Democratic Backsliding and Multiracial Democracy: A Response to the 2021 Jorde Symposium Lecture by Steven Levitsky

This nation is not now, never has been and now never will be a white country.

– James Baldwin[2]

Introduction

We live in an anxious era, particularly about the possibility of multiethnic democracy. The polarization of American democracy in general, accelerated by Trumpism in particular, has challenged narratives of race as gradually declining in significance. Instead, conventional wisdom suggests that Trumpism results directly from rising racial resentment of a White population that fears losing its relative power.[3] “Dog Whistle Politics” have been discarded in favor of openly nativist appeals, including by media figures such as Tucker Carlson.[4]

We are not alone. In France, the theory of the Grand Remplacement (Great Replacement) has spread from the fringes to the open world of presidential politics through candidate Eric Zeymour who placed fourth in first-round voting in 2022.[5] And other European countries have seen an uptick in racial anxiety. All this lends a sense of urgency to the project of establishing a multiracial coalition committed to democracy as an ongoing concern. Otherwise, we risk succumbing to the “authoritarian backlash” identified in Professor Levitsky’s Jorde Symposium Lecture. Racialized politics tend toward the winner-take-all variety, and they do not allow for the fluid coalition shifting that has been celebrated by analysts from James Madison to Robert Dahl.[6]

In this Response, I make three points, drawing on the scholarship about constitutional mechanisms for mitigating ethnic conflict. First, multiethnic democracy is not as rare as has been suggested. Indeed, it is increasingly the normal state of affairs in rich countries, fueled by patterns of global migration.

But I will argue there are some multiethnic countries whose particular histories lend themselves to authoritarian backlash and democratic backsliding. These are countries with a history of mass slavery, of which the United States and Brazil are the most prominent examples. Slave-legacy societies are especially prone to existential racial appeals as an instrument of political mobilization, and the perception of high stakes can lead to attacks on institutions.

Second, the traditional mechanisms of “managing” ethnic conflict are in some flux and do not suggest an easy way out of the current crisis, even if we could redesign constitutional arrangements from scratch. The large literature coalesces around two main constitutional approaches: consociationalism and centripetalism.[7] Consociationalism involves peak-level bargains among leaders who represent relatively homogenous ethnic groups; centripetalism emphasizes institutions that encourage cross-ethnic cooperation. Whatever the merits of these approaches in theory, the debate has not operated at a fine enough level of detail to suggest many plausible solutions for the United States. Centripetalism would operate here mainly through the design of multiracial electoral districts and perhaps the adoption of ranked-choice voting, which is emerging in mainstream discourse as a lynchpin institutional reform. But because elections management is constitutionally decentralized in the United States, this is unlikely to be a magic bullet for the country as a whole. I predict increasing divergence in democratic practices, as well as the resulting policies, across the United States.

Third, society—rather than law or politics—may provide the best pathway to address the challenges of multiracial democracy in the United States. We are feeling our way to a multiracial culture, which I designate using the term mestizaje, and in which fluidity and mixing also extend to demographic understandings. But the success of that project will hinge on whether our institutions can withstand the assault on democracy President Trump and his allies have unleashed. The future of our multiracial democracy hinges on whether the Republican Party returns to democratic competition, and there are signs this could occur. Nothing, of course, is guaranteed. Subnational autocracy remains a very real option for parts of the United States, even as others end up being bellwethers of multiracial democracy.

Each of these points intersects with Professor Levitsky’s lecture. The first point, that multiracial democracy is rather common, undercuts the claim that this would be an “unparalleled achievement.”[8] While this is true for the United States, it is not unknown elsewhere, and so directs us to ask why the United States is somewhat exceptional in its democratic trajectory. The second point, that conventional approaches to constitutional design do not provide off-the-shelf solutions, is consistent with Levitsky’s argument that we may be in real trouble. The third point about the growing complexity of racial categories suggests that more optimism may be warranted, at least for the middle term and beyond. There are surely signs of difficulties for the next election cycles, but the future of the United States is likely to be both multiracial and democratic.

Table of Contents Show

I. Is Multiethnic Democracy Possible?

In his 2020 Foreword for the Harvard Law Review, Professor Michael Klarman stated that multiracial democracy is exceedingly rare.[9] The Foreword is an extremely forceful brief describing the Trumpist attack on American democracy, published before the treasonous attempt to overturn the election on January 6. Klarman argued that racial resentment, particularly by White people fearful of losing their status as a demographic majority, is the biggest factor explaining the recent degradation of American democracy.[10] In this, he follows what we might call conventional New York Times wisdom that Trumpism is a White nationalist movement, connecting the attack on democracy as one further step in our long history of White supremacy.

I suspect the reality is a bit more complicated, as Klarman’s account ignores the substantial shares of Hispanic-Americans and Asian-Americans who voted for Trump. Presidential coalitions are always complex and varied. But it is true that disagreements over how to grapple with race are among the most polarizing issues in the United States today, so the possibility of multiethnic democracy is of great concern.

In fact, multiethnic democracy is possible, and indeed it is becoming a general condition of industrial democracies in an age of mass immigration. Australia and Canada have welcomed substantial numbers of immigrants in recent decades, with the foreign-born share of their populations rising to 25 and 20 percent, respectively. These are liberal societies like the United States, in which voluntary membership rather than ethnic ascription is the nominal basis of the country. Prior generations of immigrants in each of these places have overcome racialized exclusionary movements to enrich their culture and community, even while native populations in all three countries were displaced and marginalized. Among the fastest growing groups in each society are those of Asian origin. The share of the overall populations of Chinese or Indian descent is roughly 8 percent in each country.

In the United States, the process of ethnic integration has always been multiracial, in the sense that newcomers were racialized by the dominant White majority.[11] Contra Whoopi Goldberg, groups such as the Irish, Germans, Italians, and European Jews were all characterized as racial at various points. From our contemporary, socially contingent view of race, racial diversity expanded radically with the Immigration and Nationality Act of 1965, which replaced the prior system of seeking to preserve demographic homogeneity with a system of national quotas. This facilitated a wave of immigration from Asia and Africa that has, two generations later, significantly diversified the United States. In 1965, non-Hispanic White people composed over 85 percent of the American population, but that has dropped to below 60 percent today.[12] As a collateral consequence, immigration has diluted the electoral power of African-American descendants of enslaved persons.

Beyond settler societies that were former British colonies, other rich democracies are diversifying. Germany is 12 percent foreign born and more than 10 percent of the population has African or Asian ethnic origin.[13] France keeps no statistics by race, but it is estimated that roughly 10 percent of the population is of North African descent and another 3.5 percent is from Sub-Saharan Africa.[14] Only two-thirds of Swedes are born of two Swedish parents, and the country includes substantial non-White populations.[15] None of these countries has seen an attack on its democratic institutions, even if populist and anti-system parties have grown.[16] While Levitsky emphasizes the prospect of minority rule in the United States, these other political systems are largely majoritarian. Anti-immigrant parties seek to influence the majority, rather than capture the government and eliminate electoral competition.

Other diverse societies have also retained democratic institutions. Setting aside its undemocratic rule over millions of Palestinians in the occupied West Bank and Gaza, Israel proper is 21 percent Palestinian Arab. The remainder of the population, though primarily Jewish, includes phenotypically diverse groups of Sephardic Arabs and Ethiopians that combine to form 25 percent of the total population, as well as a group of ethnic Russians that are 14 percent of the population. This internal diversity creates a good deal of tension, but democratic institutions have held despite recent risks.[17] India is an utterly diverse country, with 121 languages spoken. It has remained a democracy until very recently, when Narendra Modi’s Hindu nationalist project led several analysts to reclassify the country as an electoral authoritarian system. And many developing country democracies are quite diverse, as they were formed through colonial exercises in line drawing that paid little heed to existing boundaries of ethnic groups, or else involved large transfers of population from other areas to suit colonial labor needs.

In short, we observe several multiethnic countries that might be considered to be fragile or at risk of democratic backsliding. But few of these have succumbed to pressure on democracy: only India really qualifies as a high-risk case at this writing.[18] It is true that there have been political forces, like the Alternative for Germany (AfD) or the National Front in France, that have protested racial diversification. But none of these countries has seen attacks on its very democratic institutions. In contrast, the countries that have suffered democratic backsliding have frequently been relatively homogenous.[19] Indeed, among four of the more prominent cases of democratic backsliding in recent years—Venezuela, Poland, Hungary, and Turkey—only the latter can be said to have a major racial cleavage, with 19 percent of its population being Kurdish.[20] Venezuela is almost entirely White or mestizo, and Hungary’s population is officially 94 percent Hungarian.[21]

We thus have examples of homogenous countries that have suffered democratic backsliding and rich, heterogenous countries that have not. Only the United States seems to be a rich, democratic, heterogenous country that is at risk of backsliding. Why?

The answer may lie in the peculiar racial and demographic history in the United States as compared with these other places. Postwar Germany and Sweden were democratic before they became multiracial. This means that immigrant populations were being absorbed into a set of relatively stable institutions. This project is not without challenge, but there is some literature on which approaches are successful and which are not.[22] Australia has a similar history, with the complication of a displaced indigenous population. Israel and India were born as pluralistic democracies, and they have grappled with difference since their earliest days.

Unlike these countries, the United States was born multi-racial and undemocratic. The presence of mass slavery shaped the negotiation of the constitution, and, of course, the country’s politics in subsequent decades. Regression was as common as progress. The birth of the world’s first mass democracy in the Jacksonian era corresponded with new restrictions on the franchise of women and free Black men. It also gave rise to the Indian Removal Act of 1830, which led to the forced displacement of tens of thousands of Native Americans.

Despite attempts after the Civil War, the country has never been completely re-founded in the way that Germany was after World War II. The franchise has continually expanded even as the country has absorbed waves of immigrants, but because of the Jim Crow regime, subnational authoritarianism obtained in much of the country for most of the twentieth century.[23] The United States was arguably not a national democracy until the Voting Rights Act of 1964, and it retains a legacy of social exclusion that provides an easy basis for race-based mobilization.

In this regard, the country most similar to the United States is not one of the diversifying industrialized Western European ones, but rather Brazil: another large, federal country with a history of slavery. Brazil was the world’s largest importer of enslaved Africans, and the country did not abolish the practice legally until 1888. As a result, the country is majority “Black,” with other groups of European immigrants, native populations, and large numbers of what we call mixed-race people. In the early twentieth century, authorities adopted a policy of encouraging mixing, or “whitening,” as a way of resolving the perceived race problem after the demise of slavery.[24] In 1933, a young Brazilian academic named Gilberto Freyre argued that miscegenation, not separation, was a key to Brazilian identity.[25]

The country has been a democracy since 1985, but after the United States, it is the most important country to experience democracy-threatening backsliding. President Jair Bolsonaro, a populist firebrand who openly expressed authoritarian musings, led attacks on the country’s institutions and sought to degrade them.[26] Although he lost a close election to former President Ignacio Lula de Silva in late 2022, at this writing he has still not conceded defeat.[27]

While the factors providing an opening for Bolsonaro are complex and myriad, one issue is the expansion of affirmative action provisions to address significant under-representation of the country’s Black majority. The left-wing governments of the early 2000s introduced quotas for universities and state employment. Backlash against these fledgling efforts has been one of the factors that led to Bolsonaro’s rise. On the campaign trail, he regularly made openly racist statements.[28] Since in power he has slashed budget support for Black and indigenous communities, including for COVID-19 relief.[29]

To be sure, race in Brazil is conceived of quite differently from race in the United States. Census categories in Brazil include White, Black, Asian, Indigenous and Mixed. But there are at least three different ways of categorizing the population on the Black–White continuum.[30] A folk method in popular use includes seven major and many minor categories.[31] According to the census, Black people are a minority of around 10 percent and the mixed category comprises 43 percent of the population.[32] But beginning in the early twentieth century, activists rejected the census categories and argued for a combined “negro” category of mixed-race and Black Brazilians.[33] Today the term afrodescendente sums up both, and this is the basis for estimates claiming that 56 percent of the population identifies as Black.[34]

Whether defined narrowly as in the census or more broadly, this group is subject to severe under-representation in the political and economic spheres, as well as greater vulnerability to police violence.[35] Rising racial consciousness has exposed many of the myths of an earlier generation. Yet Bolsonaro’s Vice-President Hamilton Mourao was able to claim with a straight face that the country was free of racism.[36]

To summarize, many of the world’s leading democracies are becoming increasingly plural in racial makeup. This means that multiracial democracy is not the non-sequitur that some would suggest. But diversity also introduces new tensions, as well as tempting levers for mobilization by political entrepreneurs. The two multiracial societies most affected by democratic backsliding, I would submit, are Brazil and the United States, both of which have legacies of mass enslavement that make them particularly susceptible.

The precise mechanisms of authoritarian attraction here are myriad, but the deeply seeded cultural fear of slave rebellion must rank high among them. Early America featured numerous slave revolts, and a strong culture of repression against the threat led to a rise in conspiracy theories.[37] The Haitian Revolution in particular raised the prospect of the overthrow of the Southern system and led to a heightening of repression in the United States.[38] Brazil too experienced violent slave revolts, which were by some accounts a major factor in the ultimate abolition of slavery in 1888.[39] While Brazil’s response was different from that of the United States, both experienced waves of immigration after the demise of slavery, as free labor came in to substitute for the enforced labor of slavery.

The relevance of slave rebellion to today’s politics might seem distant, but in many ways, it is still present. As Kimberlé Crenshaw wrote of Black Lives Matter protests,

The reflexive treatment of erratic outbreaks of looting as a more ominous threat than the organized massing of state terror also speaks volumes about the nation’s real civic priorities . . . . The fear of being outnumbered and overrun by the indignant and unforgiving masses is that thing that goes bump in the night[.][40]

The political response to the protests over the George Floyd’s murder has polarized the United States further, with Republican-led state legislatures passing dozens of laws that inhibit the freedom of speech on matters of race, focused on the poorly specified signifier of “Critical Race Theory.”[41] A violent insurrection bent on overthrowing the constitutionally mandated process for transferring power was retroactively justified by violence that occurred during some Black Lives Matter protests. These central cleavages in American politics cannot be understood outside the race framework. And they set the United States apart from other advanced industrialized democracies of the world. In my view, they explain the peculiarly American vulnerability to democratic backsliding.

II. Can Constitutional Design Help?

This Section considers what might be done to address the problem of race-based democratic backsliding, and whether constitutional design might help. My 2018 book with Aziz Huq, How to Save a Constitutional Democracy, focused on the general issues of constitutional design and democratic backsliding.[42] But it did not address the specific context of race and its role in stimulating institutional decline. While American exceptionalism is not a particularly useful lens for most matters of democratic performance, it might be that we are exceptional—not in our resilience but in our specific vulnerabilities.[43]

A large literature considers the mechanisms by which constitutional design might ameliorate ethnic conflict and facilitate cooperation under conditions of democracy.[44] The challenge is that democracy is legitimated through majority rule, but if politics are organized around ethnicity, there is the possibility that one group will persistently win and others will be permanently excluded. This dynamic can lead to violent conflict. Levitisky’s discussion of trends within the Republican Party, in which its largely White Christian base feels threatened, suggests that the ethnic lens is a useful one.[45] This Section considers the two leading approaches, consociationalism and centripetalism. The first emphasizes ethnicity as a basis for politics; the second tries to reduce its salience.

A. Consociationalism

The leading advocate of consociationalism is Arend Lijphart.[46] In consociational constitutional systems, such as those in Bosnia, Malaysia, and Lebanon, elite bargains keep the racial peace. Individuals typically live in isolation from each other, and communities can exercise a degree of self-determination on important issues like speaking their own language. The leaders at the top of the system bargain with each other over the distribution of government resources. As Brad Roth noted, this model “expressly presupposes party leaders who are more tolerant than their mass bases, but nonetheless able to bring their mass bases along.”[47]

Institutionally, the way consociationalism works is to produce a grand governing coalition that includes all sides across the major cleavages in society, with constitutional vetoes for minorities. For example, in Bosnia-Herzegovina, the presidency is held by a set of three people, one of whom represents each of the major ethnic groups. Decisions by the collective presidency require unanimity. Another institutional feature of consociationalism is proportionality in the allocation of political offices, public employment, and spending, as well as in representation.

One of the problems with this model is that, once established, it tends to reinforce rather than reduce the relevance of ethnicity. While it may be necessary to establish peace, it does not allow evolution away from the cleavage that triggered the conflict in the first place.[48] Discussing Bosnia, Roth noted that, “[a]lthough Bosnians of all ethnicities understand the dysfunctionality of the governing structure, they cannot escape it.”[49] Furthermore, elites have an interest in remaining in the coalition, implying higher levels of public spending and corruption.

Lebanon is a classic example of a consociational system. The country was founded as a Christian enclave within a broader Muslim Middle East, but has always been religiously diverse.[50] With independence from France in 1943, a power-sharing arrangement was put in place that persists to this day. Constitutionally, the president is always to be a Maronite Christian, the Speaker of the House a Shia Muslim, and the Prime Minister a Sunni Muslim. After demographic changes put pressure on the system, the Taif Agreement in 1989 promised to adjust the scheme, but the changes were never implemented. The result is a corrupt state in which leaders of the three sects divide up spoils and distribute them through patronage. The military, historically led by Christians, makes no pretense of a monopoly on the use of the force, as the Hezbollah militia retains territorial control of Southern Lebanon. The failures of Lebanon have led to a general critique of the consociational model.[51]

Does consociationalism have any application to the United States? Roth, in his review of Lijphart’s argument, noted that Lijphart identified the post-Civil War South as a kind of consociational arrangement, an assertion that Roth noted “can only be characterized as baffling.”[52] The post-Redemption South was not considered a democracy by any stretch of the imagination, and African Americans were not a part of the governing coalition at all. The post-Redemption South should be considered an instance of racial domination, not accommodation or integration.

Perhaps the better American analogy to the consociational model is the urban machine politics of large multiethnic cities such as Chicago. Groups in such cities tended to congregate geographically, giving them a base for mobilization and a focus for patronage-based distribution.[53] Like national-level consociational systems, some large cities remain hotbeds of corruption, and urban politics sometimes depend on deals among top leaders who purport to represent ethnic constituencies. But our national-level institutions do not reflect elite racial bargains.

Even if political agreement at the top is unlikely to work in the United States, consociational analysis has some relevance. Roth noted that consociationalism can create a Hobbesian problem: even when all individuals would rather de-emphasize group identity, it becomes in their interest to instead emphasize it, since the society explicitly prioritizes identity.[54] This is, arguably, a fairly good description of the United States today.

B. Centripetalism

The chief antagonist to Lijphart has been Donald Horowitz, who put forward what he calls the centripetal theory.[55] Centripetal approaches, like consociational ones, recognize that group differences are salient social constructions and often a basis of political organization. However, centripetalists seek to design institutions that foster cross-ethnic cooperation at the electoral level rather than grand coalitions at the top. Horowitz emphasized something called the alternative transferable vote, in which election systems are designed to encourage voters to consider candidates from multiple ethnic groups. This, in his view, would reduce the incentives of politicians to “out-bid” each other in appealing only to ethnic solidarity, as opposed to other bases of political mobilization.

In the United States, the best example of such a system is ranked-choice voting, now in use in national elections in the states of Maine and Alaska, as well as many localities. Ranked-choice voting is viewed by many as a lynchpin reform that could reduce polarization and help to fix American politics.[56] In a ranked-choice voting system, individuals do not just choose one candidate, but rather they assign ranks to all of the candidates. If one candidate obtains more than 50 percent of the first-place votes, that person wins. But if not, the candidates at the bottom of the first-place choice are eliminated and their votes redistributed to those voters’ second preferences. The process is repeated until someone has a majority of votes. Under this system, the incentive to take extreme positions is reduced since one must appeal to voters who prefer other candidates. Under plausible conditions of diverse polities, candidates will need to appeal to members of other ethnic groups, and so appeals based purely on ethnicity are likely to fail.

While this is an attractive mechanism, it still means that some candidates with more extreme racial and political views could be elected, if districts are drawn in such a way as to make their supporters a majority. Thus ranked-choice voting likely would have to be paired with boundary-drawing exercises undertaken by a neutral commission in order to really pull United States politics back from the brink. Constitutionally, electoral boundaries are clearly a matter of state-level policy, and this is only becoming more true as the Supreme Court guts the 1965 Voting Rights Act.[57] But centripetal institutions do have much promise. Given the gridlock that prevails in Washington, they would likely have to be institutionalized at the state level, and states that do so may experience de-polarization and moderation of their politics. But some states will not institutionalize centripetalism, leading to further divergence across red and blue states. And this means that national-level politics could remain quite polarized, with spillovers into the politics of states with sound democratic institutions.

III. Two Social Bases for Optimism

This Section argues that, whatever the fate of American constitutional reforms at the state level, there are some developments in society that might, under certain conditions, reduce the appeal of race as a leverage point for those forces to erode democratic institutions. These are the continued unsettling of race science in light of the developments in DNA technologies and the declining salience of race in individual-level demographic choices.

A. The Unsettling of “Race Science”

Efforts to categorize people into phenotypically defined races are a relatively recent development. Beginning in the eighteenth century, “race science” led to a whole line of research that had, in many cases, deadly consequences. What is in some ways remarkable about race-talk is its persistence and durability. The misconception that the concept of race has a biological rather than social origin remains widespread, even among physicians.[58] Socially defined racial groups have far more internal genetic variation than is popularly understood.

Yet the widespread availability of DNA technology may undermine this myth more quickly than the public schools have been able to. DNA technology is able to identify people’s geographic origins with high degrees of specificity, and it reveals that many Americans have very diverse ancestries. For example, the mean level of European-origin DNA in people categorized as Black in the United States is 16 percent.[59] Many White people have discovered that they have Native American blood (and many, like Elizabeth Warren, have discovered that they do not). At this early stage of knowledge, the phenomenon is surprising to people, but as more people complete DNA tests, it may give rise to a more subtle language about race.[60]

To be sure, this is somewhat speculative. The public schools are likely to remain a major battleground over the issue of how we teach race for the next few election cycles. But it is at least possible that truth will come out, both about our troubled racial history and about the biological uselessness of race as a predictive category.

B. Growing Mixed-Race Population

Another major social change around race that could have consequences for our democracy is the rapid growth of Americans with more than one socially defined race. That is, despite a history of anti-miscegenation law and norms that persisted into my own lifetime, social barriers have been loosening.[61] Polls show that opposition to mixed-race marriages is nearly vanishing; in 1958, only 4 percent of respondents approved of Black-White marriages, while in 2021, only 6 percent dis-approved.[62]

This means that a growing percentage of people have relatives across racial lines. 10 percent of Americans currently identify as multiracial, which is up an astounding 276 percent since the last census in 2010. 13 percent of African Americans are of mixed race, in that they designate another race in the census. Two-thirds of these are with the White category, and the numbers are rising in the younger cohorts. Might the number of non-Black people with a Black relative change attitudes? Might mestizaje help us get out of our dead-end? It certainly suggests that a consociational solution, in which ethnic groups remain separated and rigidly defined, is not an option. And some sociologists speak of an emerging multiracial mainstream, in which racial isolation is declining for many, but not all, minority groups.[63]

C. The Endogeneity of Identity to Public Policy

Identity is endogenous to policy. The census has changed its categories over time, and self-reporting is the basis of categorization. This can lead to anomalies that, at one level, reinforce the idea that race and identity are social constructions, but also can be conflict-generating. To give one example, Native American identity jumped 70 percent between 2010 and 2020, a rate of increase unexplainable by demographic factors alone. No doubt, some of this is due to DNA testing that has revealed to many Americans that they have Native blood, sometimes in percentages that would allow for tribal enrollment. Perhaps others seek to obtain benefits associated from checking the box on applications for employment or education. The Navajo nation, for example, saw record enrollment during the coronavirus pandemic, as people rushed to get benefits distributed through the tribe.[64] Tensions can arise between new members and those who grew up on the reservation. These tensions reflect the difference between genetic and cultural ideas of membership.

Consider a completely different set of government policies around race, those of post-genocide Rwanda. In reaction to the long-standing distinction between Hutus and Tutsis, exacerbated by Belgian colonialists, the authoritarian government of Paul Kagame has set about trying to eliminate racial difference. The Constitution seeks to ban ethnic “divisionism.”[65] People are not identified by race or ethnicity, and parties cannot be based on these factors. Those who do seek to discuss ethnicity can be jailed. There is a good deal of evidence that the policies have had an impact and promote a sense of national identity.[66] Again, identity responds to policy.

The future direction of our multiracial democracy may depend on public policy and law, which helps to construct and incentivize identity. Indeed, as I write, the U.S. Supreme Court is considering whether affirmative action policies, in particular Harvard’s soft quota on Asian applicants but also similar mechanisms utilized at many other elite private institutions, are constitutional.[67] Most readers of this Essay will be members of university communities whose racial politics diverge from those in the broader society. Diversity policies are popular on campus but contested among members of the general public. For example, California’s Proposition 16, which would have reversed the state’s ban on affirmative action, failed by a large margin in 2020.[68] At the same time, affirmative action programs have been used with some success in other countries.[69]

How these issues should be resolved in the United States is one of the great issues of politics today. The GOP seeks to focus on schooling as an issue for the culture wars, and it has used this to win elections in places like Virginia, without attacking democratic institutions per se. But the Republican party has also taken a heretofore obscure academic approach known as Critical Race Theory (“CRT”), developed in law schools in the 1980s and 1990s, and turned it into a bogeyman for purposes of electoral mobilization.

The assault on CRT exhibits an authoritarian inclination like that of Paul Kagame. Like Rwanda’s leader, anti-CRT movements seek to suppress teaching about race and chill the freedom of research and teaching on which an open society depends. It is surely easy to caricature some CRT practitioners, but an active debate over these issues is by and large healthy and necessary. Instead, the Republicans wish to chill all discussions of race, which itself is not consistent with the views of the American mainstream.[70]

From the beginning of the Republic, American leaders have emphasized the connections between education and democracy. Thomas Jefferson famously thought that education was essential for a free society.[71] Today, we are seeing race and education as a major battleground. When combined with institutional attacks on the right to vote, it is apparent that there is a genuine threat of authoritarian backlash, and so Levitsky’s call for vigilance is most welcome.

Conclusion

One cannot save multiracial democracy without saving democracy itself. The means defending democratic institutions, including freedoms of speech and association, that are under assault. State-level voting rights, and a shift toward more centripetal institutions such as ranked-choice voting, would make a good deal of difference in the quality and responsiveness of our democracy. Having courts that took a more substantive view of democracy reinforcement would also make a good deal of difference. Regardless, our counter-majoritarian constitution will remain in place, meaning that the battles for democracy will be ongoing for some time to come.[72] We are lucky to have guides like Levitsky to help us understand the fight.

Copyright © 2022 Tom Ginsburg, Leo Spitz Professor of International Law, University of Chicago Law School. Thanks to Aziz Huq for ongoing discussions on this and related topics. DOI: https://doi.org/10.15779/Z38NK3660V.

2. James Baldwin, On Language, Race and the Black Writer, L.A. Times (1979), https://marktwainstudies.com/wp-content/uploads/2017/02/Baldwin-1979-.pdf [https://perma.cc/WN3X-XF2Q] (last accessed March 3, 2022). ↑

3. John Sides, Michael Tesler & Lynn Vavreck, The 2016 U.S. Election: How Trump Lost and Won, 28(2) J. Democ. 34 (2017). See also Jon Green & Sean McElwee, The Differential Effects of Economic Conditions and Racial Attitudes in the Election of Donald Trump, 17(2) Persp. Pol. 358 (2019) (racial attitudes a major factor predicting voting behavior of whites). ↑

4. Cf. Ian Haney Lopez, Dog Whistle Politics: How Coded Racial Appeals Have Reinvented Racism and Wrecked the Middle Class (2015). See also Tina Nguyen, “The America We Know Doesn’t Exist Anymore”: Fox’s Dog Whistle Becomes an Air Horn, VanityFair.com (Aug. 10, 2018), https://www.vanityfair.com/news/2018/08/laura-ingraham-white-nationalist-rhetoric [https://perma.cc/E45B-EEZU]. ↑

5. . 2022 French Presidential Election First-Round Results, FranceTV https://www.france24.com/en/live-2022-french-presidential-election-%E2%80%93-first-round-results [https://perma.cc/YEW3-3X8C] (last visited Nov. 28, 2022). ↑

6. See The Federalist Papers No. 10 (James Madison); Robert Dahl, Polyarchy: Participation and Opposition (1972) (pluralism means that no one rules permanently in a democracy). ↑

7. Sujit Choudhry, Constitutional Design for Divided Societies: Integration or Accommodation? (2008). ↑

8. Steven Levitsky, The Third Founding: The Rise of Multiracial Democracy and the Authoritarian Reaction Against It, 110 Calif. L. Rev. 1991, 1992 (2022). ↑

9. Michael Klarman, The Degradation of American Democracy—And the Court, 134 Harv. L. Rev. 1, 107 (2020) (“The world has almost no experience with true multiracial democracy.”). ↑

10. Id. at 107. ↑

11. Sahar Aziz, The Racial Muslim 71–73 (2021). ↑

12. . Quick Facts, U.S. Census Bureau, https://www.census.gov/quickfacts/fact/table/US/PST045221 (White, not Hispanic category is 59.3 percent) [https://perma.cc/46GV-QE78] (last visited Nov. 2, 2022). ↑

13. . Germany, World Population Rev., https://worldpopulationreview.com/countries/germany-population (12 percent foreign born) [https://perma.cc/5J8Y-4SQS] (last visited Nov. 2, 2022); Demographics of Germany, Wikipedia, https://en.wikipedia.org/wiki/Demographics_of_Germany#Ethnic_minorities_and_migrant_background_(Migrationshintergrund) [https://perma.cc/49ZU-G643] (last visited Nov. 2, 2022). ↑

14. France, World Population Rev., https://worldpopulationreview.com/countries/france-population [https://perma.cc/6XSG-F6QC] (last visited Oct. 31, 2022). ↑

15. . Demographics of Sweden, Wikipedia, https://en.wikipedia.org/wiki/Demographics_of_Sweden#Origin [https://perma.cc/RQT9-5DNP] (last visited Nov. 2, 2022). ↑

16. Fiona Hartleb, Here to Stay: Anti-Establishment Parties in Europe, 14 Eur. View 39 (2015). ↑

17. Tom Ginsburg & Aziz Z. Huq, How to Save a Constitutional Democracy 87–89 (2018) (describing the risks); Israel Celebrates its Democracy’s Center, Christian Sci. Monitor (June 13, 2022), https://www.csmonitor.com/Commentary/the-monitors-view/2022/0613/Israel-celebrates-its-democracy-s-center [https://perma.cc/X3KP-5HZM] (describing centrist coalition that emerged). ↑

18. Ginsburg & Huq, supra note 16, at 28–29. ↑

19. . See Robert Kaufman & Stephen Haggard, Democratic Decline in the United States: What Can We Learn from Middle-Income Backsliding?, 17(2) Persps. Pol. 417, 423 (2019) (noting that racial resentments are mobilized by polarizing leaders). ↑

20. . Turkey Demographics Profile, Indexmundi, https://www.indexmundi.com/turkey/demographics_profile.html [https://perma.cc/EV39-LVVL] (last visited Nov. 2, 2022). ↑

21. . Demographics of Hungary, Wikipedia, https://en.wikipedia.org/wiki/Demographics_of_Hungary#Ethnic_groups_and_language [https://perma.cc/LED6-XR6N] (last visited Nov. 2, 2022). ↑

22. . See, e.g., Andreas Wimmer, Power and Pride: National Identity and Ethnopolitical Inequality Around the World, 69 World Pol. 605 (2019). ↑

23. Edward L. Gibson, Boundary Control: Subnational Authoritarianism in Democratic Countries, 58 World Pol. 101 (2005). ↑

24. Kim D. Butler, Up from Slavery: Afro-Brazilian Activism in Sao Paulo 1888-1938, 49 Americas 179, 179–80 (2018). ↑

25. Gilberto Freyre, The Master and the Slaves: A Study in the Development of Brazilian Civilization (Samuel Putnam trans., 1946) (1933). ↑

26. Emilio Neder Meyer, Constitutional Erosion in Brazil (2021) (elaborating on Bolsonaro’s attacks on the courts, the media, and the electoral bureaucracy). ↑

27. Matt Rivers, Aicha El Hammar Castano & Rachel Alves de Lima, Brazil’s Elections are Over but the Fight is Just Beginning: Analysis, ABC News (Oct. 31, 2022), https://abcnews.go.com/International/brazils-elections-fight-beginning-analysis/story?id=92424064 [https://perma.cc/4SLF-NSSZ]. ↑

28. Ciara Nugent & Thaís Regina, How Black Brazilians Are Looking to a Slavery-Era Form of Resistance to Fight Racial Injustice Today, Time (Dec. 16, 2020) https://time.com/5915902/brazil-racism-quilombos/ [https://perma.cc/ZV7G-HWYR]. ↑

29. . Id. ↑

30. Edward E. Telles, Racial Classification, in Race in Another America: The Significance of Skin Color in Brazil (2004). ↑

31. . Race and Ethnicity in Brazil, Wikipedia, https://en.wikipedia.org/wiki/Race_and_ethnicity_in_Brazil#cite_note-Jos%C3%A9_Luiz_Petrucelli-60 [https://perma.cc/2KB2-KDN6] (last visited Nov. 28, 2022). ↑

32. Id. ↑

33. Butler, supra note 23, at 179–80; Tianna S. Paschel, Becoming Black Political Subjects: Movements and Ethno-Racial Rights in Colombia and Brazil (2019). ↑

34. . Race and Ethnicity in Brazil, supra note 30. ↑

35. Nugent & Regina, supra note 27. ↑

36. Id. ↑

37. Jason Sharples, The World that Fear Made: Slave Revolts and Conspiracy Scares in Early America (2020). ↑

38. Tim Matthewson, Jefferson and the Non-Recognition of Haiti, 140 Proc. Am. Phil. Soc’y 22, 24 (1996). ↑

39. João José Reis, Slave Rebellion in Brazil: The Muslim Uprising of 1835 in Bahia (Arthur Brakel trans., 1993). ↑

40. Kimberlé Williams Crenshaw, Fear of a Black Uprising, New Republic (Aug. 13, 2020), https://newrepublic.com/article/158725/fear-black-uprising-confronting-racist-policing [https://perma.cc/T5GR-G2ND]. ↑

41. Sarah Schwartz, Map: Where Critical Race Theory is Under Attack, Edu. Week (June 21, 2021), https://www.edweek.org/policy-politics/map-where-critical-race-theory-is-under-attack/2021/06 [https://perma.cc/77ZC-F7ET]. ↑

42. Ginsburg & Huq, supra note 16. ↑

43. . Id. at 4–5. ↑

44. . See, e.g., Choudhry, supra note 6. ↑

45. Levitsky, supra note 7, at 1996–97. ↑

46. Arend Lijphart, Consociational Democracy, 21 World Pol. 207 (1969); Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (1977). ↑

47. Brad R. Roth, Advancing Peaceful Settlement and Democratisation: The Doubtful Usefulness of International Electoral Norms, in International Law and Peace Settlements 333, 346 (Marc Weller, Mark Retter & Andrea Varga eds., 2021). ↑

48. See Donald Horowitz, Ethnic Power Sharing: Three Big Problems, 25 J. Democ. 5, 14­­–18 (2014). ↑

49. Roth, supra note 46, at 353. ↑

50. Fawwaz Traboulsi, A History of Modern Lebanon (2d ed. 2012). ↑

51. Eduardo Wassim Aboultaif, Revisiting the Semi-Consociational Model: Democratic Failure in Prewar Lebanon and Post-Invasion Iraq, 41 Int’l Pol. Sci. Rev. 108 (2020). ↑

52. Roth, supra note 46, at 347. ↑

53. . See, e.g., Tomasz Inglot & John Pelissero, Ethnic Political Power in a Machine City: Chicago’s Poles at Rainbow’s End, 28 Urban Affs. Q. 526 (1993). ↑

54. Roth, supra note 46, at 347. ↑

55. Donald L. Horowitz, Ethnic Groups in Conflict (1988). ↑

56. Anne-Marie Slaughter, Ranked-Choice Voting, Politico (2019), https://www.politico.com/interactives/2019/how-to-fix-politics-in-america/polarization/ranked-choice-voting/ [https://perma.cc/NXT2-T2BG]. ↑

57. . See Shelby Cnty. v. Holder, 570 U.S. 529 (2013); Brnovich v. Democratic Nat’l Comm., 141 S. Ct. 2321 (2021). ↑

58. Kelly M. Hoffman, Sophie Trawalter, Jordan R. Axt & M. Norman Oliver, Racial Bias in Pain Assessment and Treatment Recommendations, and False Beliefs About Biological Differences Between Blacks and Whites, 113 Proc. Nat’l Acad. Sci. 4296 (2016); Andrea Deyrup & Jospeh L. Graves, Racial Biology and Medical Misconceptions, 386(6) N. Engl. J. Med. 501 (Feb. 10, 2022). ↑

59. Deyrup & Graves, supra note 57, at 502. ↑

60. Henry Louis Gates, Jr. & Andrew Curran, DNA Provides New Language to Talk About Race, N.Y. Times, Mar. 6, 2022, at 7. ↑

61. . See generally Loving v. Virginia, 388 U.S. 1 (1967). ↑

62. Justin McCarthy, U.S. Approval of Mixed-Race Marriage at New High of 94%, Gallup (Sept. 10, 2021), https://news.gallup.com/poll/354638/approval-interracial-marriage-new-high.aspx [https://perma.cc/2A3R-CUV3]. ↑

63. . See generally Richard Alba, The Great Demographic Illusion: Majority, Minority and the Expanding American Mainstream (2020). ↑

64. Simon Romero, Navajo Nation Becomes Largest Tribe in U.S. After Pandemic Enrollment Surge, N.Y. Times (May 21, 2021), https://www.nytimes.com/2021/05/21/us/navajo-cherokee-population.html [https://perma.cc/XQ2L-BPB6]. ↑

65. Rwanda’s Constitution of 2003 with Amendments through 2015, art. 10 (country committed to “eradication of discrimination and divisionism based on ethnicity, region or on any other ground as well as promotion of national unity”). ↑

66. Rwanda Reconciliation Barometer 2020 xi (2021) (“[A]lmost all the pillars and indicators have seen the status of reconciliation progress between 2016 and 2020 . . . .”). But see Filip Reyntjens, From Ethnic Amnesia to Ethnocracy: 80% of Rwanda’s Top Officials are Tutsi, Afr. Arguments (Nov. 24, 2021), https://africanarguments.org/2021/11/from-ethnic-amnesia-to-ethnocracy-80-of-rwanda-top-officials-are-tutsi/ [https://perma.cc/2HPU-WS25]. ↑

67. Adam Liptak & Anemoa Hartocollis, Supreme Court to Hear Affirmative Action Challenges at Harvard, U.N.C., N.Y. Times (Jan. 21, 2022) https://www.nytimes.com/2022/01/24/us/politics/supreme-court-affirmative-action-harvard-unc.html [https://perma.cc/A6UY-HRXV]. ↑

68. Fabiola Cineas, Affirmative Action Just Failed in California—Again, Vox (Nov. 4, 2020),

https://www.vox.com/2020/11/4/21537590/california-proposition-16-affirmative-action-results [https://perma.cc/564K-9TNR]. ↑

69. See generally Saad Gulzar, Nicholar Haas & Benjamin Pasquale, Does Political Affirmative Action Work, and for Whom? Theory and Evidence on India’s Scheduled Areas, 114 Am. Pol. Sci. Rev. 1230 (2020) (showing how beneficiaries in named classes improve welfare after affirmative action). ↑

70. Jamelle Bouie, What Americans Really Think About ‘Critical Race Theory,’ N.Y. Times (Feb. 26, 2022), https://www.nytimes.com/2022/02/26/opinion/critical-race-theory-survey.html [https://perma.cc/32SN-XX6G]. ↑

71. Thomas O. Jewett, Thomas Jefferson and the Purposes of Education, 61 Educ. F. 110 (1997). ↑

72. Sanford Levinson, Our Undemocratic Constitution (2006). ↑