In Part One of this series, we began discussing a lawsuit filed by Oklahoma attorney Stephen Jones in the Oklahoma Supreme Court seeking to prevent the state from holding a special election this year to replace U.S. Senator James (Jim) Inhofe, who has announced that he will be retiring from the Senate when the 117th Congress ends on January 3, 2023. We examined a few big questions raised by the episode, and in the space below we take up a few more.
Can Inhofe’s Promise to Resign be Enforced?
As we noted in Part One, Oklahoma state law provides that “if a vacancy or irrevocable resignation occurs in the office of a member of the United States Senate from Oklahoma” the state must hold an election on a prescribed timeline to fill the empty or to-be-empty seat. State law, though no model of clarity, defines an “irrevocable resignation” as “a written letter of resignation to the Secretary of State . . . that serves notice of the Senate member’s resignation on a date certain.” In his letter, dated February 28, Inhofe gave an “irrevocable pledge” to step down next January 3 even though his elected term has years left on it.
But what if Inhofe tries to renege on his promise? One of Jones’s arguments in the Oklahoma Supreme Court for why no special election can be scheduled or held until after Inhofe is actually gone from the Senate is that if Inhofe fails to follow through on his promise to leave office, there is nothing anyone can do about that, and any replacement election held in the meantime thus would have been for naught. (This part of Jones’s brief brought to mind two famous episodes of planned retirement—Justice Thurgood Marshall’s 1991 retirement letter that said he would “retire as an Associate Justice of the Supreme Court of the United States when [his] successor [wa]s qualified,” causing some people to wonder, wrongly in our minds, whether he was trying to retain an option to stay on the Court if he didn’t approve of the person the President nominated and the Senate confirmed to fill his seat; and NFL Quarterback Tom Brady’s suggestion last week that he might not retire after all.)
Jones asserts that under Oklahoma state law, Inhofe can change his mind without any punishment by the state. We are not experts on Oklahoma law, but it is not clear to us that under the federal Constitution, a state could not, if it wanted, impose consequences upon its own sitting senator for breaking an irrevocable pledge. For example, if a senator sent the state’s Secretary of State a resignation letter effective immediately, even during a Senate recess, and then tried to change his mind a few days later, we doubt anything in the federal Constitution would confer a right to remain in office or displace authority of the state to select and seat a replacement. Even putting aside the question of whether Inhofe could be forced—by a state court injunction or other means—to live up to his promise and vacate his seat, Oklahoma might still be able to sue Inhofe (if he reneged) for damages, insofar as the state would have relied on Inhofe’s irrevocable pledge in spending millions of dollars on a replacement election process. Likewise, individuals who launched campaigns and ran to fill the seat Inhofe had promised to leave might also have promissory-estoppel-type claims that could be vindicated under state law. It isn’t obvious that the federal Constitution would prohibit such actions any more than it immunizes senators from the reach of other ordinary state laws. In this regard, we do note there have been recent legislative proposals in Oklahoma to recoup election administration costs from state (as opposed to federal) officeholders who choose to leave office before the ends of their terms.
It is of course true that state control over federal officeholders is always tricky. But some might try to draw a rough analogy to the situation at issue in the Supreme Court’s 2020 decisions in Chiafalo v. Washington and Colorado Department of State v. Baca. In those cases, the Court unanimously upheld the power of a state to punish (and even replace) members of the state’s contingent in the so-called Electoral College who failed to cast their votes for the candidate who won the state’s popular-election contest for President. One of us, Amar, is on record as believing (and still believes) these two cases were terribly reasoned and wrongly decided. But these cases do suggest some leeway, at least in the view of the Supreme Court, for states to enforce certain kinds of discrete promises made by individuals serving in federal roles. One shouldn’t push the cases too far because (as Amar noted in his critique of them) members of the U.S. Senate have never (even in the days of legislative election) been subject to binding “instruction” by the state legislature or the state’s electorate in their discharge of senatorial duties. Query, though, whether a promise to leave office altogether is different than promises to vote certain ways on certain bills—especially if in reliance upon a retirement promise the voters of the state choose a replacement senator.
It is quite possible that the U.S. Senate itself would be the one to decide whether Inhofe must keep his promise. Jones rightly points out that expelling Inhofe from the Senate requires a 2/3 majority of the Senate. But expulsion is not the only senatorial tool. A simple majority of the Senate might invoke its separate power under Article I, section 5 of the Constitution to “judge” election returns and declare the person who prevailed in a special election to have rightly won the Oklahoma seat, so that there would be no seat for Inhofe to continue occupying. The Senate might also exert other pressures to persuade a colleague to follow through on a retirement promise: crummy committee assignments, refusals to co-sponsor bills, and formal censure are among the options.
Does Oklahoma Have an Appointment Workaround of the Inhofe Problem?
Let’s suppose that the Oklahoma Supreme Court, because the court believes Inhofe’s promise is unenforceable or because of the problem of “unripe” replacement elections we discussed in Part One (a problem perhaps not presented on the facts of the Inhofe episode but raised by the text of current Oklahoma law), rules in Jones’s favor, holding no special election can take place until Inhofe actually leaves office. Is there anything else Inhofe (and the state) could do to vindicate his (and its) wishes? Suppose Inhofe were to retire immediately, allowing the governor to trigger the special election process. Right after that (in order to avoid having Oklahoma represented by only one senator for several months), the governor could then appoint Inhofe to fill the vacancy Inhofe himself created until the special election is held and its results are certified. (This wouldn’t quite be identical to what Inhofe seems to prefer, since presumably he would be replaced in November of 2022 after his successor is certified, rather than—as he appears to desire—in January 2023 at the start of the next Congress.)
The problem with this route is that the current Oklahoma Senate-vacancy-filling statute authorizing (pursuant to Section Two of the Seventeenth Amendment) the governor to make temporary appointments also imposes unconstitutional conditions on those very appointments. First, the statute requires that the governor pick only from people of the same political party as the incumbent whose departure created the vacancy (a provision presumably aimed at respecting the partisan wishes of the voters at the last regular election until voters can be heard at another election). But as one of us, Amar, explained in a law review article over a decade ago, such requirements of party continuity, while seemingly benign, run afoul of the Seventeenth Amendment’s clear rejection of the power of a state legislature to constrain gubernatorial choice. The Seventeenth Amendment evinces a strong and general distrust of state legislatures’ motives and ability to represent statewide voters, and therefore lays down a bright-line allocation to governors (and corresponding exclusion of legislatures) of substantive decision-making power regarding temporary Senate appointments. In this vein, note that a state legislature certainly couldn’t constrain the state electorate at a special vacancy-filling election to pick only a person from the party of the departed senator. State law could not, for example, provide that only persons from the same party as the departed senator are permitted to have their names placed on the special election ballot. The reason for this seems intuitive: any legislative interest in maintaining or predicting the wishes of the electorate vanishes when the electorate has a chance to express its wishes itself.
Even if state law constrains only gubernatorial interim appointments, and not the state electorate at special elections, there are difficult (indeed seemingly insurmountable) constitutional hurdles standing in the way of constraints imposed by the state legislative. Among the most important is the fact that limiting the governor to appoint individuals of a single party amounts to adding “qualifications” to the office of U.S. Senator. U.S. Term Limits, Inc. v. Thornton makes clear that neither Congress nor states, including state legislatures, are entrusted to add qualifications for the U.S. Senate beyond those (of age, citizenship, and residency) already provided for in Article I.
But is maintaining the party identity of a state’s U.S. Senate contingent between elections really prescribing a “qualification”? Thornton suggests it is. The Court distinguished sharply between “procedural” time, place, and manner regulations over which states have power, on the one hand, and “substantive” qualifications for office over which they do not, on the other. Surely, prescribing party affiliation is not a “time, place, or manner” regulation. A state legislature, as noted above, certainly could not impose such a limit on the state electorate; nor could Congress impose such a limit on a state’s regular or special elections.
To be sure, the “time, place, and manner” clause (of Article I, section 4) by its own terms concerns only regular congressional elections—not gubernatorial appointments. An argument can be made that state legislative power over special elections under the “as the legislature may direct” language of Section Two of the Seventeenth Amendment should track legislative power to regulate “times, places and manners” of regular federal elections. But, again, neither of these provisions speaks directly to the “process” of gubernatorial appointments.
Still, if anything, the absence in the Seventeenth Amendment of specific textual state legislative power to do anything in senatorial appointments other than “empower” governors to make them would indicate less leeway for state legislatures to regulate governors than to regulate voters. Even if there is some intuitive appeal for allowing state legislatures some leeway in structuring gubernatorial appointment processes in ways that seem truly procedural, specifying party identity would fall outside such leeway.
Another set of challenges to a state law mandating party consistency goes to the very premise of these laws in the first place. Does maintaining party consistency really implement the will of the past voters? Suppose, for example, that Senate vacancies are caused by scandals that tar one political party or group of individuals, or that call into question the legitimacy of the (now departed) incumbent’s election itself. Shouldn’t governors be able to take account of evolving attitudes by the state electorate about who is fit to serve? Isn’t that why the Seventeenth Amendment designates governors as the people’s surrogates? After all, governors (unlike members of the legislature) are accountable statewide and ignore what the statewide voters want at any given moment at their (and their party’s) peril. So, might a legislatively mandated party consistency provision largely be a solution to a non-existent problem?
Relatedly, if the premise behind “maintaining the voters’ wishes until the next election” is a sound one, why limit ourselves to party affiliation? Could a state legislature require the governor to pick a temporary replacement of the same race, sex, age, or occupation of the departing senator, or somebody with the same views on abortion, vaccines, or defense-spending? Proceeding down this slippery slope would very quickly take us into the out-of-bounds realms of impermissible “qualifications for office.”
Another aspect of Oklahoma’s vacancy-filling statute presents similar problems. Oklahoma requires that anyone who accepts a temporary gubernatorial appointment sign a pledge not to run in the special election. Enforcing a voluntary pledge like Inhofe’s is one thing; forcing would-be appointees to promise not to run seems to us very close also to imposing a term limit (a quite short one at that) directly upon individuals who hold Senate office, in obvious violation of Thornton. (In law-professor style, we flag here two variants on these questions: Could a state enforce a promise a candidate in the special election chooses—but is not required—to make not to run for another term? Or a promise a candidate for the next full term makes to retire early?)
The problematic features of the Oklahoma statute are perhaps not necessarily fatal insofar as state courts (or other bodies, like the Senate itself?) might rule the unconstitutional conditions to be severable, thereby leaving the governor free to use the statute to appoint Inhofe to his own (former) seat until the November election. That would accomplish (nearly) all of what Inhofe and the state currently seem to want.
A final few possibilities bear mention. To get around the question (raised by Jones) of whether a full-fledged special election can under the Seventeenth Amendment be scheduled and held before there is an existing vacancy, could state law be amended to provide that when a senator makes a promise to step down in the near future the state schedules and holds a non-election formal poll, in which qualified voters (and only qualified voters) can express their views about possible replacements? Certainly states can hold polls whenever they want, irrespective of the Seventeenth Amendment’s election timelines. And could state law further provide that if the resignation does come about on the expected (and ripe) timeline, the governor is morally obligated to fill the temporary Senate vacancy (until someone is duly elected) with the person the voters favored at the poll (say, under pain of some politically damaging ballot designation when it comes time to consider the governor’s reelection, to the effect that he ignored the wishes of the voters)? Or could he be required (rather than just encouraged) under state law to follow the wishes of the voters and appoint the person most favored at the poll? We said earlier that the legislature constraining the governor in appointment choices is problematic. But query whether constraining the governor’s appointment discretion to directly vindicate the recently expressed wishes of the people, whose views the Seventeenth Amendment seems to elevate above all else, is a different matter. In this regard, note that prior to the Seventeenth Amendment, many states made use in Senate-election processes of Scarlet-Letter ballot designations and similar devices to impose pressure on state legislatures. In early twentieth-century Oregon, for example, voters passed an initiative that called for such a statewide poll and purported to bind (as a matter of state law) state legislators to elect as Senator the person who gained the greatest electoral support from the state’s general electorate in the poll.
As we said at the outset, we are spotting and analyzing rather than fully resolving issues, so we’ll simply allow our last few suggestions to bounce around a bit in readers’ heads.