Mark Duncan / AP

In 1960, a town of 38 residents in Vermont elected the same number of representatives—one—as Burlington, population 33,000. In Georgia, house districts contained between 1,876 and 185,422 constituents. In California, more than 6 million residents of Los Angeles County elected just one state senator, as did 14,294 inhabitants of three counties on the eastern slope of the Sierra. Legislative malapportionment produced staggering inequality in virtually every state in the union.

It was to address this situation that the Supreme Court established “one person, one vote” as a bedrock of American democracy. Now, for the first time since that era, the “reapportionment revolution” is under threat. This fall, in Evenwel v. Abbott, the Court will weigh whether or not “one person, one vote” allows states to base apportionments on all persons living within a given district, or whether the phrase really means “one voter, one vote” and requires states to count only voters for the purposes of representation. A ruling in favor of the challengers, who claim the weight of their votes has been diluted because Texas counts all persons, threatens to undermine one of the great achievements of 20th-century American democracy.

Until now, states have had the leeway to define their own method and have overwhelmingly opted to consider all persons; only a handful of states base apportionments on the number of voters. A ruling in favor of the challengers threatens to disqualify, at least for the purposes of representation, persons under the age of 18, documented immigrants who are not yet citizens, and former felons who have lost the right to vote. Latino immigrants and African-American males, who disproportionately live in urban areas, would be most affected.

Chief Justice Earl Warren considered the reapportionment rulings, which are now at risk of being overturned, the most important of his tenure—more important than Brown v. Board of Education, more important than Gideon v. Wainwright. They were intended to correct what had become, in effect, a system of minority rule.

Prior to the 1960s, American democracy was a deliberately misshapen enterprise. Because most states based representation in at least one chamber of their legislatures on factors other than population, district boundaries could readily be drawn to give some citizens a far greater voice than others. A lawmaker from a rural area of a state, for example, might have represented 10,000 voters, while a colleague from a more urban region might have had two times, or five times, or even ten times as many constituents. And since state lawmakers typically control districting for the U.S. House of Representatives, malapportionment guaranteed that residents of more populous districts lacked adequate representation at the federal level as well.

Most states mandated the reapportionment of at least one branch of the legislature at regular intervals, and more than half of the states required the regular reapportionment of both branches. But as urban and suburban populations swelled dramatically throughout the 20th century, lawmakers allowed district boundaries to remain fixed for decades.

Finally, the Supreme Court, led by Chief Justice Earl Warren, intervened. In a series of decisions handed down between 1962 and 1964, the Court concluded that each state’s districts for the U.S. House of Representatives must contain, as nearly as possible, the same number of people, and that both chambers in a state’s bicameral legislature must be apportioned according to the principle of “one person, one vote.” By the end of the decade, the Court had determined that apportionments for all elective municipal and county offices—including city councils, boards of supervisors, and school boards, among others—must adhere to the same standard.

In requiring that all apportionments be based on population, the Warren Court did not specify whether “persons” meant people or voters and thus the issue now before the Court. At the time—a year before the immigration act of 1965 opened the nation’s borders to millions of non-citizen residents—the distinction did not seem crucial to the justices.

A look at the practice in the states provides some insight as to why. In the early 1960s, only Massachusetts, Rhode Island, and Tennessee based apportionments in both legislative branches on the number of legal or qualified voters, and Massachusetts consistently ranked among the most equitably apportioned states in the nation. Texas counted qualified electors in reapportioning its state senate, while Arizona counted votes cast in the previous gubernatorial election in apportioning its lower chamber. (Indiana relied on population, but counted only males over the age of 21. For the purposes of apportioning legislative seats, women in Indiana literally did not count.) The overwhelming number of states then, as now, counted persons rather than voters.

The origin of the phrase “one person, one vote” also sheds light on the intentions of the Warren Court. It first appeared in a Supreme Court opinion in 1963, a year prior to the reapportionment decisions. Drafting the majority opinion in a challenge to Georgia’s notorious county-unit system, William O. Douglas made a seemingly modest, but crucial edit. Striking the last five words of his initial draft, Douglas wrote the words that transformed the parameters of subsequent reapportionment jurisprudence: “The conception of political equality from the Declaration of Independence, to Lincoln’s Gettysburg Address, to the 15th, 17th, and 19th Amendments can mean only one thing—one person, one vote.” Just before he circulated his draft to his colleagues, Douglas dismissed the suggestion of a law clerk who had recommended replacing “person” with “voter.”

Until the Supreme Court sits for oral arguments and issues an opinion in the Texas case now on its calendar, it is difficult to discern the justices’ motives for weighing the meaning of “one person, one vote.” Even if the Court decides that voters are a more appropriate measure than persons, it is not clear which voters would count. Eligible voters? Registered voters? Or actual voters? Given the appallingly low voter turnout in the United States, such a ruling could lead us back to a time when a relatively small number of American citizens enjoyed disproportionate political power. Given the Supreme Court’s recent assault on the Voting Rights Act and the commitment of the majority to neutering its great accomplishments, it is troubling to contemplate the Court’s motives. A decision in favor of the challengers indeed has the potential to threaten many of the gains that resulted from the reapportionment revolution of the 1960s.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.