Election subversion is, perhaps unsurprisingly, one of the hottest topics in election law. It’s a term that can apply to many things, but I wanted to focus on one risk: some election administrator or board of elections refusing to certify the results of an election (or, worse, signing some alternative results apart from the results that come to the administrator or board through the ordinary canvassing and recount process).

It’s a topic that crops up time and again in the context of Electoral Count Act reform, as if it’s a novel problem. But this problem actually happens with some regularity (although it is by no means a common occurrence). The solution resides in a simple, longstanding mechanism that state courts routinely enforce: a writ of mandamus.

Mandamus is an instruction from a court to an executive official or board (for purposes of this blog post, at least) to perform a ministerial duty. It is famously the writ at issue in Marbury v. Madison. And it is used to direct recalcitrant officials—including election officials—to perform their ministerial duties.

In 2004, the Michigan State Board of Canvassers deadlocked 2-2 on whether to certify an initiative petition proposing an amendment to the state constitution defining the legal recognition of “marriage.” The petitioners had secured enough signatures, but two members of the board believed that the proposed amendment was “unlawful and unconstitutional.” After they deadlocked, a complaint seeking mandamus was filed in state court. A unanimous per curiam opinion of the court of appeals granted mandamus. The board then approved the initiative. The conclusion was, essentially, that the board had no authority to review the constitutionality of a proposed initiative. Its role was simply ministerial, and the court could instruct the board to certify the amendment.

It happened again in Michigan in 2021–just last year!–after the Board of Canvassers deadlocked 2-2 on whether the signatures in an initiative petition were adequate. The Michigan Supreme Court granted mandamus, found the board had a “clear legal duty to certify the petition,” and ordered the board to certify.

Or consider an Ohio case in 1984, where the Ohio Supreme Court concluded that one candidate was not eligible for the office. But the Secretary of State later instructed the county board of elections to count the votes for both candidates. Back at the Ohio Supreme Court, a majority granted mandamus relief to order the county board of elections to count votes only for one candidate and to declare that candidate the winner.

States have extensive rules for canvassing, recounting, auditing, and contesting elections. Sometimes mandamus cannot be brought against an executive official precisely because that process is playing out. Consider Franken v. Pawlenty, where Al Franken prematurely sought a signed certificate of election from Governor Tim Pawlenty after the contested Minnesota Senate election of 2008. The court wouldn’t issue such an order, because the process in Minnesota was still playing out under state law. Governor Pawlenty ultimately issued the certificate of election after the contest at the Minnesota Supreme Court concluded.

By the end of the canvass, recount, audit, and contest, however, the results are in. There is a fixed set of results for a winning candidate and a losing candidate. The decision to “certify” an election is a ministerial act. And if an election official opted to refuse to certify, or certify some other result, the remedy is to file in state court (not federal court) for mandamus.

I’ve been a little surprised at the lack of discussion or analysis about how mandamus fits within this context. It’s a tried and tested remedy that has long existed in state courts. Admittedly, it cannot prevent an election official from subverting the process by refusing to certify the results, sowing public confusing and undermining public trust. But as discussions about the need to address election subversion grow, it’s worth considering what federal courts or Congress might do (1) that states cannot do (or are not doing) or (2) that would be better than the mechanisms already at state courts’ disposal.

Of course, the next question is, “What if an official defies the mandamus?” Some states might have mechanisms to instruct a different official to perform the task, escalating, for instance, a county board’s defiance to a state board or the Secretary of State. And the prospect of jailing a state executive official for contempt of a judicial order is not exactly where we want to go in the extreme cases.

But it’s simply the point that judicial process can only go so far. At some point, the law runs out, and we’re dealing with, say, defiance of a judicial order that might lead to riots or armed hostilities… in short, there’s never going to be a system insulated from the possibility that some actors act lawlessly. Mandamus helps constrain the first acts of defiance, the basis for this post. It’s a routine mechanism available in state courts. But continuing to circle down toward ever-wilder hypotheticals is, in my judgment, an impossible thing to legislate.

That said, there’s a separate intriguing and lingering question out there. The Electoral Count Act anticipates that the executive of the state will send a certificate identifying the ascertainment of the electors appointed in that state to the Archivist of the United States. Given that this is a provision of federal law, what role could Congress have in ensuring that the state governors follow this provision of law dutifully? Some potential avenues (which may introduce their own challenges) appear in the draft Electoral Count Modernization Act, which I discuss here. But that’s a topic of another future post.