North Carolina’s disputed race for governor: historical context

Federal courts have a power to protect voting rights that they lacked until recently.

Rick Hasen makes the correct and important observation that, if North Carolina’s General Assembly were to overturn the state’s gubernatorial election based not on the evidence of the actual valid votes, but solely because of a partisan desire to keep control of the governorship, then the federal judiciary would have the power to invalidate that “brazen power grab” as a violation of the Fourteenth Amendment.

In this post, I wish only to supplement Rick’s point with a historical perspective drawn from my new book, Ballot Battles: The History of Disputed Elections in the United States.

At the moment, no one knows for sure that the still-unsettled governor’s race in North Carolina will end up in the scenario that Rick envisions: a federal-court order, based on the precedent of Bush v. Gore or Roe v. Alabama, that nullifies the state legislature’s attempt to overturn the administratively certified result of the vote count. But if it does end up that way, it would underscore the 180-degree reversal of jurisprudence that has occurred over the course of the twentieth century concerning the power of the federal courts in this kind of case.

In 1900, the U.S. Supreme Court had a case involving exactly this same situation. In Kentucky’s gubernatorial election, the Republican candidate had been certified the winner, but the state’s legislature—controlled by the Democrats—overturned that outcome. Awarding the election to the Democratic candidate, the legislature’s move was a “brazen power grab” of the kind that Rick envisions occurring in North Carolina this year. Under the Kentucky constitution, the state’s judiciary was powerless to prevent this transparently partisan theft of the election. The Republicans, therefore, sought relief from the U.S. Supreme Court, arguing that the legislature’s conduct was the functional equivalent of stuffing the ballot box with fraudulent votes and thus a violation of fundamental rights protected by the Fourteenth Amendment of the federal Constitution.

The U.S. Supreme Court, however, rejected the Republicans’ claim, ruling it was barred by the so-called “political question” doctrine, meaning that the federal courts lacked jurisdiction to review the conduct of state institutions, including a state’s legislature, over the counting of votes in a state election. The Court’s decision, denominated Taylor v. Beckham, was 8-1. Only Justice John Marshall Harlan dissented.

“The overturning of the public will, as expressed at the ballot box, without evidence or against evidence,” Justice Harlan exclaimed, “is a crime against free government” and thus a violation of due process. He continued: “I cannot believe that the judiciary is helpless in the presence of such crime. The person elected, as well as the people who elect him, have rights that the [federal] courts may protect.” But Justice Harlan did not prevail, and the Kentucky legislature got away with its theft of the state’s gubernatorial election.

The jurisdictional barrier announced in Taylor v. Beckham prevailed in the U.S. Supreme Court throughout the twentieth century. This jurisdictional barrier, for example, is the basis on which Justice Hugo Black in 1948 ordered the federal court in Texas to shut down its investigation into whether Lyndon Johnson was the beneficiary of ballot-box stuffing in the crucial runoff election that was part of his race for the U.S. Senate that year. The same jurisdictional barrier is the reason why Richard Nixon in 1960 could not go to federal court to claim that John Kennedy’s victory in that year’s presidential election rested on election fraud perpetrated in Illinois and, again, Texas.

If the precedent of Taylor v. Beckham were still good law, it would block any federal court review of North Carolina’s proceedings concerning the counting of votes in the state’s gubernatorial election this year. This jurisdictional barrier would be absolute—no matter how egregious the partisan theft of the election by the state legislature might be. Yes, a “brazen power grab,” in Rick’s words, but indistinguishable in this respect from what the Kentucky legislature did in the gubernatorial election at issue in Taylor v. Beckham itself.   The case, as lawyers like to say, is “on all fours”—meaning that there is no plausible argument that the precedent, if still valid, is inapplicable.

But Taylor v. Beckham is no longer good law in light of Bush v. Gore. Although the latter case did not explicitly overrule the former, the two are logically irreconcilable. If the jurisdictional barrier announced in Taylor v. Beckham still prevailed, then the U.S. Supreme Court in Bush v. Gore could not have intervened on Fourteenth Amendment grounds to stop the recount ordered by the Florida Supreme Court to determine whether there were “hanging chads” (or dimpled ones) still to be counted in the 2000 presidential election.

No matter how much the majority of the U.S. Supreme Court might have thought that the Florida Supreme Court was attempting to steal the election for Gore, that theft would have been indistinguishable in principle from the one perpetrated by the Kentucky legislature in the gubernatorial election at issue in Taylor v. Beckham. The reasoning of Taylor v. Beckham was that a state’s vote-counting proceedings were off-limits to the federal judiciary, including the U.S. Supreme Court itself, because of the “political question” doctrine, and that reasoning—if still valid—would have been fully applicable to the 2000 presidential election, just as it was to the 1960 presidential election, or to the 1948 U.S. Senate race. Thus, the fact that the U.S. Supreme Court did intervene in Bush v. Gore necessarily demonstrates that the jurisdictional barrier articulated in Taylor v. Beckham is no longer good law.

How to explain this 180-degree transformation? The answer lies in Baker v. Carr and the “reapportionment revolution” that occurred in the Warren Court during the 1960s. Baker v. Carr was a radical reconceptualization of the “political question” doctrine, drastically curtailing its domain, and specifically rendering it inapplicable to Fourteenth Amendment claims concerning the malapportionment of state legislatures. Baker v. Carr quickly led to Reynolds v. Sims and its “one person, one vote” doctrine, which in turn the Warren Court invoked to invalidate Virginia’s poll tax.

Although none of these Warren Court precedents involved a state’s procedures for the counting of ballots, the new “one person, one vote” doctrine easily could apply to that particular aspect of election law as much as it does to redistricting or to prerequisites for the right to cast a ballot, like payment of poll tax. Indeed, the Warren Court itself signaled its potential applicability in a 1966 case involving a special runoff procedure that Georgia used in its gubernatorial elections, Fortson v. Morris.   The Court there divided 5-4 over whether Georgia’s runoff mechanism violated one-person-one-vote, with the majority concluding that it did not; but all nine justices signaled that the one-person-one-vote doctrine in principle was just as applicable to the back end of the voting process, where recounts and runoffs occur, as to the front end of the process, over matters like redistricting and the prerequisites for casting a ballot.

In the decades between the sixties and Bush v. Gore, the lower federal courts heeded the Court’s signal and began to apply the new one-person-one-vote jurisprudence to vote-counting disputes. As Rick notes, the most significant of these lower-court cases is Roe v. Alabama, which involved the Eleventh Circuit’s intervention in Alabama’s 1994 election for the state’s Chief Justice. The federal district court viewed manipulation of the state’s vote-counting rules as “the functional equivalent of altering ballots or stuffing the ballot box” and thus a violation of due process. The court’s reasoning was essentially identical to Justice Harlan’s dissent in Taylor v. Beckham—and thus to the position that did not prevail in that case. Indeed, if Taylor v. Beckham were good law, the federal judiciary had no role to play in the counting of ballots in Alabama’s election for its chief justice. But the Eleventh Circuit affirmed, not reversed, the federal district court’s ruling on the strength of the “one person, one vote” jurisprudence, thereby indicating just how thoroughly that jurisprudence had repudiated the jurisdictional barrier that had previously existed pursuant to Taylor v. Beckham. Moreover, unlike Justice Hugo Black in 1948, Justice Kennedy refused to order the federal court to withdraw from interfering in the Alabama vote-counting process. In this way, Justice Kennedy—and the U.S. Supreme Court as a whole—indicated that the precedent of Taylor v. Beckham no longer prevailed, and thus would not be a jurisdiction barrier six years later in Bush v. Gore.

It is an understatement to say that Bush v. Gore was a controversial ruling. But it is important to evaluate Bush v. Gore in historical context. Taylor v. Beckham was not a wise decision. Justice Harlan’s lone dissent had the much better argument in the case—in the same way as did his lone dissent in Plessy v. Ferguson (the infamous “separate but equal” decision). The episodes analyzed in Ballot Battles reveal that America’s ugliest experiences with vote-counting disputes have been in those instances where a state legislature brazenly steals a gubernatorial election in order to keep its own party’s candidate in power. Indeed, in the Kentucky gubernatorial election at issue in Taylor v. Beckham, one of the two candidates for governor was assassinated while the legislature was in the midst of its proceedings to overturn the certified election results. Thus, insofar as Bush v. Gore repudiated Taylor v. Beckham, thereby vindicating Justice Harlan’s dissent and confirming the power of the federal judiciary to thwart a state legislature’s transparently partisan theft of a gubernatorial election, that aspect of Bush v. Gore is actually a salutary development in American election law.

As for this year’s gubernatorial election in North Carolina, let’s hope that the current controversy never gets near the point that it begins to look like a repeat of Kentucky’s gubernatorial election in Taylor v. Beckham. Let’s hope, in other words, that the state legislature refrains from perpetrating anything looking like a partisan theft, solely to keep its preferred candidate in power. But if the state legislature were to go down this despicable road, then we must place our confidence in the federal courts to recognize that today, unlike in 1900, the words of Justice Harlan state the governing principle of law: “The overturning of the public will, as expressed at the ballot box, without evidence or against evidence,” as he put it, “is a crime against free government” and “[t]he person elected, as well as the people who elect him, have rights” protected by due process and enforceable by federal courts.