Back in 2011, I wrote a law review article discussing concerns that a limited convention for proposing amendments called under Article V could propose one or more amendments outside the scope of the application upon which it was called. Among the many safeguards against such a “runaway convention,” I pointed to the ability of a state legislature to restrict the authority of its delegates to an Article V convention, to require from its delegates an oath or pledge to restrict their deliberations to the subject of the application, and to provide for disqualification of and/or sanctions against any delegate who violates these restrictions. See Michael L. Stern, Reopening the Constitutional Road to Reform: Toward a Safeguarded Article V Convention, 78 Tenn. L. Rev. 765, 786 (Spring 2011). Since that time, at least seven states have passed such “delegate limitation acts” or “faithless delegate” laws. See David F. Guldenschuh, The Article V Movement: A Comprehensive Assessment to Date and Suggested Approach for State Legislators and Advocacy Groups Moving Forward 19 & n. 77 (Nov. 2015).
The concept of DLAs (as I shall refer to them collectively) was in part based on analogous laws that had been passed or proposed to bind presidential electors. See Stern, 78 Tenn. L. Rev. at 786 n. 111. For example, the Uniform Faithful Presidential Electors Act (UFPEA), completed by the Uniform Law Commission in 2010, “provides a statutory remedy in the event a state presidential elector fails to vote in accordance with the voters of his or her state.” The UFPEA provides “a state-administered pledge of faithfulness, with any attempt by an elector to submit a vote in violation of that pledge, effectively constituting resignation from the office of elector.” While only a handful of states have passed the UFPEA thus far, a total of 30 states plus the District of Columbia have some sort of law purporting to bind presidential electors to vote for their state’s popular vote winner.
The constitutionality of the UFPEA and other “faithless elector” laws has been debated and litigated over the past several months. Following the November 8, 2016 presidential election, a concerted effort was made by the “Hamilton electors” to argue that (a) presidential electors are free, as a constitutional matter, to vote in accordance with their own judgment and conscience; and (b) whatever norms might ordinarily compel electors to vote in accordance with the popular vote in their state were overcome by the unique and extraordinary facts of this election. These facts fell into three categories: (1) the failure of Donald Trump to win a majority or plurality of the national popular vote; (2) facts which allegedly showed Trump’s unfitness for office (some of which cannot be mentioned on this family-friendly blog); and (3) facts which allegedly showed foreign interference in the election.
As best you can, clear your mind of any passion these arguments may incite in it. Our objective here is not to pass judgment on President Trump, the Hamilton electors, or even the merits of their respective legal arguments per se. Instead, it is to see what the effort to secure an unbound electoral college, a “runaway college” if you will, tells us about the practical, real-world prospects for an analogous effort with regard to an Article V convention.
Comparing the Constitutional Arguments
Let’s begin with the constitutional arguments. There is a strong originalist argument in favor of the proposition that states cannot prohibit their presidential electors from exercising independent judgment in casting their ballots. The argument begins with the constitutional text, which specifies that the “Electors shall meet in their respective states, and vote by ballot for President and Vice-President . . . .” U.S. const., amend. XII; see also art. II, § 1, cl. 3 (superseded by the 12th amendment). The term “elector” simply means “voter” or, as Johnson’s Dictionary states, “[h]e that has a vote in the choice of any officer.” Indeed, the Constitution also uses the term “electors” to refer to ordinary voters. See art. I, § 2, cl. 1. As a textual and historical matter, it would make no more sense for states to direct presidential electors in how to vote than to direct citizens how they should vote in congressional elections. See Robert J. Delahunty, Is the Uniform Faithful Presidential Elections Act Constitutional?, 2016 Cardozo L. Rev. De Novo 129, 138 (2016).
This textual evidence is strongly bolstered by contemporaneous evidence that the Framers intended and expected presidential electors to exercise judgment in casting their votes. This evidence includes, though it is not limited to, Alexander Hamilton’s words in The Federalist No. 68 (it is from this work, not the musical, that the “Hamilton electors” took their name). See Delahunty, 2016 Cardozo L. Rev. De Novo at 131-37. As Professor Delahunty explains:
On Hamilton’s account, the presidential electors are chosen for the specific purpose of “analyzing the qualities” needed in a president; they will “act[ ] under circumstances favorable to deliberation;” and their decisions will display a “judicious combination of all the reasons and inducements . . . proper to govern their choice.” It would be difficult to affirm more clearly that the electors must exercise judgment and deliberation.
Id. at 136-37 (citation omitted).
To be sure, there are constitutional arguments on the other side as well. From an originalist perspective, Professor Michael Ramsey argues that constitutional text and history support the authority of states to select presidential electors pledged not to exercise independent judgment but to vote for a specified candidate. This argument, however, probably cannot be extended to forcing the electors to actually vote in accordance with their pledges. In any event, the weight of originalist opinion seems to be on the side of the Hamilton electors (see, for example, this by Professor David Post and this by Professor Michael Stokes Paulsen).
The principal arguments against elector discretion would seem to be non-originalist (though Professor John McGinnis notes they have not been as forcefully made as they might have been under other circumstances). These include the general (though not invariable) historical practice of electors voting for the popular vote winner of their state, the evolving societal understanding that the president is selected by a vote of the people, and pragmatic concerns that faithless electors could undermine confidence in the presidential election system. In addition, the Supreme Court’s decision in Ray v. Blair, 343 U.S. 214 (1952), provides support for the proposition that states may restrict the discretion of electors to some degree, though the Court did not reach the issue of whether states could actually force electors to vote for a particular candidate.
The goal here is not to reach a conclusive answer as to the constitutionality of faithless elector laws (or how the courts might ultimately resolve the issue). Rather, the point is that the constitutional argument of the Hamilton electors is not a weak one, and it is in fact much stronger than the analogous argument which might be made regarding an Article V convention.
Why? For one thing, there is nothing in the constitutional text which even mentions delegates to an Article V convention, much less says anything about their exercise of judgment or discretion. This can be seen by examining the principal argument against a limited Article V convention. Professor Paulsen has argued that the term “convention for proposing amendments” used in Article V implies that the convention cannot be constrained in the amendments it proposes; he contends that “[t]he text supplies no basis for inferring a power, on the part of either Congress or applying state legislatures, alone or in concert, to limit what the convention may consider.” Michael Stokes Paulsen, A General Theory of Article V: The Constitutional Limits of the Twenty-Seventh Amendment, 103 Yale L. J. 677, 738 (1993).
As I have explained elsewhere, Professor Paulsen’s attempt to infer from the text of Article V an unbounded discretion on the part of the convention is weak. He reads the words “a convention for proposing amendments” as meaning “a convention for proposing such amendments as that convention decides to propose,” but it is at least equally natural, I contend, to read it as “a convention for proposing such amendments as the state legislatures have applied for.” See Stern, 78 Tenn. L. Rev. at 772.
For present purposes, however, the important point is that Paulsen’s argument is directed solely at the discretion of the Article V convention as a whole. He makes no claim that individual delegates have the discretion to vote in contravention of law or instructions from the state they represent. Thus, while Paulsen’s arguments on the Article V convention and the electoral college may seem similar, they are actually quite different. In the latter case, he does not merely argue that the electoral college as a whole may validly elect any person who meets the constitutional qualifications (a point no one disputes), but that individual electors have the discretion to vote for such a person even if their state law purports to forbid them from doing so. Paulsen makes no such argument with regard to delegates to an Article V convention because there is simply nothing in the constitutional text on which to hang that argument.
Equally important, while there is abundant evidence that the Framers expected and intended presidential electors to exercise discretion, no such evidence exists with regard to Article V convention delegates. To the contrary, the commentary of the Framers indicates that they expected the delegates to reflect the views of the state legislatures they represented. James Madison, for example, “adhere[d] to the view that the proposing power given to the convention was merely a ministerial extension of the state’s power to initiate amendments.” Stern, 78 Tenn. L. Rev. at 770-71. Thus, in The Federalist No. 43, he explained that Article V “equally enables the general and the State governments to originate the amendment of errors.” It would be difficult to say that the power of state legislatures to originate amendments was “equal” to that of the Congress if one expected the Article V convention to act as an additional and independent check on those legislatures. Hamilton’s comments in The Federalist No. 85 similarly imply that state delegations will reflect the state’s viewpoint (as established by the legislature in its application), rather than exercising independent judgment. See Stern, 78 Tenn. L. Rev. at 771. This contrasts markedly with Hamilton’s discussion of presidential electors in The Federalist No. 68.
In short, the argument against the constitutionality of the UFPEA and other faithless elector laws is much stronger than that against the DLA.
The 2016 Litigation
So what good did their strong constitutional argument do the Hamilton electors in court? In December 2016 four separate cases were brought by electors seeking declaratory and injunctive relief from enforcement of the UFPEA and similar laws. The electors, all of whom were Democrats pledged to vote for the Hillary Clinton/Tim Kaine ticket, argued that these laws would unconstitutionally interfere with their right to seek and vote for a compromise candidate broadly acceptable to Republican and Democratic electors.
None of these cases were dismissed on the merits. The courts mostly acknowledged that the elector plaintiffs had plausible constitutional arguments that might (or might not) ultimately prevail. See Baca v. Hickenlooper, No. 16-1482 (10th Cir. Dec. 16, 2016) (there is a “debatable” argument the constitutionality of Colorado’s faithless elector law); Abdurrahman v. Dayton, No. 16-cv-4279 (D. Minn. Dec. 23, 2016) (merits of constitutional challenge to Minnesota’s UFPEA are a “close call” with “compelling” arguments on both sides); Koller v. Brown, No. 5:16cv7069 (N.D. Cal. Dec. 16, 2016) (noting that California faithless elector law “cannot be neatly reconciled with the duties Hamilton attributed to the Electoral College in Federalist No. 68,” but contending there is an “equally viable counter-argument” in favor of the law’s viability).
Nevertheless, none of the courts in the four cases granted the relief sought by the Hamilton electors. Not one of the 10 judges who considered the arguments (at the district and appellate levels) favored granting relief.
Reasons varied. In the Minnesota case, the court found the claims non-justiciable because it was too late to grant the relief sought (preventing the transmission of the electoral ballots to Congress). In the other cases, the courts found that the plaintiffs had failed to establish some or all of the elements required for the issuance of a preliminary injunction, such as proving that such an injunction was necessary to avoid irreparable harm. For example, in Chiafolo v. Inslee, No. 2:16cv1886 (W.D. Was. Dec. 15, 2016), aff’d No. 16-36034 (9th Cir. Dec. 16, 2016), the court stressed that Washington law neither prevented electors from voting for the candidate of their choice nor prevented those votes from being counted. Although Washington law also provided that faithless electors could be subject to a civil fine of up to $1,000, the court found that the mere possibility that plaintiffs might face this minimal monetary penalty did not qualify as irreparable harm.
If the Hamilton electors had had more time, perhaps they would have enjoyed greater success. Perhaps. In order to win the immediate objective of having enforcement of a particular faithless elector enjoined, the plaintiffs would need to persuade the court not only that they had the stronger argument on the merits, but that the case was justiciable and met all the requirements for issuance of injunctive relief (including the showing of irreparable harm, public interest, etc.). Even if they could persuade a court to issue such relief, however, it would only prevent enforcement of a single state’s law. In order to truly eliminate the effect of faithless elector laws, the Hamilton electors would have needed to bring a case before the Supreme Court and then convince the Court to issue a decision broad enough to apply to all faithless elector laws. Such an effort could take years.
Even then, it might not have the desired effect. If the Court were to follow Professors Rappaport and Paulsen, it could conclude that states may require electors to take an oath to vote for the winner of the state’s popular vote, but may not compel the electors to vote in compliance with that oath. Such a conclusion would leave electors legally free to violate their oath but might reinforce, rather than undermine, the moral obligation and constitutional norms that have almost always impelled electors to fulfill it.
The experience of the Hamilton electors tells us that any attempt to invalidate DLAs would almost certainly be futile. To begin with, the pool of potential plaintiffs would be limited. Only convention delegates would likely have standing, and even a delegate would not have standing to challenge a DLA enacted by another state. Thus, it would be necessary to recruit plaintiffs (let’s call them the “anti-Madison delegates”) from the delegation of states that have passed a DLA; the delegates from these states would be among those least disposed to bring a challenge.
There is also the question of when a suit could be brought. Courts would almost certainly view a suit as unripe until such time as Congress actually calls a convention. Even then, a court might view the alleged injury as too speculative to support a justiciable case. After all, until the convention meets there is no assurance that delegates will even have the opportunity to consider amendments prohibited by the DLAs (e.g., the convention itself might adopt a rule limiting the subject matter to be considered). Thus, a court might consider any case brought before the convention actually assembles to be unripe.
Apart from justiciability, the plaintiffs would face the same hurdles to a grant of preliminary relief as did the Hamilton electors. As in the elector cases, the possibility that convention delegates might face disqualification, civil penalties or even criminal prosecution may not qualify as irreparable injury sufficient to justify a preliminary injunction. Similarly, the courts may find that the balance of hardships and public interest do not support the grant of relief because the public reasonably expects that delegates to an Article V convention will adhere to their pledge to confine their deliberations to the subject of the state’s application. Cf. Baca v. Hickenlooper, No. 16-cv-02986 (D. Colo. Dec. 21, 2016) (“The public interest tilts substantially in favor of the public expecting and requiring the electors to do what they agreed to do”), aff’d No. 16-1482 (10th Cir. Dec. 16, 2016).
Of course, as already discussed, the argument on the merits for invalidating the DLAs is weaker than that against the faithless elector laws. Thus, the odds for our anti-Madison delegates are even longer.
Even if one or more suits were successful in securing injunctive relief against DLA(s), this would be of little effect in and of itself. As in the case of the Hamilton electors, securing injunctive relief against the law of one or two states would be of minimal significance. In the compressed time available for litigation, it would seem nearly impossible to obtain a Supreme Court ruling with broad implications for enforcement of DLAs generally. And even if the anti-Madison delegates were to pull off this feat, it would still not trigger a runaway convention for the reasons discussed in the next section.
Structure, Norms and Political Incentives
At the end of the day, the failure of the Hamilton electors was not caused by the UFPEA or other faithless elector laws. After all, there are 20 states that have no faithless elector laws at all. In those states, only two electors (both Texas Republican electors who voted for Ron Paul and John Kasich respectively) failed to vote for their state’s popular vote winner.
Moreover, even in most of the states that do have faithless elector laws, there is no penalty or other enforcement for violation of an elector’s duty to vote in accordance with the law. The law establishes that the elector’s duty is legal as well as moral, but there is no other practical consequence. In those states there was only one faithless elector (a Hawaii Democratic elector who voted for Bernie Sanders).
Ironically, the the most faithless electors were in Washington, one of the few states that actually purports to penalize such electors (as we saw in the Chiafolo litigation). Four Democratic electors failed to vote for Clinton, who won the popular vote in Washington. Three of them voted for Colin Powell. One voted for Faith Spotted Eagle (maybe he thought this would be a defense to a charge of being a “faithless” elector—hat tip, Justice Don Willett).
History provides further evidence that faithless elector laws were not critical to the failure of the Hamilton electors. The UFPEA is a recent phenomenon, having been proposed by the Uniform Laws Commission in 2010. (I suspect other faithless elector laws are also of relatively modern origin, though I have not pinned that down.) The uniform act was not prompted by a spate of faithless electors in recent presidential elections. To the contrary, the last time there were multiple faithless electors in a single election was in 1912, when eight Republican electors declined to vote for their party’s vice-presidential nominee. They did have a pretty good reason, though. He died before the election.
According to the website FairVote, prior to 2016 there were only 160 faithless votes in the entire history of the electoral college, of which 41 related to the vice presidency. Of the remaining 119 votes for the president, 63 came from the election of 1872, where again a candidate (this time Democratic presidential nominee Horace Greeley) died before elector balloting day. Thus, quite apart from any laws or potential penalties, faithless electors have been extremely rare, and they have never changed the outcome of any presidential election. (They did, sort of, make a difference in the vice presidential election of 1836).
The UFPEA was designed as a fail-safe mechanism to deal with a “black swan” type event where it was thought a small number of faithless electors could change the outcome of a presidential contest. Professor Robert Bennett, who worked on the UFPEA, explained in 2014 the circumstances that concerned the Uniform Laws Commission:
If the apparent election day electoral college outcome is either indecisive or very close, a number of possibilities would encourage faithlessness, with real—and painful—bite. First, the apparent electoral college winner might have lost the nationwide popular vote. That happened as recently as the 2000 election. Second, the popular vote outcome in the elector’s own state might have been quite close. That too happened in a few states in the 2000 election, and indeed is not at all uncommon. So, an elector might see some political return from voting faithlessly. Third, an elector might well have disapproved of the party’s candidate for the presidency. That was the basis for the AP story [about electors potentially defecting from Mitt Romney] in 2012. Fourth, some unfavorable news about the apparent victor might become public after election day (or even before). Fifth, the (apparent) loser’s campaign might make promises to court faithlessness (an ambassadorship, say, or embrace of a policy position favored by the elector).
The 2016 presidential election presented a perfect storm of the events feared by Bennett. First, Donald Trump lost the national popular vote. Second, he won the popular vote in key swing states by a razor thin margin. Third, because Trump was a complete outsider to the Republican Party, he had little support among the party regulars and activists who typically serve as electors. At the very least, he was not the first choice of many Republican electors, and it seems likely that a fair number would have been personally disposed to replace him if they could. Fourth, unfavorable information about Trump emerged after (and before) the November election, most notably the intelligence report regarding alleged Russian interference in the election. Finally, though the Hamilton electors did not (as far as I know) offer anyone ambassadorships or the like, they sought to find a compromise candidate who would appeal to wavering Republican electors.
Despite this confluence of factors in favor of a “runaway college,” the Hamilton electors were able to persuade only two Republican electors to defect from Trump. They thus fell miserably short of the 37 votes they needed to flip in order to prevent Trump from winning an electoral college majority.
What explains this failure? As we have seen, the UFPEA and other faithless elector laws cannot be the primary explanation because many Republican electors were not subject to any legal prohibition, and few faced any realistic possibility of legal action if they defected. (However, these laws may still have contributed at the margins to the prevention of a runaway college by making it more difficult to peel off some Republican electors who otherwise might have been tempted to step forward as faithless electors. They also may have provided a convenient excuse for some electors who wanted to avoid personal responsibility.)
The more important reason lies in the social norms and expectations that influence presidential electors. These individuals owe their selection to the state political party to which they belong. They tend to be party loyalists with deep ties to the political establishment and fellow activists in their state. Whether or not they have any loyalty to the presidential nominee, they understand that electors have one job—to vote for their party’s nominee.
Some flavor of the political pressures facing electors may be found in this Politico article by Kyle Cheney, who interviewed electors and other Republican sources in the days leading up to elector balloting day. One insider explained, “It’s a matter of personal pride for a state party chairman and a state party to ensure that all the electors that their people elected vote, and vote in the manner in which they are supposed to.” The article goes on to say:
The Republican electors hail from states where Trump won the popular vote and, in interviews, many said that rejecting the will of the voters in their state would be unthinkable. Most are also longtime party loyalists, selected by state GOP leaders for the sole purpose of confirming Trump’s election, which means violating that expectation would all but assure permanent excommunication from the party.
The difficulty of persuading even Trump-skeptic delegates to defect was colorfully summarized by Charlie Potts, an elector from Oklahoma:
“I would prefer that another person had been nominated by the Republican Party and had won the election, but am I going to go against the 14 million [primary voters] who voted for Trump?” Potts said. “And am I going to vote against all the people in Oklahoma who voted two-to-one for Trump? No, that’s just stupid.”
Potts noted that his home state, Oklahoma, makes it a misdemeanor to oppose the statewide popular vote winner—a crime he doesn’t want to commit. But even if he believed those laws can’t be enforced . . . it would take a far more flawed candidate that Trump to convince him to disenfranchise Oklahoma’s voters.
“Let’s say that somehow the American people nominated a guy who murdered 47 people, carved ‘em in a ditch and hadn’t been caught yet, and he got nominated for president,” Potts said. “Well, you know, then I might change my position.”
Remember, the choice of president was one Hamilton believed “should be made by men most capable of analyzing the qualities adapted to the station, and acting under circumstances favourable to deliberation, and to a judicious combination of all the reasons and inducements that were proper to govern their choice.” The Federalist No. 68. There you have the result—our president is to be someone who has killed and carved up less than 47 people. 47. That’s the cutoff.
That gives you some sense of what it would take for the electoral college to run away.
The fact that the public expectations and political pressure take place on a state-by state basis provides an additional firewall against a runaway college. If a group of electors from one state were to announce their intention to cast faithless votes, this might provide political cover for other electors from that state to do the same. But it would be of less comfort to an elector from another state, who would still be left alone to face the legal, political and social consequences there.
As this suggests, there is a significant collective action problem that hindered the efforts of the Hamilton electors. Casting a faithless electoral vote, or announcing an intention to do so, involves a substantial personal cost to that elector. But there is little corresponding benefit unless the elector has some assurance that enough other electors will follow suit to change the outcome of the election. This meant it was difficult to persuade any faithless electors to come forward until there were already 36 others who would do so.
Moreover, it seemed unlikely that even 37 faithless electors could change the outcome of the election. Merely voting for someone who was neither Trump nor Clinton would simply have thrown the election into the House of Representatives, which would have almost certainly still elected Trump. And even if it had been possible to identify a compromise candidate who could have attracted 270 electoral votes from a combination of Republican and Democratic electors (which it was not), there was no guarantee that Congress would have accepted this result when it came to the electoral vote count.
[As an aside, Professor Derek Muller has pointed out that Congress could have established a significant precedent in its counting of the 2016 electoral vote. It could have rejected the votes cast in violation of state law by faithless electors in Hawaii and Washington. Alternatively, it could have rejected votes from Minnesota and Colorado, where faithless electors were removed and replaced because of their attempt to violate state pledges. However, none of these situations resulted in an objection that required Congress to vote on whether or not to sustain it. Since Congress never voted on these questions, the precedential effect of the counting is questionable.]
The same institutional, political and social factors that have prevented a runaway electoral college will also prevent a runaway Article V convention. Like the UFPEA, the DLA is simply an additional fail-safe mechanism protecting a system that is already extremely secure.
To begin with, Article V convention delegates will be selected on a state-by-state basis. They will be appointed directly or indirectly by the state legislature. The delegates are therefore likely to be either state legislators themselves or drawn from the same class of politically active people who run the political parties and fill state and local offices.
These individuals, like presidential electors, will be highly responsive to the concerns and expectations of their state’s political class (particularly, in the case of delegates, its legislature). For the vast majority of states, including but not limited to the two-thirds that applied for a limited Article V convention in the first place, this means confining their deliberations to potential amendments that can fairly be said to fall within the subject matter for which the convention was called. If there is one thing that can be said with confidence from the many debates in state legislatures regarding the Article V convention, it is that no one favors an open or unlimited convention with authority to propose amendments on any subject it pleases.
The same can be said of public opinion more broadly. While admittedly the general public knows less about the Article V convention than about the electoral college, there is every reason to believe it will be instinctively against a runaway convention that could (in theory) propose changes to cherished constitutional rights. And we know for sure that grassroots activists on the right (Eagle Forum) and left (Common Cause) would be fiercely opposed to any effort to expand the convention’s authority.
All of which means that any effort to procure a runaway convention will face the same insuperable barriers as did the Hamilton electors. The effort would have to proceed on a state-by-state basis because each state’s delegation will face different considerations. Some states will have passed DLAs, some will require their delegates to take an oath or pledge, and some will simply have strong norms against a runaway convention.
Even assuming this effort has the kind of political impetus and energy as did the Hamilton electors (which seems unlikely), it will face the same kinds of collective action problems. Few delegates will want to be the first in their state to announce they will ignore the convention’s subject matter limitations, thereby incurring legal, political and social costs without any assurance they will ultimately prevail in proposing an out of scope amendment. Indeed, to some extent the risk-reward proposition for faithless delegates is worse than for faithless electors. A faithless elector’s vote counts (at least unless the state or Congress takes action to stop it), but a faithless delegate’s vote only counts if a majority of the state delegation agrees. Moreover, there is a high likelihood that Congress will view any out of scope amendment as invalid and therefore not submit it to the states for ratification. Thus, the efforts of the “anti-Madison” delegates are even more likely to be fruitless than were those of the Hamilton electors.
In theory, a runaway electoral college could choose anyone as president (at least anyone who meets the Constitution’s minimum requirements) without regard to the outcome of the November popular vote or even whether the individual selected had been a candidate. In practice, however, no runaway college has ever happened, and, as the experience of the Hamilton electors shows, there is no serious risk that it will ever happen. For similar reasons, there is no serious risk that an Article V convention would ever run away.