about race-related questions will be mere skirmishes over a problem whose deepest sources lie elsewhere.
By Cass R. Sunstein