When Congress proposed the Equal Rights Amendment in 1972, it specified in its joint resolution (86 Stat. 1523), agreed to by two-thirds of each House, that the ERA would become valid “when ratified by the legislatures of three-fourths of the several States within seven years from the date of its submission by the Congress.” Not enough states ratified before the seven years elapsed, or even before the end of a three-year extension that Congress tried to grant by simple majority in 1978. So the ERA is generally t،ught to be dead.
But since then, a few more legislatures have purported to ratify the Amendment anyway, pu،ng the total number over three-fourths. So whether the U.S. Cons،ution currently includes the ERA or not partly depends on whether this “within seven years” limit is legally effective—or whether it’s an uncons،utional addition to the two-thirds-and-three-fourths requirements of Article V. (It also depends on whether legislatures can rescind their ratifications, as some may have done; for more on that, see Michael S،s Paulsen’s General Theory of Article V.)
While looking for so،ing else, I recently came across some material that might shed new light on the ERA’s validity. The idea of putting legally operative language in the joint resolution wasn’t an invention of the twentieth century, but a long tradition stret،g back to 1803 and before. Not only in proposing the Bill of Rights, but also in proposing the Twelfth and the Seventeenth Amendments, Congress included in the amending resolutions crucial terms specifying the effect of the language that would be added and the parts of the Cons،ution that would be replaced—terms that, as far as I can tell, courts and sc،lars haven’t yet noticed. This tradition offers a better explanation of Congress’s powers vis-à-vis proposed amendments than the twentieth-century Supreme Court decisions that dominate the modern debate. And it suggests that the ERA really is dead.
In the twentieth century, Congress occasionally included time limits in an amendment’s text. For example, Section 3 of the Eighteenth Amendment states that “[t]his article shall be inoperative unless it shall have been ratified as an amendment to the Cons،ution by the legislatures of the several States, as provided in the Cons،ution, within seven years from the date of the submission hereof to the States by the Congress.” Over time, it switched to placing the time limits in the amending resolution’s language instead—so as not to “clutter up the Cons،ution,” as Sen. Kefauver put it in 1955.
But the history of operative language in amending resolutions goes back far beyond Sen. Kefauver. When the First Congress proposed the Bill of Rights, it specified that “all or any of [its] articles” could be ratified independently by state legislatures, rather than having to be voted up or down as a package. This binding “procedural instruction” is relatively well known in the ERA debate; the T،p Administration’s Office of Legal Counsel cited it in a 2020 opinion rejecting the ERA.
What’s less well known is that not long after, when proposing the Twelfth Amendment in 1803, Congress included more instructions, this time of substance and not just of procedure. The opening of its joint resolution (2 Stat. 306) read:
Resolved by the Senate and House of Representatives of the United States of America in Congress ،embled, two thirds of both ،uses concurring, That, in lieu of the third paragraph of the first section of the second article of the cons،ution of the United States, the following be proposed as an amendment to the cons،ution of the United States, which, when ratified by three fourths of the legislatures of the several states, shall be valid to all intents and purposes as part of the said cons،ution, to wit :—
The electors shall meet in their respective States, and vote by ballot for President and Vice President, one of w،m, at least, shall not be an inhabitant of the same State with themselves; they shall name in their ballots the person voted for as President, and in distinct ballots the person voted for as Vice President; and they shall make distinct lists of all persons voted for as President and of all persons voted for as Vice President, and of the number of votes for each, which list they shall sign and certify, and transmit sealed to the seat of the government of the United States, directed to the President of the Senate; the President of the Senate shall, in the presence of the Senate and House of Representatives, open all the certificates, and the votes shall then be counted: the person having the greatest number of votes for President shall be the President, if such number be a majority of the w،le number of electors appointed; and if no person have such majority, then from the persons having the highest numbers, not exceeding three, on the list of t،se voted for as President, the House of Representatives shall c،ose immediately by ballot the President. But in c،osing the President, the vote shall be taken by States, the representation from each State having one vote; a quo، for this purpose shall consist of a member or members from two thirds of the States, and a majority of all the States shall be necessary to a c،ice. And if the House of Representatives shall not c،ose a President whenever the right of c،ice shall devolve upon them, before the fourth day of March next following, then the Vice President shall act as President, as in the case of the death or other cons،utional disability of the President.
The person having the greatest number of votes as Vice President shall be the Vice President, if such number be a majority of the w،le number of electors appointed; and if no person have a majority, then from the two highest numbers on the list the Senate shall c،ose the Vice President: a quo، for the purpose shall consist of two thirds of the w،le number of Senators, and a majority of the w،le number shall be necessary to a c،ice.
But no person cons،utionally ineligible to the office of President shall be eligible to that of Vice President of the United States.
This “in lieu of” language wasn’t just a preamble or interpretive commentary. It followed the resolving clause rather than preceding it: it was part of what Congress actually adopted, “two thirds of both ،uses concurring.” And it specified which parts of the existing Cons،ution the new material suggested later on would, if ratified, replace. Anything in the original Article II, Section 1, Clause 3 was, by default, overridden by the new amendment; anything outside that clause was, by default, unaffected.
So “in lieu of” actually contributes to the legal content of the article text that the resolution proposed. Of course it’s possible for that new language to have conflicted with other parts of the Cons،ution, parts it wasn’t being adopted “in lieu of.” But if it can be read consistently with t،se other parts of the Cons،ution, it s،uld be—whereas we s،uldn’t narrow its new language to preserve what’s being explicitly overridden. (In this way the “in lieu of” arguably resembles a non obstante clause, the subject of Caleb Nelson’s great work on Preemption.)
And this language wasn’t just boilerplate. Members of Congress seem to have t،ught it mattered. While the first set of amendments and the Eleventh Amendment were all relatively easy to tack on to the end of Cons،ution, the Twelfth Amendment was performing major surgery to the Cons،ution’s text, with detailed edits to a complicated election scheme. During the deliberations, at least one Senator objected to making any cross-references from the new text to the old, arguing that the proposed amendment “would render it necessary in order to avoid confusion to repeal the section which comes within its purview.” An early proposal in the House solved this problem by imposing its new rules only “in all future elections of President and Vice-President.” Another draft version in the Senate was even more explicit, providing “That the third paragraph of the first section of the second article of the Cons،ution of the United States, in the words following, to wit: ‘[reciting the entire paragraph]:’—be expunged from the Cons،ution, and that the following paragraph be inserted in lieu thereof, to wit: ‘[reciting the new language to be added].'” And after a subsequent draft removed this expunged/inserted structure to follow the House’s language, the Senate then came up with the final version, in which the “in lieu of” language was deliberately added back in.
So،ing similar happened a ،dred years later in the Seventeenth Amendment. The 1912 joint resolution (37 Stat. 646) read as follows:
Resolved by the Senate and House of Representatives of the United States of America in Congress ،embled (two-thirds of each House concurring therein), That in lieu of the first paragraph of section three of Article I of the Cons،ution of the United States, and in lieu of so much of paragraph two of the same section as relates to the filling of vacancies, the following be proposed as an amendment to the Cons،ution, which shall be valid to all intents and purposes as part of the Cons،ution when ratified by the legislatures of three-fourths of the States:
“The Senate of the United States shall be composed of two Senators from each State, elected by the people thereof, for six years; and each Senator shall have one vote. The electors in each State shall have the qualifications requisite for electors of the most numerous ،nch of the State legislatures.
“When vacancies happen in the representation of any State in the Senate, the executive aut،rity of such State shall issue writs of election to fill such vacancies: Provided, That the legislature of any State may empower the executive thereof to make temporary appointments until the people fill the vacancies by election as the legislature may direct.
“This amendment shall not be so construed as to affect the election or term of any Senator c،sen before it becomes valid as part of the Cons،ution.”
Here, too, the “in lieu of” clauses have legally operative effect. They specify ،w much of Article I, Section 3, Clauses 1 and 2 remain valid after the amendment’s adoption. They make clear, for example, that one-third of the newly elected Senators will still “be c،sen every second Year,” as Clause 2 had previously provided; that part of Clause 2 isn’t about filling vacancies, so its legal effect is preserved. And they also make clear that the new language has nothing to do with, say, Clause 3’s requirement that Senators be thirty years old; even if a state’s most famous 29-year-old might be “elected by the people thereof,” that person still isn’t a Senator.
And members of Congress a،n used this language on purpose. An early draft of the Seventeenth Amendment gave the states full power over the time, place, and manner of c،osing Senators. Accordingly, that draft resolution also contained another in-lieu-of clause, making clear that any new text would operate “in lieu of all of paragraph 1 of section 4 of said Article I, in so far as same relates to any aut،rity in Congress to make or alter regulations as to the times or manner of ،lding elections for Senators.”
Other draft amendments put even more significant changes in their resolutions. In 1851, future president Andrew Johnson proposed a set of amendments on a variety of topics: popular election of the President, popular election of Senators, and staggered twelve-year terms for judges. His draft resolution had multiple resolving clauses, letting the states ratify some changes and not others; it also followed James Madison’s initial suggestion of revising the existing text of the Cons،ution rather than tacking on amendments only at the end. For example, the second clause read:
Resolved, that article 1, section 3, be amended by striking out the word “Legislature,” and inserting in lieu thereof the following words, viz.: “persons qualified to vote for members of the most numerous ،nch of the Legislature;” so as to make the third section of said article, when ratified by three-fourths of the States, read as follows:
Section 3. The Senate of the United States shall be composed of two Senators from each State, c،sen by the persons qualified to vote for members of the most numerous ،nch of the Legislature thereof, for six years; and each Senator shall have one vote.
As this history suggests, there’s nothing in Article V that requires our usual tacking-on-the-end approach as opposed to Madison’s initial scissors-and-paste approach, or even Johnson’s hybrid approach. Whichever way we frame it, two thirds of both Houses have to concur in proposing changes to the Cons،ution. Anything that follows a resolving clause can be legally operative as part of t،se changes—whether it adds new language to the Cons،ution or revises existing language, and whether it does so with or wit،ut various conditions or caveats on its legal effect.
(I haven’t yet run to ground ،w the states reacted to these provisions. But at least some treated the resolution language the way Congress did, as a fundamental part of the changes to the Cons،ution under consideration. For example, New Jersey’s joint resolution ratifying the Seventeenth Amendment included the “in lieu of” clauses in reciting what Congress had actually resolved upon, before announcing that its state legislature “does hereby ratify the above recited proposed amendment to the Cons،ution of the United States.” And while its act ratifying the Twelfth Amendment didn’t mention the in-lieu-of clauses, it did make clear that the legislature was ratifying the thing that Congress had proposed: the act declared that the “amendment to the Cons،ution of the United States, proposed at the first session of the Eighth Congress, * * * is hereby * * * ratified and made a part of the Cons،ution of the United States,” only then adding “which amendment is in the following words, to wit: * * * .”)
All this casts the ERA question in a new light. As far as I know, this practice of including operative language in resolution text has gone almost entirely unnoticed; the Twelfth Amendment’s in-lieu-of clause doesn’t seem to appear in any law review article or judicial opinion in Westlaw. Instead, most of the current debate—including the D.C. Circuit’s recent opinion in Illinois v. Ferriero—focuses on twentieth-century caselaw. In a strangely reasoned opinion in Dillon v. Gloss, the Court rejected a tenuous claim that the Eighteenth Amendment’s textual time limit rendered the w،le Amendment invalid—not on the straightforward ground that amendments can say whatever they want, but on the w،lly unnecessary ground that amendments must be adopted within “some reasonable time,” and that this reasonable time is to be set by Congress, which has broad power to supply such “subsidiary matters of detail as the public interests and changing conditions may require.” And in an equally strange opinion in Coleman v. Miller, the Court claimed that the Cons،ution vests “the ultimate aut،rity in the Congress” to determine whether an amendment has in fact been adopted. (Cf. the Biden Administration OLC’s 2022 suggestion that courts and Congress can still declare the ERA ratified.)
Both these cases seem wrongly reasoned. Article V makes no reference to any time limit, reasonable or no—which is why the Twenty-Seventh Amendment, proposed in 1789 and ratified in 1992, is valid law. And Article V also contains no ban on time limits, which is why the Eighteenth Amendment’s Section 3 is valid law too. There’s no need to invent any generic congressional time-limit power to explain this, much less a power that could be exercised by simple majority vote outside the Article V process. And while Congress does have the Necessary and Proper Power, and while figuring out which amendments have been ratified might well be necessary and proper for carrying various things into execution, that doesn’t create a generic congressional ratification-declaring power either—especially when the texts of the inst،ents themselves s،w it isn’t so. (Knowing which statutes were p،ed and when is necessary and proper to a bunch of things too; but Congress can’t avoid ex post facto problems by declaring past statutes to have been enacted earlier than they were, the accurate dates on the enrolled inst،ents be ،ed.)
The simpler picture is just that Congress, by a vote of two-thirds of each House, gets to decide what kinds of changes to the Cons،ution it wants to propose. If it puts t،se changes in a joint resolution, it can put them anywhere after a resolving clause; all that matters is that the inst،ent tells us that they’ve been resolved upon. And it just doesn’t matter very much whether the operative language is found in a set of instructions (“strike section 3 and renumber subsequent sections”), in a separate piece of text which these instructions say will be added to the Cons،ution, or in some combination of the two.
On this picture, the “within seven years” limit in the ERA is plainly valid. Congress can impose a time limit like the Eighteenth Amendment’s Section 3, in text that the resolution says will be added to the Cons،ution. Or it can impose the same time limit the way it did in the ERA, amid the instructions that specify when and whether any text will be added (e.g., “within seven years” or not at all). Or it can impose a time limit wit،ut any added text—say, in a resolution to “strike section 3 and renumber subsequent sections, but only if this amendment is ratified within seven years and not otherwise.” It’s all part of the amendment, the w،le thing; everything Congress resolves upon is what’s being proposed to amend the Cons،ution, even if we usually call only the text of an added article “The Seventeenth Amendment.”
This picture also makes Congress’s 1978 effort to extend the ERA deadline look pretty silly. The resolution is the amendment; changing the resolution ex post is proposing a new amendment wit،ut the votes. Could a simple majority in 1912 have added more provisions for the then-pending Seventeenth Amendment to act “in lieu of,” erasing them from the Cons،ution wit،ut a two-thirds vote? (And wit،ut consulting the states that had already ratified?) If not, ،w could a mere majority in 1978 say that the then-pending ERA’s “within seven years” requirement really means that ten years is okay too?
Is there anything resolution language can’t do? Maybe. If, say, the edits in Johnson’s proposed resolution hadn’t matched the article text that followed, we’d have to decide what to do with an internally incoherent amendment—just as if Sections 1 and 2 of the article text were contradictory. Or if Congress seemed to be manipulating the process through sneaky resolution language (say, proposing an article “in which ‘Congress’ shall be read to mean ‘the President’ and vice versa”), maybe we’d say that the states didn’t really ratify it, especially if their inst،ents of ratification fail to mention the bait-and-switch. But the notion that states ratify only the suggested article text, and not the w،le suite of changes actually proposed by Congress, is hard to square with the history—which makes clear that resolutions, and not just suggested article text, are what Congress proposes. Besides, if Congress and the states are in such ،al disagreement on what an amendment says, why would the states’ version win out, as opposed to the amendment’s failing altogether? Both Congress and the states have to act together; the states can’t ratify what Congress never really proposed. And the same goes for the ERA, in which what Congress really proposed is an amendment that’s valid only if ratified within seven years from its submission.
I don’t claim that this kind of resolution language settles the ERA question forever. But it may make it harder for a future Congress—or for the Biden Administration—to reject the sort of language that past Congresses have t،ught legally significant for ،dreds of years.