I’ve been wrestling with the class basis of German fascism. It’s a huge literature, I’ve hardly scratched the surface, but—in typical blogging ethos—here’s my summary.

I divide it into three parts: first, Nazi voters as primarily lower-middle-class Germans; second, Nazi voters as a broad cross-class coalition; and third, Nazi voters as assembled by economic incentive. This trichotomy is neither complete nor precise, and I ignore religion in favor of focusing on class.

If this synthesis achieves anything, it is to show just how difficult it is to perform inference under profound data scarcity. There were no opinion surveys in Weimar. Nor were there individual voting records. Everyone therefore uses electoral results by place to make inferences about individuals. Let’s say neighborhoods with more wealthy people voted disproportionately for Nazis; perhaps more wealthy people were Nazis. But perhaps not—maybe there’s something distinct about non-wealthy people who lived near wealthy people. Even employing modern statistical techniques, it’s hard to infer individual characteristics of Nazi voters.

First wave: The middle-class thesis

Rudolf Heberle’s work is unusual: it’s contemporaneous. He was the first to quantitatively study Nazi electoral patterns, and the only to do it starting in 1932, as Nazis marched the streets.

Heberle combined three types of data. He had official election returns from 1919 to 1932, mostly at the county level. For 18 rural counties, he obtained records of the share of agricultural workers employed on family farms, large farms, and estates, as well as occupational statistics.

Nazi support was strongest where small family farms were predominant and weakest where wage laborers were predominant. Socialist support exhibited the reverse pattern. Similarly, Nazi support was stronger where there were more proprietors and fewer wage earners, and socialist support went the other way.

Why were small proprietors particularly susceptible to fascism? Wage laborers “held firmly” to socialist ideas. Unions cultivated a relatively stable social view. Family farmers, meanwhile, felt threatened from above—by banks and big business—and from below, by unions and socialists. Nazis promised protection against both. “The main mass support came,” the Heberle concludes, “from the middle layers of society.”

Seymour Lipset echoed the same ideas in Political Man, arguing fascism was a distinct expression of the lower middle class—“extremism of the center.”

Second Wave: From middle-class to cross-class

Richard Hamilton set Lipset as his foil. German fascism, he believed, was not primarily a lower-middle-class phenomenon. Analyzing 14 cities, Hamilton found that areas with lower-middle-class folks were not conspicuously more fascistic; more affluent neighborhoods exhibited higher fascist support. “The ‘best districts,’” Hamilton concludes, “gave Hitler and his party the strongest support.”

Thomas Childers was unconvinced by Hamilton’s critique. Childers found that areas with shopkeepers, artisans, and small farmers were the most consistent bases of Nazi support, as Lipset had predicted. The Nazi coalition, however, was a lot broader than small proprietors alone. Nazis pulled significant support from civil servants, pensioners, and workers in small workshops. The nucleus of Nazi support was made up by small proprietors, Childers concludes, but the Nazi party had also become the “catchall party of protest.”

Jürgen Falter, in 1991, completed this second wave. He assembled a comprehensive national election dataset supplemented with statistics on occupation, class, religion, industry, and other variables. Falter finds that the Protestant old middle class—the artisans and proprietors emphasized by Lipset and Childers—were overrepresented among fascist voters. Nazis primarily pulled support from former liberals, conservatives, and nonvoters, with fewer socialists switching to the fascist side. Yet, because the Nazi coalition was far broader than just the old middle class, Falter agrees with Childers: the Nazis were a catchall protest party with broad cross-class support.

Third Wave: Economic incentives

Several decades later, Gary King, along with three coauthors, were unsatisfied with Falter’s conclusion. The catchall theory “applies to most groups and almost all big or growing parties in almost all countries.” Every large party, in other words, is a catchall party. (Falter responded that, to the contrary, most Weimar parties were socially homogeneous.) The coauthors asked which groups moved disproportionately towards Nazis from 1928 to 1932. They found that the “working poor—self-employed shopkeepers and professionals, domestic workers, and helping family members”—disproportionately moved towards Nazis. Self-employed voters, the authors proffer, were incentivized by Nazis’ promise to protect property and extend cheap credit; agricultural workers were incentivized by protective tariffs and debt relief. The unemployed, by contrast, moved disproportionately towards communists, whose policies appealed to them more directly.

In 2024, Björn Brey and Giovanni Facchini found that areas particularly affected by declining exports shifted less towards Nazis, counter a simple economic-pain-to-fascism pipeline. The effect was mediated by the type of worker. Counties with blue-collar workers experienced a larger swing towards Nazis; counties with white-collar workers saw a smaller swing. Brey and Facchini argue that Nazi public-works policies were advantageous to blue-collar workers but benefits-cutting policies threatened white-collar workers. An economic shock, the authors conclude, does not secure a political outcome on its own; “rather, the proposed policy response to deal with the aftermath of the shock plays a keys role.”

Closing Thoughts

When Heberle wrote in the 1930s and 1940s, he personally collected or tabulated all of his data and principally relied on small-N correlations. Thanks largely to Jürgen Falter and his collaborators, we now have relatively comprehensive national election, occupational, and demographic data. We also have more sophisticated statistics.

Better data and statistics have not produced a wholly different Nazi voter. Small proprietors remain central to the Nazi coalition and effectively everyone agrees that Nazis assembled a broad coalition. The questions that remain are those on the margin: who moved towards Nazis more, who moved towards Nazis less, and why.

Nearly a century after Heberle began studying Nazi voters, the field remains defined by the absence of individual data.

Ours has been called the “anti-social century,” marked by “spiritual crisis.” Among young Americans, time spent physically with friends fell 70 percent from 2003 to 2020. The share of Americans who report having three or fewer close friends has nearly doubled since 1990, while the share of children who meet friends daily has halved. Men who watch television spend seven times more time in its unblinking company than in the company of blinking humans. Women with pets spend more time with them than with fellow political animals.

It’s not just a hit on health and happiness—it’s a political phenomenon, too. In 2016, Yoni Appelbaum wrote that Trump appealed more to the “disconnected” than the “disaffected.” His article was based on a survey that found civically disengaged Republicans were considerably more likely to support Trump.

Michael Bender followed some of Trump’s most fanatical supporters. Many of them—people who trailed Trump from one rally to the next—were retired and had “little to tie them to home.” Reviewing the book, Michelle Goldberg concludes that a “socially healthy society would probably never have elected Trump in the first place.”

Derek Thompson, in his landmark essay on loneliness, devotes a section to “your politics on solitude.” In her 2023 Atlantic piece, titled “The Weaponization of Loneliness,” Hillary Clinton describes how Steve Bannon activated an army of “rootless white males.” “We’re enmeshed in some sort of emotional, relational, and spiritual crisis, and it undergirds our political dysfunction and the general crisis of our democracy,” writes David Brooks. “For people who feel disrespected, unseen, and alone politics is a seductive form of social therapy.”

They build on a prestigious lineage. Alexis de Tocqueville thought associations—formed to establish seminaries and schools, build inns and churches, administer hospitals and prisons—were the lifeblood of American democracy: “In democratic countries the science of association is the mother science.” Associations transform isolated, weak men, powerless against despots, into a powerful collective. “Americans of all ages, all stations in life,” he observes, “are forever forming associations.”

In the shadow of the Holocaust, Hannah Arendt, William Kornhauser, and Theodor Adorno attributed fascism to atomization and alienation. Adorno observed that a fascist radio host appealed to Americans suffering “despair, isolation, and loneliness.” People “available” to mass movements, explains Kornhauser, are those with “the fewest ties to social order.” Arendt is even clearer: totalitarian movements are “mass organizations of atomized, isolated individuals.” “The chief characteristic of the mass man”—the fanatic, the fascist—“is not brutality and backwardness, but his isolation and lack of normal social relationships.” The movement acts as an ersatz community, a “fictitious home,” the externalization and projection of the disoriented. Loneliness is “the essence of totalitarian government.”

Robert Putnam brought this story into the 21^st century. Associations “broaden the participants’ sense of self.” The groups do not have to be political; even a choral society or bird-watching club can teach “an appreciation for the joys of successful collaboration.” Neighborly communities, once cradles of social trust, now wither.

But Trump supporters are not particularly alienated. Appelbaum based his conclusion on a single poll without any controls. A poll taken a few months later found no relationship. A 2025 poll again found no relationship between number of close friends and support for Trump. Even the poll Appelbaum used weakly supports his conclusions: Trump was more popular among Republican disconnected voters, but across parties, a larger share of disconnected voters favored Clinton than Trump.

Perhaps it’s not a matter of individual loneliness but of community cohesion. Andrés Rodríguez-Pose, Neil Lee, and Cornelius Lipp test this theory. They measure cohesion using a composite index that includes, among other things, labor unions, sports clubs, and golf courses per capita. Putnam should expect that less cohesive counties, all else equal, experience a bigger Trump swing.

But no. Communities that most swung toward Trump—those experiencing employment and population decline—are “precisely those where social capital—the very form of capital that, according to Putnam (2000), was supposed to provide the glue for America’s democratic institutions—has held stronger than elsewhere.”

On December 18, 1931, Werner Willikens, a prominent Nazi, was elected to a newly created fourth seat on the presidium of Germany’s largest agricultural organization. The Reichslandbund (RLB) had well over a million members, linked by local offices, agricultural newspapers, and professional staff.

The following spring, in the second round of the presidential election, the RLB openly supported Hitler. In November, its managing president joined a petition calling on President Hindenburg to appoint Hitler chancellor. When Hindenburg appointed General Schleicher instead, the RLB mounted a campaign against his government, in close coordination with Nazi officials.

Willikens’s election was the product of systematic Nazi infiltration. In August 1930, Richard Darré, Hitler’s agricultural advisor, began constructing the Agrarian-Political Apparatus, or ApA. Darré described it as “a weapon” to be employed “with utmost force in agrarian-political wars”—really, a network of Nazi farmer-whisperers.

Darré instructed advisors to “regard every non-National Socialist organization . . . as a fortress to be conquered.” In 1929, the party had forbidden Nazis to join the RLB. The RLB was closely connected to the established nationalist right, a political rival of the Nazis. But after the Nazi electoral breakthrough, Darré concluded the RLB was of more use captured than killed. Hitler called on Nazi farmers: “Into the Landbund!”

In November 1931, Darré met with the RLB director, dubiously claiming more than half of the Landbund’s members were Nazis. He threatened to form a competing Nazi agricultural association. By the end of the meeting, Nazis had been promised a seat on the presidium; Willikens was elected six weeks later.

Historian Horst Gies argues that the Nazi-infiltrated RLB helped persuade Hindenburg to appoint Hitler—making Darré’s ApA one of the most important political machines behind Hitler’s rise. Hitler, in a letter to Darré, expressed a similar belief: “The incorporation into our movement of the millions who made up the German peasantry . . . contributed substantially to creating the conditions for National Socialism’s legal conquest of power in Germany.”

The associational life of Weimar Germany, writes Sheri Berman, was “extraordinarily vigorous.” The Germans had a word for it, naturally: Vereinsmeierei, roughly associational “fetishism” or “mania.” Retailers, bakers, gymnasts, singers, and folklorists all assembled into clubs. Contemporaries joked that if three Germans met at a bar, they began drafting bylaws.

These networks did not protect against fascism—they helped facilitate it. At a moment when institutions were young and weak, the “vigor of associational life . . . served to further undermine and delegitimize the republic’s political structures.” The groups offered an alternative to national political parties; they “created a sense of purpose and transcended socioeconomic divisions.” The fragmentation of political power across small parties is often noted, but there was a symmetric vertical fragmentation; even successful parties had only a loose hold on coalitions of subparty organizations. “The result,” writes Berman, “was a highly organized but vertically fragmented and discontented society that proved to be fertile ground for the Nazi’s rise.”

These groups not only weakened national parties but also provided an opening for disciplined outsiders. This is the point of the RLB story. Put simply, it’s easier to organize a society of several hundred organizations than 65 million individuals. Nazis were “an unusually activist bunch” who exploited the latent political power of the associations. They came to power “not by attracting alienated, apolitical Germans, but rather by recruiting highly activist individuals and then exploiting their skills and associational affiliations.”

After Hitler became chancellor, these same organizations enabled his rapid consolidation of power. The RLB leader—who expected influence in the regime he helped bring about—resigned after Nazi newspapers accused him of speculation. The Gestapo arrested the leader of agricultural cooperatives—which had proven hard to capture without state power—and Darré seized control by threat of force. By September 1933, all agricultural groups were either dissolved or folded into the Reich Food Estate, headed by Darré personally. The entire agricultural web: united, subjugated, and instrumentalized.

Social alienation is not good for democracy—what is unhealthy for a person is usually unhealthy for a polity. But social connectedness is not straightforwardly good for democracy either.

What matters for democracy is a certain kind of social life—the kind that assembles into stable national politics. Associations that transcend political boundaries are depolarizing. That’s good—so long as they don’t undermine the efficacy of the national system. Robust associations connected Germans across party and class, but they failed to create a national equilibrium. They eroded national parties and created a new path to power. Through associations, and on high.

It feels plausible: Nazis as lonely people, lost in a newly industrialized world, searching for meaning in uniform. Trump voters, too: lonely and lost, voting for their man, but voting alone. I want it to be true. I’m sure Arendt and Adorno did too. Then these expressions, these sick expressions of sick societies, whether clad in Hugo Boss or drenched in gilded extravagance, would be different from ordinary politics. If only.

Parties have been growing weaker for decades. A political party still acts as a nexus of rules, funding vehicle, and electoral heuristic. But it exercises far less authority over policy coordination and candidate choice.

1. Associational party building: Parties should build permanent local offices, offer citizen services, and build local relations.

Two scholars call for “associational parties”, defined as “organizationally robust intermediaries that allow for meaningful interactions between elites and everyday members of the electorate” (the idea was picked up by a 2026 report). Concretely, they want parties to build year-round local offices, hold public meetings, partner with civic organizations, and create voter feedback channels. They briefly trace the “hollowing out” of political parties via TV/internet, democratized primaries, and Super PACs. An advantage of their proposal is that it does not require legislation. They hope to increase the influence of ordinary voters over activists and donors.

It’s an attractive picture—parties as embedded civic organizations, citizens who show up. At times, however, it feels like political nostalgia, colorized and imported to our time. Their all-too-brief treatment of history is sufficient to see the problem: the internet and direct primaries are here to stay. And parties are hardly the only declining local organization. I’m skeptical you can fundamentally change party organization without changing rules, since parties are probably ~rational players in the political economy.

But maybe I’m being cynical. It would be great if parties could cultivate more member social investment. And, if there is a revival of local society, parties should do everything to be a part of it.

2. Institutional veto: Parties should require candidates to obtain endorsements from party officials before entering the primary.

After the 2016 election, roughly 20,000 women, interested in running for office, reached out to EMILY’s List. That’s notable because it’s a lot of women—but also because an independent organization had become a portal into electoral politics.

Jonathan Rauch and Raymond La Raja trace how third-party groups gained influence in both parties: the rise of individual donations to candidates, accessibility of media to insurgents, decline of trust in parties. This has created a thriving market for subparty associations that recruit, train, fund, staff, and legitimize candidates (Club for Growth, Tea Party, Justice Democrats, Run for Something, Indivisible, etc). As subparty groups have become core organizational infrastructure, ideological groups and individuals have become primary funding vehicles. In 1990, business PACs provided 62 percent of all PAC funding in competitive primaries. By 2016 they provided just 22 percent, and individual contributions now dwarf PAC contributions in competitive primaries.

All of this, Rauch and La Raja argue, reinforces polarization. When the party—judged by its ability to win and govern—relinquishes control of primary candidates to subparty groups, whose aims tend towards ideology and activism, the result is predictable.

The authors suggest a number of possible interventions. One possibility is pre-primary party endorsements, which might then be printed on ballots. The stronger rule would be to require candidate approval of some number of party insiders before primaries.

The report was written a decade ago but could’ve been written yesterday. The story feels important and true, and since 2016, the DSA has grown especially powerful on the left. The right is more muddled. Trump might be the single best argument for stronger gatekeeping—but, paradoxically, that would enable the now-victorious Trump to cast a longer shadow over the party. And there’s the obvious objection: it’s not, in the traditional sense, very democratic.

Worth thinking about. Two additional thoughts:

First, a 2026 report from the Bipartisan Policy Center corroborates many of the findings. Parties avoid spending on safe seats, creating space for subparty groups to challenge incumbents. Party operatives report that candidate recruitment is shared with “funded interests”—subparty groups. Between elections, some state parties are essentially reduced to a bank account, while subparty groups continue engaging. Not one of 12 interviewed political operatives believed the party should act as an ideological gatekeeper.

Second, it’s become conventional wisdom that primaries polarize because primary voters are more extreme. But the empirical tests don’t support a strong conclusion (although there is some evidence). Maybe polarization is caused less by primary voters, than by the menu from which they choose. The menu, after all, is increasingly prefixed by organized activists.

3. Public financing: Provide public funding for political parties.

Most people agree large donors wield asymmetric political power. But small donors, often held as a solution, are equally extreme to large donors, and probably contribute to polarization. Fundraising messages tend to be toxic. Candidate-centered fundraising substitutes fundraising capacity for electoral capability.

Constitutional jurisprudence limits regulation of campaign finance. If we can’t limit donations, we can make them less important. Providing some public vehicles for political parties could help rebuild some of the capacity subparty organizations have taken over, and reduce their need to appeal to their base. There are a whole bunch of questions here—which parties qualify, how much do they get, etc. Lee Drutman proposes that every party that clears some threshold should receive funding proportional to its level of public support.

It would be a substantial public expenditure. Parties may continue to compete for funding. And it’s not clear how much the under-institutionalization of political parties is a money problem.

1924. A national hate group—which aims to solve the Jewish question, if necessary by elimination—boasts millions of members. Signs appear outside social clubs: “No Jews, Dogs, or Consumptives”. Universities erect quotas.

1929. The Great Depression hits as hard as anywhere. Industrial production halves, unemployment reaches 25 percent.

1938. The most prominent religious leader in the country broadcasts to millions. In his first broadcast after Kristallnacht, the priest explains that Jews are “a powerful minority in their influence; a minority endowed with an aggressiveness and initiative.” Most citizens partly blame Jews for the persecution of Jews.

Except fascism never comes. This is the United States.

America had economic collapse, organized racial violence, mass antisemitism, radio demagogues, and protofascist organizations—why did the center hold?

Daniel Ziblatt offers one explanation. German fascism succeeded partly because extremists captured the country’s mainstream conservative party and drove out moderates. Once a barrier against fascism, the party became the bridge to it.

The collapse of the institutional right as a direct cause of German fascism.

Reichstag, June 27, 1922. The chamber’s dark wood paneling adorned by wreaths of red roses and blue hydrangeas. A plain oak coffin. In it Walter Rathenau: liberal, scholar, Jew, the nation’s foreign minister, assassinated by nationalist antisemites. Deputies from all sides, communists and monarchists alike, dressed in solemn black.

If democracy comes in moments, this is one of them.

Two years after the funeral, Weimar entered a four-year “golden age”. Cabinet turnover slowed, currency stabilized. The right-wing party, the DNVP—founded as a monarchist opponent of Weimar—behaved as a mainstream conservative party. It entered coalition governments. Half of its deputies supported a reparation plan, heresy among nationalists. Most strikingly, the officially still monarchist party supplied decisive votes for legislation banning monarchist organizations.

The erosion of the institutional right opened the way to Hitler.

Between 1928 and 1930, the DNVP’s vote share fell from 14 to 7 percent. Nazi support moved the other way, from 3 to 18 percent. About a third of 1928 DNVP voters supported Nazis two years later, and Nazi support grew most where the DNVP declined most.

DNVP leaders also legitimized Nazis. In 1929—with Hitler’s party still marginal—the DNVP included Hitler as equal partner in a national campaign against reparations. In 1931, the DNVP organized a nationalist rally at which 70,000 SA men displayed fascist military discipline.

The DNVP’s electoral collapse and radical turn made a center-right coalition acceptable to President Hindenburg impossible. Presidential governments increasingly ruled by emergency decree. A few years later, Hitler would do the same.

The DNVP had once functioned as a “border party”, keeping potential radicals attached to parliamentary politics. It had become a bridge to fascism.

The German right was structurally vulnerable to capture.

Heinrich Class led the ultranationalist Pan-German League. Alfred Hugenberg was an industrial executive and media magnate who controlled Germany’s largest private news agency. Incensed by the party’s moderation, they staged a takeover.

Power lay with 45 regional associations, which set candidate lists and elected the party chairman. Many were underfunded and lacked mass participation. Class had the people: his Pan-German League had around 16,000 disciplined members. Hugenberg had the capital: he controlled at least 30 percent of the DNVP’s 1928 spending. In October, Hugenberg was elected chairman unopposed.

Hugenberg quickly consolidated power. He replaced moderate regional leaders, threatened to expel dissenting deputies, and asserted the authority to dictate parliamentary votes. His radical turn alienated core elements of the party’s coalition, accelerating the DNVP’s electoral collapse.

Limits of the argument

Hugenberg did not initiate the DNVP’s decline; its support had already fallen from 21 percent in 1924 to 14 percent in 1928. Nor can the causal direction be cleanly inferred: the DNVP’s collapse created space for Nazism, and Nazism accelerated the DNVP’s collapse. And the opening did not necessarily have to benefit extremists—under different conditions, it could have been filled by moderates.

American protofascists never acquired a common political vehicle.

Perhaps 100,000 Americans belonged to explicitly anti-democratic organizations during the 1930s. The KKK enrolled several million earlier in the interwar period. Father Coughlin retained millions of listeners even after his broadcasts became openly antisemitic and sympathetic to Hitler. Antisemitism extended further: even in the late 1930s and 1940s, 10 to 15 percent of Americans favored severe measures against Jews, and more than half of Americans partly blamed Jews for their own persecution. Yet these constituencies never assembled into a common political movement.

Roosevelt retained much of the potential insurgent constituency.1

Huey Long and Father Coughlin were the most formidable American populists of the interwar period. Both advocated extensive economic intervention and redistribution, militantly opposed communism, and commanded enormous followings.

Yet they had a “single, debilitating weakness”: the “inability to wean their followers from Franklin Roosevelt”.2 Half of survey respondents who approved of Coughlin in 1936 reported voting for Roosevelt.3

The New Deal appropriated some of their most salient arguments. In proposing a more graduated tax plan, President Roosevelt was partly motivated by “stealing Long’s thunder”. FDR’s utility regulation and monopoly-busting competed with Long and Coughlin’s polemics against economic concentration.

White supremacists also found an institutional home in Roosevelt’s Democratic Party. Agricultural and domestic workers—disproportionately Black—were excluded from New Deal protections to accommodate Southern Democrats. White supremacists remained integrated into the national party. In 1936, Roosevelt won 87 percent of the vote in Georgia, 97 in Mississippi, and 99 in South Carolina.

The strength of Roosevelt’s hold on the populist electorate was never fully tested: Huey Long was assassinated before an anticipated 1936 campaign.

The Republican Party was difficult to capture.

The Republican Party was in trouble. In 1932, Roosevelt swept 42 states and Democrats won three-quarters of House seats. Democrats again defied political gravity in the midterms. 1936 was worse yet: FDR carried all but two states, while Republican congressional representation fell under 20 percent. Republicans were sidelined from national politics and nearly broke—the RNC ended the 1934 campaign with $352 in liquid assets.4

Father Coughlin sensed opportunity. In May 1936, Coughlin said he hoped for “a renovated Republican Party” of “contrite heart”. Hoover aides, eager for his anti-Roosevelt legions, approached the priest (some, however, were wary: one aide warned that encouraging Coughlin was “playing with nitroglycerin”). The romance was short-lived. Alf Landon, a bitter opponent of Coughlin’s financial program, won the nomination.5

Why didn’t Coughlin attempt a takeover of the Republican Party? Whatever his reasons, orchestrating a takeover would have required an organization Coughlin did not have.

The Republican Party offered no single point of capture. The representatives who chose the RNC chair were selected through 53 separate state and territorial party systems. The presidential nominee was chosen by convention delegates. Delegates emerged from a state-by-state patchwork of primaries, caucuses, conventions, and committees. Legislative candidates were generally selected through direct primaries. Party authority was therefore divided among state organizations, primary voters, and the RNC.

Scale was also forbidding. More than 3.3 million voters participated in the 1936 Republican presidential primaries—five times the DNVP’s entire membership—even though those contests selected only about a third of convention delegates. A national insurgent had to compete both for a mass primary electorate and within state party processes.

Coughlin had millions of listeners and a loose political association. But he did not have Class’s disciplined ideologues, cadres embedded in state parties. Nor did he have Hugenberg’s extraordinary wealth. And the Republican Party would have been a far grander target than the DNVP.

Nor could Coughlin bypass the major parties.

After Landon won the Republican nomination, Coughlin quickly endorsed William Lemke, the candidate of the new Union Party. Despite Coughlin’s enormous audience, Lemke secured just 2 percent of the national vote. The structural domination of two parties has been a persistent fact of the American system: plurality elections, ballot access laws, and the Electoral College reinforce the pressure to organize within major parties. Unlike Weimar’s proportional system—where roughly 60,000 votes could yield a Reichstag seat—an American radical movement unable to capture a major party was likely to remain marginal.

When he took over Weimar’s largest right-wing party, Alfred Hugenberg was not popular. His ascension may have been a fatal blow—a rational and self-interested party would have spit him right back out.

But a party is as it is constituted. The DNVP was constituted as a confederation of regional associations. With a little dedication and a lot of resources, these associations could be controlled. Once captured, the party ceased containing the radical right and began legitimizing it. Hitler had waited long enough to get lucky: the depression, the right’s collapse, the fragmentation, Hindenburg’s imprudence.

In October 1935, Sinclair Lewis, pen in hand, imported fascism to America. It took little imagination. His antagonist is unmistakably Huey Long. The priest who legitimizes and endorses the fascist is a renamed Father Coughlin. The book’s title: It Can’t Happen Here.

It could, of course, and that was Lewis’s point. But it didn’t. It didn’t because of the enormity of Roosevelt’s coalition. It didn’t because of the mundane complexity of party machinery. The left accommodated, and the center held.

Subscribe now

In February 1939 some 20,000 people assembled in Madison Square Garden. George Washington stands, haloed by a cone of light, between swastikas.

How many Americans were fascists, sympathizers, or anti-Semites? The short answer (approximations within order of magnitude; America’s population ~130M):

1. About 100,000 Americans were part of an expressly fascist organization.

2. Major far-right organizations enrolled memberships in the millions.

3. Before Pearl Harbor, about 2 - 5% of Americans supported Hitler or Mussolini.

4. About 10 - 15% favored severe action against Jews.

5. About half blamed Jews, wholly or partly, for their own persecution.

Few backed Hitler, but many embraced antisemitism.

Eleven national polls reveal a large gap between explicit fascist sympathy and broader antisemitism. Assembled here:

Fascist orgs had ~100,000 members.

• Black Legion: With “60,000 to 100,000 armed and disciplined men,” the Legion “had become the most formidable nativist organization around.”1

• Bund: German-Americans for Hitler.2

• Silver Legion: Aimed to “lead a nationwide movement of Silver-Shirted men to drive the Jews and revolutionaries from power!”3

• Fascist Circles: Italian-Americans for Mussolini.4

• Christian Front: They “saw themselves as the advance guard in a holy war against Communists and Jews.”5

Far-right movements reached millions.

This data I’d put the least weight on—historical accounts not only disagree on membership numbers, they also disagree on the political identities of the groups.

• Mothers’ Movement: Characterized by one historian as a “mass movement of profascist women”6 but more cautiously a “loose coalition” of conservative anti-interventionists.7

• KKK: Capitalized on “white Protestant fear of Catholicism, Judaism, and racial amalgamation.”8

• National Union for Social Justice: Combined populism, Catholicism, and radical monetary reform.9 Included because of Father Coughlin’s personal control & authoritarian bend.

• Committee of One Million: Gerald Smith’s Christian-nationalist organization—Coughlinites, antisemites, and antiwar activists.10

• America First Committee: Anti-interventionist coalition with an influential antisemitic far-right faction.11

Andy: I have a headache, I’m seeing double, I have a song stuck in my head and I’m hungry.

Nurse (Anne): Okay, well, some of those things are symptoms and some of them are just being a person.

(from Parks and Recreation)

It’s common among certain journalists and academics to read Adorno to understand Trump. He is after all among the most important theorists of fascism. Adorno I mean not Trump.

In the half-nation of West Germany Theodor Adorno transcended the detached stature of most theorists. Jewish, born in Frankfurt, Adorno returned from American exile in 1949. He was a staple guest on radio shows, speaking of music and popular culture but with most urgency of the cause and meaning of fascism. He helped forge a German identity around institutions over blood, an Americanized nationalism: and became therefore “the postwar conscience of Germany.”

Psychological Technique is Adorno’s analysis of Martin Luther Thomas—radio host, pastor, fascist, man of many hats. Adorno dissects the patterns of his speech and the causes of his appeal.

I must say my reading shed little light on fascism. It shed light instead on the limits of social philosophy, at least as practiced here.

The problem is it’s not clear what is distinct to fascism. Adorno accuses the radioman of presenting his listeners with a “fait accompli”, as if his political movement has already won. This serves the fascist’s purpose because people would like to be part of a winning movement. Fine. But a different Martin Luther—Martin Luther King—insisted the arc of history bends towards justice. This too declares inevitability. “We’ve got a shot” doesn’t ring the same.

In several places it’s easy to recognize Trump. The radioman often spoke of having no institutional backers. He stewed in the politics of nostalgia. Most resonantly, Adorno reports the pastor weaponized aversion “to any kind of expertness.”

But claiming financial independence is a feature of anti-institutional politics left right and center. Nostalgia politics—the last “A” of MAGA—is conventional conservatism. And anti-expertise was common among the American left after the Vietnam War.

Psychological Technique feels astrological: it’s a long list of descriptions that makes projection easy.

The same is common in analyses of Trump. A 2019 article used this very book to compare Trump’s rhetoric to the radioman’s. A 2025 study classifies Trump’s tweets by which fascist tendency they exemplify (promoting American manufacturing, for instance, evidences a fascistic cult of tradition).

If the first problem is the lack of a control, the second problem is falsifiability.

The theory Adorno develops in Psychological Technique is a contrast of rationality: fascist propaganda is rational—even ingenious—while fascism itself is irrational. Fascist leaders are “psycho-technics”; there’s a “shrewd soberness behind the drunken words.” The fascist crowd, in contrast, is irrational. They hope to cure loneliness and resentment by trading autonomy for collective release (a desire incubated, Adorno explains, by industrial society dislocating and suppressing the individual). Since fascism could never actually cure those feelings, fascism is irrational.

It all sounds true and that’s half the problem. It feels like Adorno is describing an aesthetic: the efficiency and order—the “rationality”—of German Nazis, contrasted by the mass hysteria (“irrationalism”) that carried them. An aesthetic can’t be tested.

(Adorno once debated Popper on falsifiability in science.)

The book is useful as a document of a particular fascist. But without a control, without falsifiability, it infinitely describes and infinitesimally explains. It’s a portrait, an aesthetic characterization, not to be confused with science.

In this conversation:

• 1819: Black Belt planters vs North Alabama plain folk at statehood

• Some plain folk refuse Confederacy’s war

• Redemption, planters take the state back

• 1901: planters disenfranchise black Alabamians and many poor whites

• 1921: black and white miners defeat another disfranchisement amendment

• Big Jim Folsom: the plain folk find their champion

• 1946: another disfranchisement amendment, Boswell, passes

• Brown makes race personal; plain folk coalition cracks

• 1958: the planters win

A War of Sections is available on UGA Press or Amazon. Follow Steve’s Substack.

Bernard Grofman and Gary King were assigned homework by the Supreme Court of the United States. Naturally they determined to complete the assignment.

Since 1986, the Court had held that partisan gerrymandering could, in theory, violate the Constitution.1 But it had never come close to invalidating a partisan map. The obstacle was the absence of a clear, objective standard. Evaluating a 2003 Texas gerrymander, the Court maintained the same position.

But hidden in the details was a slight shift. Four Justices fixed on a way to measure gerrymandering, a metric called partisan symmetry.

A fifth Justice, the swing vote, remained unconvinced. The problem, wrote Justice Kennedy, was it lacked a legal threshold—how much asymmetry is too much? Second, measuring symmetry requires simulating hypothetical elections. Kennedy would not invalidate a map based on “a hypothetical state of affairs.”

Still, Kennedy did not “altogether discount” symmetry. Solve these problems and he’d reconsider.

Bernard Grofman and Gary King, among the two most prolific political scientists in the country, had helped develop the metric. Victory was one Justice away, and Kennedy had just issued the challenge.

It was their move.

Courts can enforce speed limits because most people agree: speed is distance over time. Anyone who hoped to cure gerrymandering would first need to define it, and Grofman and King had been trying for decades.

The key concept is symmetry, also called bias. It asks whether the same votes produce the same seats for both parties. If Democrats win 7 of 10 seats with 60% of votes, but Republicans would win only 6 seats with 60%, the system is biased.

Grofman proposed measuring bias across a range of plausible outcomes.2 But that required knowing how many seats a party would win under vote shares that had not actually occurred. If Democrats win 55% of the vote and 6 of 10 seats, how do we know how many seats they would win with 54% of votes? Or 40%?

For years, political scientists used a shortcut, shifting every district by the same number of points and seeing which seats flip. This assumes that any statewide shift would hit all districts equally, a “uniform swing.”

Gary King, working with a colleague, relaxed that assumption. Instead of using uniform shifts, they estimated how each district would move under different statewide outcomes.3

For decades they labored in the academic wilderness. Over twenty-five years, Grofman published 10 papers on partisan symmetry; King published 11.

Now, finally, they had the attention of the highest court of the land.

In The Rights Revolution, Charles Epp argues that judicial change depends on people like Grofman and King.

For 125 years after the Bill of Rights, and 50 years after the 14th Amendment, the Court “largely ignored civil liberties.” That changed around 1917, two decades before the Court’s liberal shift. What caused the change, Epp contends, was the infrastructure around the Court.

Before 1915, only large businesses could sustain prolonged constitutional litigation. With the NAACP, ACLU, and AJC, among others, civil-liberties litigants began developing comparable capacity.

Around the same time, legal education shifted from apprenticeship to law schools, fostering a more critical legal culture. Professors developed new constitutional theories, the private bar grew more sophisticated, and civil liberties organizations increasingly drew on these networks.

The U.S. rights revolution, in short, “developed within a broader political economy of litigation.” Scholars helped breathe life into Bill of Rights.

In 2007, Grofman and King turned in the assignment. It was their first article together.

Partisan Symmetry as a Judicial Test for Gerrymandering is not typical scholarship.4 It doesn’t say anything new. It reads like an open letter to the Supreme Court, a memo to a law clerk.

The coauthors make their best case for partisan symmetry. There is, they declare, an academic consensus: “We are aware of no published disagreement or even clear misunderstanding in the scholarly community about partisan symmetry as a standard for partisan fairness.”

They explain concepts patiently, for the quantitatively disinclined, then address Kennedy’s objections one by one.

Addressing the where-to-draw-the-line problem, Grofman and King offer Kennedy five possible thresholds.

On hypotheticals, the authors explain that partisan symmetry requires estimating the relative vote share across districts. Gerrymandering itself depends on the same exercise. Mapmakers pack opposing voters into landslide districts and spread their own voters across winnable ones. This requires knowing the partisan lean of different parts of the state. If this was unknowable, partisan gerrymandering itself would be impossible.

They end in hope. “The Supreme Court has seeded the clouds after this 20 year drought. We think that, at last, there can now be a non-trivial probability of rain, and as a result to a new flowering of American democracy.” Non-trivial probabilities aside, it’s a nice line.

Then the academic consensus cracked.

There was precedent for what Grofman and King wanted—a generation earlier, the Supreme Court had remade American democracy.

Until 1964, legislative districts were hugely unequal in population. A 38-person town in Vermont had the same representation as Burlington.

Within days of Reynolds, Michigan drew new districts. More than half the state’s senators retired or lost their primaries. Within two years, nearly every state had followed. Los Angeles went from one senator to fourteen.

The Court accomplished this with what now looks like a simple rule: legislative districts must be “as nearly of equal population as is practicable.” One person, one vote.

But the simplicity is retrospective.

For one, population equality competes with other theories of representation. American institutions, Alexander Bickel argued, historically represented “not only people, but interests, groups, and regions.”5

Even if votes should carry equal weight, one-person-one-vote does not guarantee fair representation. States might respond with at-large elections, allowing a bare statewide majority to capture every seat.6

Scholars canvassed many possible standards. Courts could cap the ratio between the largest and smallest district, the maximum or average deviation from an ideal district, or the percent of the population able to elect a majority.7

One person, one vote—population equality—was the most radical possible outcome. Even supporters thought it unlikely the court would constitutionalize the principle.8

But that’s exactly what the court did.

Grofman and King were not the only scholars who rose to the challenge. Eric McGhee and Nicholas Stephanopoulos fashioned a new metric, custom-built to meet Kennedy’s criteria.

McGhee criticized partisan symmetry for putting weight on implausible outcomes. If Democrats earn 60% of votes and every seat in San Francisco, it doesn’t matter that Republicans would theoretically achieve the same outcome if they somehow won 60%. They never would.9

Stephanopoulos and McGhee introduced the efficiency gap. It compares the votes “wasted” by each party. A “wasted” vote is one either cast for a losing candidate, or every surplus vote, over 50%, cast for a winning candidate. In a 60-40 district, the winner wastes 10% of the total vote and the loser wastes 40%. The efficiency gap is intuitive, requires no hypotheticals, and looks administrable.

Then came the objections. Two mathematicians observed that the metric “fetishizes” 75-25 districts but penalizes proportional outcomes.10 Three economists predicted EG would reduce competition and increase polarization.11 A political scientist argued that surplus winning votes are not equivalent to losing votes.12

The efficiency gap was one of many metrics that emerged around this time. A pair of political scientists proposed repurposing an old statistical measure of skew.13 A mathematician suggested borrowing from angular geometry.14 A redistricting workshop for mathematicians attracted 1,000 applicants.15

Each new metric arrived as a possible solution to Kennedy’s challenge, and also underscored his point; there is no singular answer.

Can courts bring about social change? If not, there would be little point to the whole circus. But there’s a growing band of skeptics.

Gerald Rosenberg posed the question in The Hollow Hope, and answers a faintly qualified “no.” His central target is Brown, his thesis captured by a two-word epigraph: “nothing happened.”16 Ten years after Brown, only 1.2% of Black children in the South were attending desegregated schools.17 Southern states effectively resisted, closing public schools and funding private segregated schools.18

“Brown and its progeny,” Rosenberg concludes, “stand for the proposition that courts are impotent to produce significant social reform.”19

Michael Klarman expands the argument—courts rarely stand up for rights when it’s unpopular.20 Together, Klarman and Rosenberg question the Court’s capacity to act against prevailing opinion.

Mark Tushnet makes the normative leap: the Supreme Court should not have the final say in what the Constitution means. Political institutions should settle their own boundaries.21

The cost of litigation-based reform, Tushnet says, is deeper than popular backlash. Public debate adopts legal vocabulary. Discussion of principle becomes circuitous, doctrinal. Resources—money, elite attention—funnel into litigation, at the expense of other channels of change.

The moment arrives.

A typical story, a partisan gerrymander. Wisconsin.

The case is brought on behalf of twelve voters, represented by the Campaign Legal Center and co-counsel, including one Nicholas Stephanopoulos, co-framer of the efficiency gap.

The district court declares the map unconstitutional. The efficiency gap is central to its analysis.

The Supreme Court agrees to review the case. Now everyone is paying attention.

Eric McGhee files an amicus brief for the efficiency gap. Gary King urges the Court to rely on partisan symmetry. Bernard Grofman files separately, urging a framework of symmetry, responsiveness, and causation.

They are few among many. King’s brief is joined by the soon-to-be Dean of Yale Law School. Eight political scientists, in two briefs, assess whether the bias could be explained by geography. It cannot. Historians argue gerrymandering has long been denounced in constitutional language.

John McCain and Sheldon Whitehouse file a brief. John Danforth, Bob Dole, John Kasich, Richard Lugar, and Arnold Schwarzenegger file a brief.

In all, thirty-six political scientists, ninety-five law professors, fifteen historians, and one-hundred-fifty-four politicians join or file briefs.

It’s thirty-five years after Grofman published his first article on symmetry, eleven years after Grofman and King predicted a non-trivial probability of rain, four years after Stephanopoulos and McGhee introduced the efficiency gap.

The support structure has assembled.

It’s a punt.

Chief Justice Roberts writes for a unanimous court. The plaintiffs have not shown the right type of injury. The Court does not consider whether gerrymandering is constitutional, whether any metric can prove it.

It has been the greatest honor and privilege to serve our nation in the federal judiciary for 43 years.22 With these words, Justice Kennedy announced his retirement from the Supreme Court of the United States, effective July 31, 2018. Twelve days later, President Trump nominated his replacement, Brett Kavanaugh.

If Grofman, King, McGhee, Stephanopoulos, and all the others were wondering Kavanaugh’s position on gerrymandering, they were not left guessing for long. Rucho was decided in Kavanaugh’s first Term.

The door was slammed shut.

“There are no legal standards discernible in the Constitution for making such judgments,” wrote the Chief Justice, “let alone limited and precise standards that are clear, manageable, and politically neutral.”

And that, as they say, was that.

Reviewing studies on the effects of Reynolds—one person, one vote—Gerald Rosenberg, in Hollow Hope, concludes “any [policy] effects that can be traced to reapportionment are small.”23

More recent scholarship suggests otherwise. When suburbs and cities received more representation, they secured significantly more government funding.24 More importantly still, Reynolds weakened the Southern Democratic party and helped create space for Southern Republicans.25 It was among the key ingredients in America’s partisan realignment.

Fixing gerrymandering through courts was not inevitably doomed to fail.

Had the Court chosen to intervene, it would probably have been effective. Gerrymandering is more similar to Reynolds than Brown; enforcement would not require mass cooperation or social change.

But aren’t metrics flawed? Yes; certainly, partisan symmetry and the efficiency gap both carry defects. Other metrics, especially those that rely on simulations, are probably better. Regardless, any metric would restrict the degrees of freedom partisan mapmakers can exercise.

Wouldn’t the Court have to draw an arbitrary line? Courts draw them for a living. “Beyond a reasonable doubt,” “probable cause,” “undue burden”—as arbitrary as they are ambiguous. “Where do you draw the line?” is so commonly asked by law professors that one of them termed an acronym: WDYDTL.26

How could Grofman and King have known that the Court would punt, that Kennedy would then be replaced by a less-receptive Justice?

Who could blame them for trying?

And yet I wonder if Grofman and King, if McGhee and Stephanopoulos, would find some truth in Tushnet’s frame. That in judicializing rights, the academic dialogue came to fixate on the perfect metric.

Ending gerrymandering doesn’t require a metric.

The Supreme Court could adopt a presumption that any partisan map—any map drawn by one party, without the consent of the other—is unconstitutional. Or the Court could forbid any “self-interested insiders” in participating in the districting process.27 Neither requires a metric.

Congress could address the problem in all sorts of ways. It could require maps to be drawn by nonpartisan commissions, as did a bill the House passed in 2019. It could require multi-member districts with proportional elections, or simply statewide proportional representation.28

There is something seductive about Kennedy’s challenge. How do we measure gerrymandering without hypotheticals? How do we flatten a many-dimensional object into something easy to understand and apply?

It’s a neat little puzzle. Trivial to understand, excruciating to solve. On the other side, a fairer electoral system.

A hard math problem that matters—or so they hoped.

There are at least 18 ways to measure gerrymandering, most of them developed since 2014. For each metric, there is a wealth of criticism—most of it quite technical/academic/in-the-weeds, and specific to the metric being criticized.1

There are a few non-academic articles explaining why measuring gerrymandering is hard. They’re worth reading—Drutman explains the many facets of “fairness”, and Yglesias and Whitcomb show how a fair-looking map can, by virtue of geography, favor one party.

But metric-makers are sophisticated folks—political scientists, economists, mathematicians. Why has the problem proven intractable?

My goal is to uncover the shared conditions that make measuring gerrymandering hard. I argue it’s hard to measure for three reasons. First, the thing being measured is discontinuous: single-member districts convert tiny vote-share differences around 50% into different outcomes. Second and third, the relevant comparison is counterfactual. To tease apart partisan gerrymandering, a metric needs to know what elections are plausible, and what maps were actually available.

Consider this my grand unified theory of why shit’s hard.

Conditions that make it hard

1. Discontinuity. A district lost with 49.9% of votes, as a matter of vote distribution, essentially the same as a district won with 50.1%. But it’s a different outcome.

2. Electoral baseline. Not all statewide vote outcomes are equally likely. Some states are reliably Democratic; others are reliably Republican. A good metric accounts for the actual likelihood of statewide outcomes.

3. Geographic & legal contingency. What maps are geographically and legally available? A metric that only takes into account votes and seats fails to consider these constraints. That’s lossy. In Joe Biden’s words: don’t compare me to the Almighty, compare me to the alternative.

Points two and three can be combined into something like “counterfactual baseline”—is the metric assuming an unrealistic distribution of votes (geographic contingency), implausible maps (legal contingency), and unlikely statewide outcomes (electoral baseline)? I choose to separate electoral baseline, and group together geographic & legal contingency, but you could of course make other choices.

To see geographic and legal contingency, imagine a hypothetical state, Fractalia, divided into thousands of precincts. In every precinct, Democrats receive exactly 49 votes and Republicans receive 51. If districts are contiguous unions of precincts, then every possible district is 49% Democratic. Democrats receive 49% statewide and win zero seats—not because of a gerrymander, but because no legally available map can do otherwise. If the same voters were arranged differently, it would very much look like a gerrymander. But in Fractalia, it’s the geography, stupid.

This may seem implausible; but consider a different hypothetical state—call it “America”—where one party is overwhelmingly urban, with small pockets of support elsewhere. The urban party—with or without gerrymandering—will produce landslides in the urban districts. For a party, landslides are bad. You’d rather some of those urban folks sprawl out and help you win a few more seats.

Bring out the metrics!

Now consider the 4 most cited ways to measure gerrymandering. This gets a little technical, but the tl;dr is—each of these four very important gerrymandering metrics is injured by these conditions.

1. Partisan Symmetry: Asks whether the map treats parties equally at the same vote share. If Democrats win 9 of 10 seats with 60% of the statewide vote, would Republicans also win 9 seats with 60%?

2. Efficiency Gap (EG): Compares each parties “wasted votes”: all votes cast for losing candidates, plus surplus winning votes (every vote over 50%+1). Were Democrats or Republicans more efficiently distributed?

3. Mean-Median Difference (MMD): Compare a party’s mean and median district vote share. If its mean is significantly higher than its median, the party’s vote is probably concentrated in high-margin districts. You’d typically rather win by slimmer margins.

4. Declination: Measures the difference in magnitude between losing and winning districts, i.e. asymmetry across the 50% won/lost line. A party that is losing narrowly and winning by a lot is competing on unfavorable terms.

EG and declination inherit the discontinuity. For EG, for instance, a district won by 50.1% is essentially the “perfect district” for that party—they waste no votes—while a district lost with 49.9% of votes is the worst-possible district. Therefore, a tiny shift in votes commits a massive difference in efficiency. The same is true for declination.

The other two metrics—symmetry and MMD—smooth over the discontinuity. A tiny shift in votes does not change either of these metrics. But that creates a new problem. Now the metric is removed from actual outcomes. If Democrats win 2 of 3 districts with 40%, 51%, and 62% of the vote, MMD detects no difference from Democrats winning 3 of 3 districts with 51%, 51%, and 51% of the vote.

Electoral baseline is particularly devastating for symmetry. Who cares whether Republicans would theoretically earn 9 of 10 seats with 60% of votes, if they can never hope to earn 60%? Symmetry does not limit its investigation to plausible outcomes.2

The other metrics are also affected. Declination becomes unstable in landslide states and undefined altogether if the minority party loses every district.3 Let’s say District A tends to be 7 points below than the state average for Democrats. EG will be significantly different if the state votes 55% for Democrats (they lose the district with 48%) vs if the state votes 60% for Democrats (they win with 53%).

All four metrics require only district vote shares and seat outcomes—none account for geographic & legal contingency.

It is quite plausible Democrats are winning landslides, and losing by smaller margins, because they have overwhelming support in cities, and small pockets of support in other areas. Indeed, this is the thesis of Jonathan Rodden’s excellent book.4

These conditions would organically create an “asymmetric” map, where Democrats win fewer seats with equal vote share. It also means Democrats likely “waste” more votes (EG), have higher mean than median district vote share (MMD), and win with higher margins than they lose (declination).

These are not the only problems with these metrics. But this is, I think, a fair synthesis of the background conditions that make this a fickle problem.

There is a solution

Or at least a better framework.5 Simulate a bunch of maps. Use some basic geographic unit—e.g. precincts—and require maps meet contiguity/compactness/VRA requirements. Applying actual election returns, ideally from several elections, record how each party fares under each simulated map. Perhaps also test counterfactual vote shares.

Compare the enacted map to the simulated ensemble. Test how the enacted map compares to the ensemble on raw seat count or any other metric (including, if you’d like, symmetry, EG, MMD, etc.). If the enacted plan is more favorable to Democrats than 99.5% of simulated valid plans, that is really strong evidence of a gerrymander. The old metrics, too, can be interpreted against an ensemble.

Complexity hides in the detail. It’s not easy to model the VRA, there are different ways to define “compactness”, and there are different possible sampling distributions.

Still, simulation largely solves the baseline problem. It asks whether the enacted map is extreme relative to a stated universe of lawful, geographically plausible alternatives.

Gerrymandering is hard to measure because maps are not just partisan artifacts. Partisan manipulation occurs in the shadow of geography, law, and a particular political moment. A good test has to see the alternatives.

FairVote, Protect Democracy, and other reform organizations make the intuitive case—seats should track votes. If a party wins 35 percent of the vote, it should receive 35 percent of the seats.

That argument made me curious about the current system. How disproportional are single-member House elections? Are they more or less proportional than they used to be? I decided to chart the relationship between votes and seats over time.

I made the chart below: the Democratic share of the national vote for House of Representatives vs the Democratic proportion of House seats. When the dotted line is above the solid line, Democrats secured a higher proportion of seats than their share of the national vote.

For instance, in 1892, Democrats secured 48% of the vote and 61% of House seats. Two years later, the Democratic share of the national vote decreased by 10 percentage points—to 38%—but their share of House seats halved, to 26%.

Excited, I sent the chart to a friend. House elections look more proportional in recent years than ever before, I said.

It’s a cool chart, he responded. But I don’t know how to discern proportionality from it.

Just look at the shading, Jack!

But the closer I looked, the more I had to agree:

1. The closer the vote share to 50%, the more material any seat surplus/deficit.

2. The “vote-seat gap” flattens two distinct concepts—bias and responsiveness—into one.

3. Measuring bias is hard, interpreting it is harder.

A seat surplus is more valuable today than it used to be.

The first point is easy to see. There may be a greater political difference 51 and 55 percent of seats than between 55 and 70 percent of seats.

More clearly, there is a discontinuity at 50%. A tiny seat surplus makes all the difference to a party that earns 49% of the vote.

From 1959 through 1994, Democrats always had at least 50 more House seats than Republicans. Today thin majorities are the norm. The world of landslides and durable majorities has given way to narrow margins and regular changes in control.2

That makes any seat surplus far more valuable.

Proportionality doesn’t tell us what we want to know.

Let’s say Democrats win 55% of votes and 65% of seats. That’s a 10pp seat surplus.

Is that fair? Maybe. If Republicans won 55% of the vote, would they have also won 65% of seats?

More generally: conditional on a vote share, would each party receive the same seat share?

Responsiveness is the percent change in seats corresponding to an additional percent of votes. Bias is an asymmetry between parties in the seat-vote relationship.

Consider three different ways of translating votes to seats:

In a proportional system, 40% of votes secures the same share of seats. In a winner-take-all system it secures no seats. In the majoritarian system modeled here, it secures 30%.

The three systems are significantly different, but none of them are biased. Republicans and Democrats are treated the same.

Here are the winner-take-all and majoritarian system, with a bias:

Now Republicans need more votes to achieve the same representation as Democrats.

Measuring bias is hard, interpreting it is harder.

Bias varies with vote share. Look again at the biased majoritarian system. The Republican penalty depends on the vote share. If Republicans win 100% of the vote, they win all of the seats; there is no bias. At other vote shares, there is a bias.

So where, along the curve, should we measure bias? Most commonly, political scientists measure bias when vote share = 0.5. They define “bias” as the difference between the seats Ds and Rs would expect to secure, if each with 50% of the vote. Others sum the difference in expected seats across several possible vote shares (e.g., if each party has 45%, 46%, etc., up to 55% of the vote). The more complete approach would be to sum the entire area between the curves, although that might put too much weight on unrealistic outcomes.

Additionally, elections do not magically produce vote-seat curves. An election delivers only one point along the curve. To recover the curve, political scientists model counterfactual elections. The simplest model, “uniform swing”, moves every district by the same amount (say, increase Democratic and decrease Republican vote share by 1% in each district), counts new seat shares, and traces the curve. Newer methods soften that assumption, allowing districts to vary around the swing.

Finally—and most critically—a good measure of bias does not uncover intention. Gerrymandering can create bias, but so can different turnout across districts and natural geographic sorting. Sorting is particularly salient today, as Democrats have become an overwhelmingly urban party.

Look again at the first chart. It tells us that seats are not perfectly proportional to votes. At times, seat share dramatically departs from vote share, although the largest gap is not necessarily most important. But the chart does not tell us whether there is a partisan bias, whether Democrats and Republicans would be treated equally if they achieved the same vote share.

This complicates my analysis of the proportionality of the House. More importantly, it complicates the argument for proportional representation.

There’s nothing wrong with saying that 35% of the vote should translate to 35% of seats. But the claim is broader than it first appears. It rejects partisan bias, which I suspect most people would agree with. It also rejects the winner’s bonus. In other words, it says the system should be less responsive—a 1% change in votes should yield a 1% change in representation.

That’s more controversial, and it’s a topic for another time.

• Why reapportionment made the North more liberal, but the South more conservative

• Why the Civil Rights Act would be harder to imple

  ment without reapportionment

• How reapportionment paved the way for party realignment

“I can stay here as long as you can!” declares the senator from Northern California. “And I can stay here as long as you can,” dutifully retorts the southerner.[1]

It’s Monday, April 18, 1921. We’re in the California State Senate in Sacramento. The room is deep red. The carpet, the heavy draperies, the upholstery on the chairs—all red, all borrowed, deliberately, from the British House of Lords. It’s dim.

Forty desks face the dais in tight curving rows. An enormous chandelier hangs overhead.

Carved into the cornice, above the rostrum, the Latin motto: SENATORIS EST CIVITATIS LIBERTATEM TUERI—it is the duty of a senator to guard the liberty of the Commonwealth.[2]

The California Constitution says representation “shall be apportioned according to population.” After each census, the legislature must redraw the districts.

Senator Frank Boggs introduced the first bill of this legislative session, to do exactly that. It’s labeled “Senate Bill 1,” or S.B. 1.

***

Senator Boggs is convinced his bill will pass—if only he can get it to a vote.

But Senator Walter McDonald of San Francisco has made it his personal mission to keep the bill from reaching a vote.

For months, he has campaigned against the reapportionment plan. “If we permit a reapportionment on the proposed basis,” he warned in January, “We may as well divide the State.”[3]

At 10:30 a.m., the Senate Journal records: “Bill read second time. Committee amendments pending.”[4] S.B. 1 is now on the floor, slated for a vote today.

McDonald is worried. “They’ve slipped,” he says of his fellow opponents. “We’ve got to fight for time.”[5]

At 11:00 a.m., before the Senate can vote on any amendment, McDonald rises. “Mr. President,” he announces, “I move to place Senate Bill 1 at the foot of the second reading file.”[6]

That is: beneath every other bill, unlikely to be reached before the end of the session. Defeat through delay.

The Senate Journal records, impassively: “Motion seconded.”[7]

McDonald declares he will use “every means” to defeat reapportionment, including demanding a separate roll call on eighty-four amendments.[8]

The lieutenant-governor suggests this would cause delay.

“I object to the presiding officer of this Senate explaining this bill to the advantage of its proponents,” an opponent of reapportionment protests. “As long as we beat this bill, I don’t care if we pass no other measure.”[9]

Eventually the Secretary calls the roll on McDonald’s motion. If it passes, S.B. 1is very likely dead.

“Allen” “Nay”

“Anderson” No response

“Arbuckle” “Aye”

It’s going to be close.

***

Since the 1910 census, California’s population has grown 44 percent. There are a million new Californians.[10]

But the growth has not been even. Los Angeles County has nearly doubled. Some northern districts have shrunk.

As a result, one northern Sierra senator represents about 43,000 people, while one Los Angeles senator represents about 117,000. One voter has three times the representation of the other.[11]

This is an instance of a national pattern.

Land has always promised self-reliance—the American virtue—and wealth, the American dream. In 1910, most Americans still lived on farms or in villages of fewer than 2,500 people.

But the country is changing. Over the last decade, six million Americans have left farms for cities.[12] By 1920, for the first time in American history, most Americans live in urban places.[13]

The city throws up a new landscape—skyscrapers and apartment houses, speakeasies and department stores. The livery stable becomes a gas station.[14] Great writers portray small-town life, once the nation’s moral center, as narrow, stifling, and provincial.[15]

There’s a backlash. To a turn-of-the-century politician, writes William Leuchtenburg, “no asset was greater than that of birth in a log cabin.”[16] The Ku Klux Klan has exploded in membership, strongest in places where old-stock Protestants feel themselves being overtaken by an America becoming Catholic, Jewish, immigrant, Black, wet, and urban.[17]

The old America clings to its institutional advantage. By 1921, Alabama, Kansas, Michigan, Oklahoma, and Tennessee have all failed to redraw districts. For the first time in history, Congress has failed to reapportion after the 1920 census.

It’s a battle between two Americas, described by Walter Lippmann as “the new urban civilization with its irresistible economic and scientific and mass power,” and “the older American village civilization making its last stand against what looks like an alien invasion.”[18]

On April 18, 1921, California is at its center.

***

Back in the Senate chamber, the Secretary is still calling the roll. It’s close, but the motion loses, 18 to 15, with 7 absent. Those seven absences give McDonald his opening.

Before the secretary can proclaim that McDonald’s motion has failed, McDonald rises again.

“I move for a call of the Senate.”

The Senate Journal records dryly: “Pending the announcement of the vote, Senator McDonald moved for a call of the Senate. Motion carried.”

It continues: “Time, eleven o’clock and twenty-five minutes a.m. The President directed the Sergeant-at-Arms to close the doors.”[19] The sergeant-at-arms is sent to find the absent senators.

A call is an old attendance rule.[20] But California’s rule has a wrinkle. Ordinary business cannot continue without the unanimous agreement of every present member.[21] Many legislatures allow business to move under a call if there is a quorum—California does not.

To pass, S.B. 1 will need twenty-one votes, an absolute majority of the forty-member Senate. Both sides believe they have the votes: “We’ve got enough votes to put it through,” says Boggs. “Go ahead and try . . . . You haven’t got the votes,” responds Senator Crowley of San Francisco.[22]

Neither side has twenty-one in the room, so both sides have reason to gamble.

The vote is suspended; the chamber is sealed.[23] Since McDonald can withhold his assent, he can stop the Senate from doing almost anything else. It’s a filibuster.

The newspapers capture the stalled chamber.

The Bee says Senators were “locked up.”[24] “McDonald tied up the Senate,” says The Examiner. “Nothing was done,” the paper continued—“and it was fully demonstrated that time, if made for slaves, was not made for State Senators of California.”[25]

Some clerical business does continue, but important matters are frozen.

Eventually, the call of the Senate will be lifted. There will be a deal, or absent members will show up. But even then, McDonald wields a potent threat. The reapportionment bill carried eighty-four amendments out of committee. McDonald will demand a separate roll call on each.

“I don’t want to block good legislation,” McDonald tells his colleagues, “But I propose to have a roll call on everything.”[26]

“It looks as if he could hold us here for at least 48 hours,” admits one pro-reapportionment senator. “We cannot prevent roll call after roll call.”[27]

The Chronicle reports, “McDonald had demanded a separate roll call on eighty-four amendments to the Boggs bill. Lieutenant-Governor Young . . . suggested this would cause serious delay.”[28]

That, lieutenant-governor, is the point.

***

Thomas Jefferson opens his Manual of Parliamentary Practice with Arthur Onslow, the longest-serving Speaker of the House of Commons. Onslow taught that parliamentary rules are “a check and control” on the majority, “a shelter and protection to the minority.” The majority has numbers, the minority has rules. Procedure is a check, Jefferson says, on “the wantonness of power.”[29]

Modern constitutional scholars make a similar point. Sujit Choudhry calls some legislative procedures “opposition rights”: procedural rules that let a minority scrutinize, delay, demand answers, and force questions into public view. Democracy, he argues, needs a “regime of opposition rights.”[30] But those rights, in his account, are different from supermajority rules. Opposition rights should not replace majority rule.[31]

***

At 1:45 in the afternoon, McDonald is seized by grace, or just hunger. The rules say no recess can be taken during a call. McDonald offers a “gentleman’s agreement” for a lunch recess.[32]

What’s a rule among gentlemen?

Almost all members sign.[33] One senator from Los Angeles, asked to sign, cracks: “I don’t see the names of any gentlemen signed to it.”[34]

***

How does McDonald get anywhere close to half of the Senate to subvert a constitutional duty?

Two ingredients: local interest and North-South politics.

Los Angeles has grown the most and needs more senate districts. For every new district, some other districts have to be merged.

San Francisco is set to lose a district, having already lost two districts following the last census. The entire San Francisco delegation—seven men, including McDonald—oppose redistricting.

Santa Clara County is set to lose one of its senators. One of the county’s senators concedes the arithmetic. “Santa Clara has population for one senator and a sixth,” he says. “The trouble is we cannot make one sixth into a whole.”[35] He understands the county doesn’t have enough people to justify two districts—and votes against reapportionment anyway.[36]

Sometimes the losers are created by chain reaction. District 17 needs more people. Boggs proposes adding Santa Barbara, which pulls it out of a district with Ventura. Ventura needs a new partner, and the only adjacent county large enough is Kern. Boggs glues the two together.

The senator from Ventura is furious. The pairing is “intolerable,” he writes, “since there is practically no community of interest between the two counties.” Kern is more populous, he continues, so Ventura “would be simply the tail of the dog.”[37]

Even within counties, the politics are deeply local. Chico, the largest city in Butte County, favors a plan joining Butte with a mountain county,[38] creating a district with stronger timber and mineral interest. Gridley, a smaller farming town in the same county, wants the district to stay agricultural.[39]

In the end, the Boggs plan would essentially keep District 6 unchanged, swapping one small Sacramento Valley farm county for another.[40] Senator Duncan—a farmer’s senator—should be happy.

Yet he opposes the bill. He is one of the senators who intends to demand a roll call on every one of 84 amendments.[41] It’s he who shouts, at another senator on the Senate floor: “I can stay here as long as you can!”[42]

What explains his opposition? “I don’t see what Los Angeles is all het up about,” he tells the Bee.[43]

This is the regional dimension of the conflict. The Sacramento Bee’s front-page session preview framed reapportionment as a coming North-South fight. Under the heading “Reapportionment Battle,” it warned that Los Angeles County would seek increased representation because of its population growth, while San Francisco, “with the support of the Northern Legislators,” would fight the Los Angeles contingent “to the last ditch.”[44]

Los Angeles has its own distinct interests. It is more dry than San Francisco. San Francisco is the most unionized city in the state; Los Angeles is famous as an open-shop city. But more than any policy issue, reapportionment is a power struggle between the old northern coalition and the young southern metropolis.

McDonald says as much: “If we permit a reapportionment on the proposed basis that would give Los Angeles county twenty-one Assemblymen and ten Senators we may as well divide the State.”[45]

And the divide runs both ways. San Diego’s district is entirely unchanged. But San Diego is in southern California, and its senator confronts the San Francisco delegation: “What’s the matter with you San Franciscans?... Why do you block everything in this way?”

Crowley fires back: “What’s the matter with you Southerners? . . . You want to grab everything; why don’t you divide the State?”

“We love you too much.”[46]

***

The Senators return, “with reapportionment in the same condition as before feeding time.”[47]

Boggs still has the votes. What he does not have is a vote. “I have enough votes to pass my bill if I can ever get it to a vote,” he says. “If I cannot get it to a vote, why I have done my best, that’s all.”[48]

Several Senators try to reach the governor. If the governor promises a special session, just for reapportionment, perhaps they can find a way out of this wretched chamber.[49]

It’s 5:15 now. We’ve been locked in for six hours. The Senate “killed time all day long,” reports The Chronicle, with Senators “amusing themselves in various ways while enduring their confinement.”

Over the day, three additional senators have arrived in the chamber. Four absentees remain.

McDonald moves to end the call of the Senate. Perhaps he was hoping the newly-arrived senators would help. The Senate Journal records: “The Secretary was directed to call the roll, on Senator McDonald’s motion, of the Senators who had not answered to their names.”

McDonald loses 16 to 20.[50] Two of the newly arrived Senators vote against his motion. Boggs appears a single vote away from having enough votes to pass S.B. 1.

Has McDonald miscalculated?

McDonald rises again. He moves for the bill to be re-referred to committee.

This would delay final vote by another day. There are only four days of the session left.

Boggs, the bill’s sponsor, says sending the bill back to committee would mean its defeat. Another senator offers a similar prognosis: “If the Senate sends this bill back to committee, it will be evading its constitutional duty.”[51]

The Secretary calls a vote. One by one, the Senators respond.

McDonald’s new delay motion is on the brink of passing. Several senators have changed their vote.

Senator Chamberlin, who favors reapportionment, interrupts the vote and moves to freeze the Senate.

It’s the same tactic McDonald used six hours earlier. But this time it’s made by the other side. The Senate Journal records: “Pending the announcement of the vote, Senator Chamberlin moved a call of the Senate.”

It continues: “Time, five o’clock and forty-five minutes p.m. The President directed the Sergeant-at-Arms to close the doors.”

The Senate is sealed again, but now the pro-reapportionment side is scrambling. Over the next ninety minutes, two more Senators arrive.

At 7:10 p.m., Senator Chamberlin releases the call. The Secretary calls the roll.

***

A few hours ago, 20 Senators voted against McDonald. During the call, one more pro-reapportionment senator entered the chamber. They have an absolute majority. Boggs appears correct: if they can get it to a vote, they will prevail.

But can they hold together?

Boggs is from the San Joaquin Valley. This sprawling region between the Sierra Nevada and the Coast Ranges is neither northern nor southern California. This geographic neutrality makes Boggs a suitable author of the bill.

The day after Boggs introduced S.B. 1—on January 6, 1921—the four senators from the San Joaquin Valley met at the Hotel Land in Sacramento, and pledged their support for the bill.[52] This may be what gives Boggs confidence in the bill’s passage—if the North locks horns with the South, Central Valley Senators break the tie.

Prohibition is the other major issue on the 1921 legislative agenda. Reapportionment and prohibition, reports the Appeal-Democrat, are the “grim spectres” of this legislative session.[53]

Senator Harris is one of the four San Joaquin senators who pledged their support for S.B. 1.

The same day Boggs introduced S.B. 1, Harris introduces S.B. 4. It’s a dry bill, instructing California officials to enforce federal prohibition law.

Harris’s bill has not yet had a second reading in the Senate.[54] Harris has every reason to fear that a prolonged filibuster over reapportionment will jeopardize it.

On the other side of both fights stands the same man. McDonald is the floor leader of the wet bloc, the man who will orchestrate the opposition to Harris’s dry bill.[55] Perhaps McDonald can kill two bills with one stone.

Harris switches his vote. He cannot afford more delay.

***

The motion passes. Twenty-two to sixteen.[56] S.B. 1 goes back to committee. Officially, only until tomorrow. In practice, Boggs believes, the bill has just been killed.

Five senators have switched. Harris is joined by three more drys: Dennett, also of the San Joaquin Valley, Nelson,[57] of the state’s far north, and Newton Allen, of Los Angeles.

Inman of Sacramento switches too. Inman is a wet, but he is fed up with the delay. “We are rapidly getting into a jam,” he says, “and literally hundreds of bills will die in committee.”[58]

But it is Allen who takes the blame, because he is from Los Angeles.

“All Blame Allen” reads an L.A. Times subhead.[59] The article continues, “L.A. city and county legislators joined in warm denunciation of Senator Newton M. Allen of L.A.”

Assemblyman Lyon says, “I lay the blame . . . at Allen’s door.” Senator Chamberlin goes further: Allen, he says, has “proved his disloyalty to the south.” [60]

The pro-reapportionment side manages one last gesture. The Senate votes that S.B. 1 must be returned from committee no later than noon the next day.[61]

But several senators already understand what has happened. Sending the bill back to committee, Boggs says, “meant its defeat.”

***

The Reapportionment Committee meets the next morning and tries to redraw the map one more time.[62] The committee tries to win back Ventura by leaving it alone and moving Kern elsewhere. But Kern has to go somewhere, and now the senator from neighboring Kings County objects. It’s a one-for-one trade; S.B. 1 is no stronger than before.

At noon, a meeting takes place at the Governor William Stephens’ office.

For months, the governor had maintained a “hands-off policy” on reapportionment.[63] He was from Los Angeles, and the southern delegation hoped for his support. But support would mean choosing Los Angeles over San Francisco, South over North. Stephens stayed out.[64]

McDonald’s delay tactics create a new problem. If the filibuster continues, the governor’s own economic and government-reorganization bills may die on the calendar.[65]

The governor offers an exit. He promises to call a special session for reapportionment.

Publicly, Stephens keeps his distance. Redistricting, he tells reporters, is “exclusively the Legislature’s concern.”[66] Privately, his representatives inform key senators of the promise.[67]

It’s an off ramp.

At 5:00 p.m., a pro-reapportionment senator introduces a resolution. It reads, in part:

WHEREAS, It is the constitutional duty of the Legislature to pass legislation reapportioning the State . . . in this session . . .

WHEREAS, This Senate is of the belief that any attempt to reapportion the State . . . in this session . . . will probably result in the defeat of very important legislation . . .

Therefore, be it Resolved . . . . That further consideration of all reapportionment matters be postponed . . . and that the Governor . . . is hereby requested to call a special session . . . [68]

The resolution is adopted, 26 to 6. Boggs himself advocates in favor. He, too, wants to get to other matters.[69]

The bill is referred back to committee—this time, permanently.

***

In 1689, John Locke described what would become the problem of redistricting. As populations shift, a town where “more inhabitants than a Shepherd is to be found” can send as many legislators “as a whole County numerous in People, and powerful in riches.” If representation is fixed by the Constitution, this creates an “inconvenience . . . incapable of remedy.”[70]

American constitutional law tried to answer Locke’s problem with the census: count the people every ten years, then reapportion and redraw. In theory, courts could complete the design: if legislatures fail, courts can compel compliance.

John Hart Ely would describe reapportionment as exactly the type of democratic failure that corrupts the political process and deepens over time. When beneficiaries control the instruments of repair, judicial intervention may be necessary to break institutional capture.

But there is a rising academic chorus questioning the wisdom of judicial supremacy. Niko Bowie—who taught me constitutional law—argues that the Supreme Court should not have the power to strike down federal law. He cites a long history of the Supreme Court invalidating laws that banned the spread of slavery, protected voting rights, prohibited discrimination, and outlawed child labor, among others.[71] Elected institutions, then, would bear the ultimate constitutional responsibility. Bowie is arguing principally about federal courts striking down federal law. But he might say that state courts’ long failure to stop malapportionment shows that courts, too, are unreliable arbiters of political process.

If courts are unreliable and legislatures self-interested, who remains?

***

The L.A. Times runs the headline “DODGING THE POINT.” “Surely the property owner in the roaring town is entitled to a representation equal to the man in the field or forest.”[72]

The S.F. Chronicle quotes the Constitution verbatim, concluding: “Sectional differences are no justification for violating the Constitution.”[73]

The off-ramp was the promise of a special session. It never takes place.

For months, a special session seemed “very probable,”[74] although the governor, reported the L.A. Times, is “taking his time on the matter.”[75]

The political problem is the same: the governor worries that a special session would force him to take a sectional position. One journalist describes the governor as “somewhat nonplussed.”[76]

Ten months after the lock-in, the governor publicly polls legislators, asking whether a reapportionment measure “could be enacted.”[77] Turning a decision into a consultation—the oldest trick in the book.

On March 4, 1922, the governor announces that his poll demonstrated reapportionment to be “hopeless.”[78] There will be no special session.

The L.A. Times has the last word: “The apportionment is mandatory, but what is a mandate between politicians?”[79]

[1] Kenneth C. Adams, Redistricting Bill Sent Back To Committee, S.F. Chron., Apr. 19, 1921, at 2, col. 2.

[2] California State Senate, 1935, Calisphere, http://calisphere.org/item/ade91fb4f1f7ecf01bff6c553b5a4a3c/ (last visited June 9, 2026); California Legislature, Senate Chamber, Calisphere, https://calisphere.org/item/385e7c8d9a3ab8b0c2bcbfb672f5e7a1/ (last visited June 9, 2026).

[3] Associated Press, Want Another Farm School, L.A. Times, Jan. 05, 1921, at 2.

[4] Journal of the Senate During the Forty-Fourth Session of the Legislature of the State of California 1542 (1921).

[5] Edward W. Hamilton, Apportionment Deadlocks Senate; All Work Stopped for Hours; North Lacks Votes to Kill Bill, S.F. Examiner, Apr. 19, 1921, at 4, col. 1.

[6] Adams, supra note 1, at 1.

[7] Senate Journal, supra note 4, at 1542.

[8] Reapportionment Causes Deadlock in Upper House, Sacramento Bee, Apr. 18, 1921, at 11.

[9] Reapportionment Causes Deadlock, supra note 8, at 11.

[10] U.S. Dep’t of Commerce, Bureau of the Census, Population: California (1921).

[11] Id.

[12] William E. Leuchtenburg, The Perils of Prosperity, 1914–32 225 (1958).

[13] U.S. Bureau of the Census, Fourteenth Census of the United States: 1920, Population, Volume I: Number and Distribution of Inhabitants 51 tbl.16 (1921).

[14] Leuchtenburg, supra note 12, at 2.

[15] Id.

[16] Id. at 229.

[17] Id. at 209-10.

[18] Pamela S. Karlan, Reapportionment, Nonapportionment, and Recovering Some Lost History of One Person, One Vote, 59 Wm & Mary L. Rev. 1921, 1935 (2018) (quoting Walter Lippmann, Men of Destiny 28 (1927)).

[19] Senate Journal, supra note 4, at 1542.

[20] Paul Mason, Mason’s Manual of Legislative Procedure §§ 190-191, at 155-156 (2010 ed.).

[21] Cal. Legislature, Rules of the Two Houses of the California Legislature 75 (1921).

[22] Hamilton, supra note 5, at 4, col. 1.

[23] Rules of the Two Houses, supra note 21, at 184.

[24] Reapportionment Causes Deadlock, supra note 8, at 1.

[25] Hamilton, supra note 5, at 4, col. 2.

[26] Id. col. 1.

[27] Id. col. 2.

[28] Adams, supra note 1, at 2.

[29] Jefferson’s Manual of Parliamentary Practice 1, in Constitution, Jefferson’s Manual, and Rules of the House of Representatives, H.R. Doc. No. 107-284, at 125-128 (2003).

[30] Sujit Choudhry, Opposition Powers in Parliamentary Democracies 40 (Apr. 22, 2026) (unpublished manuscript), https://ssrn.com/abstract=6628138.

[31] Id. at 14-15.

[32] Hamilton, supra note 5, at 4, col. 2.

[33] Reapportionment Causes Deadlock, supra note 8, at 1.

[34] Hamilton, supra note 5, at 4, col. 2.

[35] Edward H. Hamilton, Young Praised for Impartiality; 3 S.F. Solons on HIs Committee; Senate Favors Boggs’ Measure, S.F. Examiner, Jan. 6, 1921, at 2.

[36] Another example of legislative self-interest is District 4: one of the least populated districts in the state, awkwardly combining agricultural and timber counties, and thus an obvious target for elimination. Its senator, naturally, saw the matter differently.

[37] Arbuckle and Hume are Making Fight on Reapportionment, Ventura Weekly Post, Jan. 21, 1921, at 6.

[38] Butte County Fight Looms In Assembly, Sacramento Bee, Jan. 19, 1921, at 11.

[39] The Gridley Chamber of Commerce argues that, since “timber, mineral and hydro-electric interests, are controlled mostly by large corporations with headquarters in the larger cities,” they are sufficiently represented, and District 6 should be made up of Valley farm districts. Sutter and Butte Best District, Gridley Herald, Jan. 19, 1921, at 1.

[40] Yolo would be swapped for neighboring Glenn. They vote similarly—in 1920, 62 percent of Yolo and 64 percent of Glenn voted for President Harding. Cal. Sec’y of State, Statement of the Vote at General Election Held On November 2, 1920 (1920) Both Chico, the city, and Gridley, the farm town, were open to the addition of Glenn. Sutter and Butte Best District, supra note 39; Butte County Fight Looms, supra note 38.

[41] Bill on State Districts May Fail In Senate, Sacramento Union, Apr. 19, 1921, at 2.

[42] Adams, supra note 1, at 2.

[43] Duncan Sees No Reason for Excitement In the South, Sacramento Bee, Sept. 17, 1921, at 30.

[44] Forty-Fourth Legislature to Open Monday, Sacramento Bee, Jan. 1, 1921, at 1.

[45] Associated Press, supra note 3, at 2.

[46] Hamilton, supra note 5, at 4, col. 2.

[47] Id.

[48] Id.

[49] Reapportionment Causes Deadlock, supra note 8, at 11; Hamilton, supra note 5, at 4.

[50] Senate Journal, supra note 4, at 1572.

[51] Boggs’ Measure Sent Back To Committee, Sacramento Bee, Apr. 19, 1921, at 11.

[52] San Joaquin Valley Representatives Meet, Sacramento Bee, Jan. 7, 1921, p. 11.

[53] New Assembly Lineup Also Scheduled By Legislature, Marysville Appeal, Jan. 6, 1921, p. 1.

[54] Senate Journal, supra note 4, at 1619.

[55] Franklin Hichborn, Story of the Session of the California Legislature of 1921 284 (1922).

[56] Senate Journal, supra note 4, at 1581.

[57] Hichborn, supra note 55, at 283-284 nn.220 & 222.

[58] Reapportionment Bill Attacked, Bakersfield Morning Echo, Apr. 19, 1921, at 4.

[59] Kyle D. Palmer, Boggs Measure is Dead Issue, L.A. Times, Apr. 20, 1921, at 2.

[60] Id.

[61] Senate Journal, supra note 4, at 1581.

[62] Boggs Bill, Amended, Reported out Again By Senate Committee, Sacramento Bee, Apr. 19, 1921, at 6.

[63] Kyle D. Palmer, Extra Session Seems Likely, L.A. Times, Apr. 18, 1921, at 2.

[64] Id.

[65] Chronicle Bureau, Redistricting Bill Is Chloroformed; Senate to Request Special Session, S.F. Chronicle, Apr. 20, 1921, at. 3.

[66] Id. at 1.

[67] Special Session of Legislature is Expected, Sacramento Bee, Apr. 20, 1921, at 1. But see Palmer, supra note 59 (”Yes,” replied Inman, “I did talk with Mr. Madsen, but I have absolutely no agreement or understanding of agreement with the Governor’s office.”).

[68] Senate Journal, supra note 4, at 1621.

[69] Special Session, supra note 67.

[70] John Locke, Second Treatise of Government ch. XIII, § 157 (1690).

[71] Nikolas Bowie, The Contemporary Debate over Supreme Court Reform 4 (June 30, 2021), https://www.whitehouse.gov/wp-content/uploads/2021/06/Bowie-SCOTUS-Testimony-1.pdf.

[72] Dodging the Point, L.A. Times, Apr. 20, 1921, at 22.

[73] Redistricting the State, S.F. Chronicle, Apr. 20, 1921, at 22.

[74] Special Session May Be Called Before June 1st, Sacramento Bee, Jan. 4, 1922, at 1.

[75] News of the Capital, L.A. Times, Jan. 22, 1922, at 4.

[76] Id. at 84.

[77] Governor Asks Views Of Legislators on Re-Apportionment, Sacramento Union, Feb. 27, 1922, at 1.

[78] Governor Decides There Will Be No Special Session, Sacramento Bee, March 4, 1922, at 1.

[79] The Artful Dodger, L.A. Times, March 7, 1922, at 22.