Explaining and Assessing the Decline of State Constitutional Conventions

Guest Blogger

This post was prepared for a roundtable on Constitutional Faith and Veneration, convened as part of LevinsonFest 2022—a year-long series gathering scholars from diverse disciplines and viewpoints to reflect on Sandy Levinson’s influential work in constitutional law.

John Dinan

Of the many aspects of constitutionalism that Sandy has done so much to illuminate and participants in Levinsonfest are addressing, I am especially drawn to Sandy’s exploration of American state constitutionalism, specifically state constitutional conventions. Sandy has highlighted Americans’ frequent reconsideration of and innovation regarding governing principles and institutions. And he has lamented the decline of state constitutional conventions that once served as forums for much of this reconsideration. And in work with William Blake, he has analyzed recent unsuccessful referenda on calling state conventions, with an eye to understanding the public’s views about conventions. 

I have also long been interested in state constitutional conventions and in my remarks, I will consider why, after more than 230 state conventions were held from the 1770s to the 1970s, none have been held in the last 30 years, and I will also discuss the consequences of this change. This is a good time to offer these remarks because in November 2022 three states – Alaska, New Hampshire, and Missouri – will vote on automatically generated referenda asking whether to call a convention. My remarks might be helpful not only in stimulating conversation about a subject that Sandy has thought and written about but also offering some perspective about these upcoming convention votes and informing the participants in these referenda. 

I will start by discussing why state conventions have been entirely absent in recent decades after they were once held on a regular basis. Conventions are less prevalent today in part because of a greater reliance on other mechanisms for changing state constitutions. In some states in the 18^th and 19^th centuries, conventions were the only means of changing constitutions. But once all states allowed for legislature-referred piecemeal amendments – New Hampshire was the last state to take this step, in 1964 – this removed some of the need to call conventions to make constitutional changes, because these changes could be made in an alternative forum and in a more focused way. The 20^th-century adoption of citizen-initiated amendments in 18 states (I am including Mississippi in this count, even though the Mississippi Supreme Court last year invalidated that state’s initiative process, in issuing a decision in a case challenging a recently passed medical marijuana legalization initiative) has provided still another vehicle for changing state constitutions and in this case avoiding any need for legislators to sign off on the changes. And then Florida has since the late 1960s allowed still another alternative to conventions, by providing that one of two constitutional commissions (one commission has broad authority and the other is limited to addressing tax and budget matters) shall convene every 10 years and have the power to submit constitutional changes directly to voters, though voters in November 2022 will vote on a measure to eliminate the more powerful of these two commissions. Efforts to bring about state constitutional changes that would at one point have been funneled into calling conventions are now channeled to other amendment mechanisms, reducing the pressure to call conventions.

But there is more to the story of the decline of conventions. After all, conventions were still called on a frequent basis in the Progressive Era and then again during the Reapportionment Revolution, even after legislature-referred amendments and later citizen-initiated amendments became viable options. The decline really took hold in the 1980s. New Hampshire’s 1984 convention and Rhode Island’s 1986 convention are the last two full-scale state conventions, and Louisiana’s short-lived 1992 convention composed entirely of legislators is the last convention of any kind. 

It is also important to consider legislators’ and voters’ calculations about the risks and rewards of calling convention and the way this balance has shifted in recent years to emphasize the risks. Legislators have for a long time perceived significant risks in calling conventions and have often opposed calling them. From the perspective of legislators, convention delegates are rivals, whether in pushing for institutional changes that limit legislative power, placing items on the agenda that legislators would prefer not to raise, or elevating convention delegates’ profile at the expense of legislators. 

Voters have long held more mixed views on conventions. In prior eras voters often viewed conventions as a way of securing beneficial reforms that would improve governance. This led voters in prior years to sometimes put pressure on legislators to submit convention calls for public approval. Voters were also willing to say yes on various occasions to convention referenda that appear automatically on the ballot at periodic intervals in 14 states (I am including Oklahoma in this count even though Oklahoma officials have ignored this constitutional requirement for over three decades). In recent decades, though, voters have been much more persuaded of the risks of calling conventions, as seen in voters’ rejection of the last 33 automatic convention referenda that have appeared on state ballots (although yes votes exceeded no votes in Hawaii in 1996 and Maryland in 2010, these convention referenda did not secure a majority of votes cast in the entire election and were therefore deemed to have failed). 

The leading critiques of conventions advanced by opponents during ballot campaigns highlight these risks, and in two key respects. First, in what has been most important in driving down support for convention referenda, in a situation where public opinion polls often show support exceeding opposition for much of the campaign, only to plummet in the final weeks before the election, voters have been persuaded that conventions could turn into run-away assemblies and open a Pandora’s box of issues and make changes that go far beyond what was anticipated. A second risk highlighted in recent convention campaigns focuses on the possibility that convention delegates might eliminate provisions that previous generations added to state constitutions, whether regarding worker protections, environmental rights, privacy guarantees, gun rights, or tax limits, among other provisions. 

Certainly, groups supporting conventions have highlighted their possible benefits, from revising outdated constitutional limitations on legislative authority to securing greater transparency and accountability in governing. However, these benefits have been outweighed in voters’ minds by the risks, leading to the rejection of every convention referendum since 1984, usually by sizable margins (fewer than 17 percent of voters supported New York’s 2017 convention referendum), though occasionally suffering narrow losses (Rhode Island’s 2014 referendum attracted support from 45 percent of voters). 

In considering what difference it has made that conventions are no longer taking place and what has been lost as a result of the decline, I am inclined on balance to view the decline of conventions and prominence of these particular critiques of conventions as a reason for concern rather than as something to be welcomed. 

A leading benefit of holding conventions is to permit a comprehensive review and scrutiny of whether constitutional provisions hold together in a coherent fashion, especially when these provisions may have been adopted in different eras and occasionally work at cross purposes. There are, of course, some particularly high-profile cases of incompatibility, most notably in Colorado and California, where tax-and-expenditure amendments were adopted at one time and amendments mandating a certain level of education spending were adopted at a later time, and with the effect of limiting budgeting discretion. But aside from these high-profile and oft-cited examples, state constitutions are in many cases the product of a number of particular foundings, in that groups and officials in different generations have adopted provisions, at times through legislature-referred amendments and at other times through citizen-initiated amendments, intended to serve diverse and occasionally conflicting goals. 

Conventions permit a comprehensive review of constitutional provisions – and in a way that is not possible through piecemeal amendments – for the purpose of not only addressing conflicting provisions but also considering whether institutions adopted by one generation are still operating in an effective fashion. State constitution-makers have through the years adopted innovations regarding the design of the legislature and executive and the structure and selection of the judiciary, whether adopting legislative term limits, a plural executive, the line-item veto, or an elected judiciary, to name just a few innovations. In some cases, it is possible in the contemporary era to reconsider these decisions via the amendment process; but in many cases adjustments to any one of these institutions have broader implications for governance and would benefit from a comprehensive review of the sort best suited to take place in conventions. 

If these are some leading benefits of conventions, how should we assess the arguments that have figured prominently in recent campaigns about the risks of conventions? To the extent that the concern is that convention delegates might stray from public opinion and adopt as part of a constitutional revision some provisions that depart from public expectations and are out of step with the public, the history of state constitutional conventions provides little support for this claim. It is true that some conventions in prior eras opted against submitting their work for voter ratification, most notably when several turn-of-the-20^th-century southern state conventions adopted constitutions replete with disfranchising provisions and simply proclaimed them to be in effect. However, in the last 100 years, since Louisiana’s 1921 convention failed to submit its work, every state convention has submitted its work product to voters for ratification, thereby ensuring that voters can provide a check on wayward conventions and giving delegates an incentive to take account of public sentiment or risk the defeat of an entire constitutional revision at the polls. 

In fact, voters haven’t hesitated to vote down the work of conventions, as in New York in 1967, Maryland in 1968, North Dakota in 1972, and Arkansas in 1970 and 1980, with these rejections fueled in some cases by concern about just one or two provisions in an entire constitutional revision. This suggests that the prospects of run-away conventions opening a Pandora’s box of unanticipated issues are dim. At least the chances are quite low that convention delegates would betray the public trust and remain unchecked by voters. 

If these concerns that have figured prominently in recent convention campaigns can be rather easily rebutted, another leading set of contemporary concerns about calling conventions poses a different and somewhat more difficult challenge. The concern here is not that convention delegates will stray from public opinion by raising unexpected issues but rather that convention delegates will respond to a current generation’s preferences for removing provisions and commitments inserted in a constitution by a previous generation. That is, the concern is not that convention delegates will be unfaithful to the voters who select them but rather that they will be unfaithful to prior generations by reconsidering their work. This is at the root of the concerns of groups who oppose calling conventions out of a fear that protective labor provisions or environmental guarantees or privacy rights or expansive right-to-bear-arms clauses or tax-limitation measures adopted by a prior generation will be reconsidered by the current generation. 

This is a more difficult challenge to assess, because it goes to the root of the disagreement between Thomas Jefferson and James Madison about how one ought to view constitutions. Jefferson’s preference that each generation should be permitted to reconsider and either renew or revise its commitment to constitutional provisions has not been influential at the federal level, at least when it comes to formal constitutional amendment (as opposed to judicial interpretation of constitutional provisions). Madison’s preference for constitutional veneration has won out at the federal level – again, in so far as we are talking about the structure of and approach to formally amending the constitution, as opposed to the role of the judiciary in interpreting the constitution. However, Jefferson’s vision has long been influential at the state level, as seen in the number of conventions that were held until the early 1980s. The recent effectiveness of these critiques of conventions and the string of voter rejection of convention referenda are therefore best understood as representing a departure from a longstanding state constitutional tradition and a development more to be regretted than welcomed. 

John Dinan is a Professor in the Department of Political and International Studies at Wake Forest University. You can contact him at dinanjj@wfu.edu.