How to Improve the Federal Amendment Process Without Formally Amending the Constitution – N.Y.U. Journal of Legislation & Public Policy

By Ethan Herenstein

April 24, 2023

*This is the tenth article in our series The Promise of an Amendable Constitution in an Uncertain Era. You can view the rest of the pieces in the series here.*

It is no secret that it is hard to amend the Constitution. Indeed, as Richard Albert has written, our Constitution is “the world’s most difficult to amend.” One common type of proposed solution is to amend how we amend the Constitution, such as by lowering the bar for constitutional change or by changing the order of operations. These proposals, however, present a paradox: How can we amend the Constitution’s amendment process if it is impossible to amend the Constitution?

A part of the solution lies in the recognition that the Constitution is difficult to amend not just because of the text of Article V—the section dealing with amendments—but also with how it has been interpreted. In a poorly reasoned case from more than one hundred years ago—Hawke v. Smith¹—the Supreme Court adopted a reading of Article V that disempowers voters in the federal amendment process. The ruling effectively divests voters of the power to apply for a constitutional convention, to ratify proposed amendments, to block state legislatures from ratifying unpopular amendments, or to otherwise intervene in the federal amendment process, leaving them at the whim of today’s state legislatures, which are, as Miriam Seifter has explained, “almost always a state’s least majoritarian branch.”

In this short essay, I offer some reasons to question Hawke. While I don’t have the space to demonstrate that it is wrongly decided, I hope to show that the opinion is worthy of reconsideration.

Start with the text of Article V. It assigns important roles to state “legislatures” in both the proposal and the ratification of constitutional amendments. The Proposal Clause specifies two ways of proposing constitutional amendments: (1) two thirds of each house of Congress can “propose” amendments or (2) two thirds of state legislatures can make an “application” to Congress to call a national constitutional convention “for proposing amendments.” Once an amendment has been proposed, the Ratification Clause empowers Congress to choose between one of two ratifiers: (1) three quarters of state legislatures or (2) three quarters of state constitutional conventions. These options yield four potential paths for constitutional change, and state legislatures play a key role in three of them.

In Hawke, the Supreme Court ruled that the term “legislature,” as used in the Ratification Clause, referred to the actual state legislature, rather than the state’s lawmaking process. The Court therefore nullified a state constitutional provision that had required that federal amendments be submitted to a popular referendum after approval by the legislature. As a result, the Court held, Ohio voters were powerless to prevent their state legislature from ratifying the Eighteenth Amendment.

Although the case was unanimous, the issue was hotly contested at the time, as the Eighteenth Amendment wound its way toward ratification. Before Hawke, at least two state supreme courts had come out the other way.² Most illustrative is the Supreme Court of Ohio’s ruling, which the U.S. Supreme Court reversed in Hawke. The Supreme Court of Ohio ruled that the term “legislature,” as used in the Ratification Clause, refers to the legislative power in a state, rather than just the formal state legislature. For this proposition, the court relied on Ohio ex rel. Davis v. Hildebrant,³ a U.S. Supreme Court opinion from a few years earlier that dealt with a similar issue in the context of Article I’s Elections Clause. Just as the Ratification Clause tasks state “legislatures” with ratifying constitutional amendments, the Elections Clause tasks state “legislatures” with “prescrib[ing]” the “times, places, and manner” for congressional elections.⁴ In Hildebrant, Ohio voters rejected the state legislature’s congressional map by a referendum vote. Even though the Elections Clause uses the term “legislature,” the U.S. Supreme Court approved the use of the referendum as a valid “part of the legislative power” in Ohio. Hildebrant thus established, as the U.S. Supreme Court would later put it, that “‘the Legislature’ did not mean the representative body alone,” but “encompassed a veto power lodged in the people.”⁵

In Hawke, the Supreme Court of Ohio concluded that if the referendum comports with Article I’s Elections Clause, then it should comport with Article V’s Ratification Clause. To be sure, the court recognized that making rules for congressional elections is “purely legislative,” whereas ratifying federal amendments is a “federal function.” But the court emphasized that ratification is a form of legislation—indeed, “[t]he making of the Constitution is the highest function of legislation.” Accordingly, the court concluded, “all of the agencies provided by the state for legislation should be empowered to act in accordance with the provisions made by the state at the time the action on the ratification should be taken.”⁶

As noted, the U.S. Supreme Court disagreed. Distinguishing Hildebrant, the Court ruled that “ratification by a state of a constitutional amendment is not an act of legislation within the proper sense of the word.” Instead, the Court explained, “[i]t is but the expression of the assent of the state to a proposed amendment,” in which expression “no legislative action is authorized or required.” The Court also emphasized that legislation and ratification find their sources in different constitutions: Unlike “the power to legislate in the enactment of the laws of a state,” which is derived from the state constitution, the “the power to ratify a proposed amendment to the federal Constitution has its source in the federal Constitution.”⁷

Hawke has cast a long shadow over Article V. Although the case dealt with the Ratification Clause, courts have since applied similar reasoning to the Proposal Clause. Applying Hawke, state supreme courts have ruled that voters may not use the ballot initiative process to make an application for a constitutional convention.⁸ For similar reasons, courts have ruled that a governor may not veto a state legislature’s application for a constitutional convention.⁹ Likewise, the U.S. Supreme Court has suggested that a President may not veto congressional amendment proposals.¹⁰

The operative question, then, is whether the act of proposing or ratifying a federal amendment is “an act of legislation.”¹¹ Unfortunately, the U.S. Supreme Court has given virtually no guidance as to what makes something an act of legislation for purposes of Article V. The most it has offered is the question-begging definition that the legislative function “is that of making laws.”¹² The Court has also implied that certain other functions assigned by the federal Constitution to state legislatures are not legislative. Examples include the “electoral function” that state legislatures used to exercise when choosing U.S. Senators under Article I, section 3 (since superseded by the Seventeenth Amendment) or the “consenting function” that state legislatures exercise in relation to the acquisition of lands by the United States under Article I, section 8, paragraph 17.¹³ But the Court has not explained why these are not legislative functions.

There is a colorable argument that proposing and ratifying constitutional amendments are legislative functions. As Michael Stokes Paulson has explained, “the actions taken by Congress and by state legislatures in voting for any particular amendment language must be understood as ordinary legislative enactments of those bodies . . . , made in accordance with each body’s usual processes and subject to the usual understanding of how legislation is made.” Recognizing that his approach is in tension with Hawke, Paulsen suggested that the best approach “probably is simply to recognize that Hawke was wrongly decided.”¹⁴

To be sure, amendment proposals and ratifications would constitute an unusual type of legislation. Jonathan Walcoff has argued that legislation should be defined as “a sovereign state enactment not subject to any other state’s action.”¹⁵ On this view, ratification would not count because it “is merely the casting of a single vote” and takes effect only when enough other states “take the same action.” And a proposal is even “less akin to ordinary legislation” since it’s one more step removed from changing “the rights and duties of the citizens.”

Walcoff’s definition, however, might be too narrow. Outside of the federal amendment process, state legislatures pass laws that do not take effect unless and until some out of-state-actor takes some other action. For example, many states have enacted into law the National Popular Vote Interstate Compact, which takes effect only if enough other states enact the same law. In addition, many states have recently passed “trigger” abortion bans that took effect only after the U.S. Supreme Court overturned Roe v. Wade¹⁶ last year. If these enactments count as “legislation,” why shouldn’t amendment proposals and ratifications? At the very least, the ambiguity should cut in favor of revisiting Hawke, since, as David Pozen and Thomas Schmidt have argued, courts should embrace “an interpretive presumption favoring ease of amendability.”

We are in an amendment dry spell. It has been more than fifty years since Congress has last proposed an amendment; states have never convened a constitutional convention. The simple story is that, in our age of hyper-polarized politics, it is nearly impossible to obtain the deep consensus required for constitutional change. But Hawke is also a part of the story. Without Hawke, voters could bypass state legislatures to apply for a constitutional convention, and they could ratify popular amendment proposals without legislative interference. What is more, with fears of a runaway convention mitigated by the fact that the voters themselves could refuse to ratify unpopular proposals, states might call for a constitutional convention for the very first time. These changes would help to reinvigorate the federal amendment process—all without formally amending the Constitution.

Ethan Herenstein is counsel at the Brennan Center for Justice at New York University School of Law.

Suggested Citation: Ethan Herenstein, How to Improve the Federal Amendment Process Without Formally Amending the Constitution, N.Y.U. J. Legis. & Pub. Pol’y Quorum (2023).

1
253 U.S. 221 (1920).
2
Hawke v. Smith, 100 Ohio St. 385 (1919); State v. Howell, 107 Wash. 167 (1919). By contrast, the state supreme courts in California, Maine, and Michigan ruled that a popular referendum violated Article V. See Barlotti v. Lyons, 182 Cal. 575 (1920); Opinion of the Justices, 118 Me. 544 (1919); Decher v. Vaughan, 209 Mich. 565 (1920).
3
241 U.S. 565 (1916).
4
U.S. Const. art. I, § 4.
5
Ariz. State Legislature v. Arizona Indep. Redistricting Comm’n, 576 U.S. 787, 805 (2015).
6
Hawke, 100 Ohio St. at 389-90.
7
Hawke, 253 U.S. at 230-31.
8
AFL-CIO v. Eu, 36 Cal. 3d 687, 703 (1984); State ex rel. Harper v. Waltermire, 691 P.2d 826, 831 (Mont. 1984).
9
Opinion of the Justices to the Senate, 373 Mass. 877 (1977).
10
Hollingsworth v. Virginia, 3 U.S. (3 Dall.) 378, 381 (1798); INS v. Chadha, 462 U.S. 919, 955 n.21 (1984).
11
I have elsewhere explained how this legislative function test undermines the so-called independent state legislature theory (ISLT). Crucially, the ISLT is wrong, even if Hawke is right.
12
Smiley v. Holm, 285 U.S. 355, 366 (1932).
13
See Ariz. State Legislature, 576 U.S. at 806-07.
14
Alternatively, Vikram Amar has suggested that Hawke is “susceptible of a narrower reading,” according to which the case “can be understood to be more about the lack of remedies for state law breaches than . . . about state law conflict with Article V.”
15
Jonathan L. Walcoff, Note, The Unconstitutionality of Voter Initiative Applications for Federal Constitutional Conventions, 85 Colum. L. Rev. 1525, 1537 (1985).
16
410 U.S. 113 (1973).