NOTE: This is an updated version of an essay first published in 2013. For a comprehensive treatment of amendment law, see The Law of Article V: State Initiation of Constitutional Amendments and postings at the Article V Information Center website.

Many lawmakers and activists, and most of the public, now favor a constitutional amendment to impose financial restraint on Congress. Because experience shows that Congress is unlikely to propose such an amendment itself, there is growing interest in the Constitution’s procedure enabling the states to propose an amendment. The Constitution calls the states’ mechanism for doing so a convention for proposing amendments.

A convention for proposing amendments has never been held. While there are a number of reasons for this, a primary reason over the last 50 years has been the “runaway” scenario, first widely popularized in the 1960s and 1970s by liberal politicians, judges, and activists eager to block suggested amendments that would have overruled some liberal Supreme Court decisions (pdf). In one of the ironies of history, a handful of deeply conservative groups subsequently decided to promote the scenario to block the process from being used for any purpose.

The essence of the “runaway” scenario is that a convention for proposing amendments would be a “constitutional convention” in which the delegates could disregard prescribed limits on their authority, and push America further along the road to perdition. The scenario seems to have misled enough people to effectively disable a core mechanism in our Constitution’s system of checks and balances.

I am a constitutional historian—a former constitutional law professor with training in history and classics—who focuses on explicating the meaning of parts of the Constitution. Before undertaking a research project, I typically scan the existing scholarly literature to determine what has been written on a particular subject, and how thorough that writing is. I began to investigate Article V questions in 2009 when I found that the relevant commentary was relatively sparse and mostly of poor quality. My legal and historical research not only corrected much of what had been written, but also forced me to change my mind about several key issues (such as the controllability of a convention). This research has resulted in a legal treatise, a half-dozen major articles and free standing studies (including a three-parter), and a fair number of writings for the general public.

Although I have addressed the “runaway” scenario briefly in a few of these studies, several people have asked me to respond in more detail. I am doing so only reluctantly: Like most scholars, I prefer to pursue my own inquiries rather than respond in detail to uninformed claims. Moreover, I’ve learned that answering “runaway” assertions is a fool’s game—rather like shooting wooden ducks in a carnival shooting gallery: Once you knock down a series of objections, a new bunch always pops up, as two-dimensional as the last.

In point of fact, many of the “runaway” writings are so confused and frantic that they are not worth answering. One of the better essays—and apparently a source for others—was posted on the Internet in early 2013 by Chuck Michaelis, a businessman and amateur historian. To be sure, the Michaelis article, like all the others, contained enough inaccuracies and misunderstandings to cause a professional to shake his head. But it did set a higher tone than many of its genre, so I have decided to frame this essay as a response to the Michaelis piece.

In order to strengthen his case, however, I have re-organized Mr. Michaelis’ argument so that each step leads more logically to the next. The results are as follows:

• A convention for proposing amendments is a constitutional convention akin to the 1787 Philadelphia gathering, which, according to Mr. Michaelis, was America’s “first constitutional convention.” He does not say it explicitly, but clearly assumes that 1787 assembly was the only federal constitutional convention ever convened.
• Mr. Michaelis finds gaps in the language of Article V that, he says, ultimately leave delegate selection and allocation in the hands of Congress, the body Article V charges with the duty of “calling” the convention. Article V, he says, thereby lodges critical power over the process in the same institution that has been abusing its authority.
• An American convention is an inherently uncontrollable body: conventions are instruments of the people’s sovereign will, and the gaps in Article V leave the convention for proposing amendments unbridled.
• A convention likely will disregard limits set in the state applications. Not only do conventions enjoy sovereign power, but the “first Constitutional Convention” proved a runaway, despite “strongly worded” congressional language to control it.
• The convention could obtain the changes it desires by altering the ratification process, as the “first Constitutional Convention” did.
• Such uncertainties render the process a “risky gamble.”
• The existing Constitution is sufficient to deal with the current federal crisis if we elect conscientious people, repeal the 17^th amendment, and reclaim the 10^th amendment.

The last claim—which Mr. Michaelis actually advanced early in this article—is more of a policy call than a matter of constitutional interpretation. It is easily disposed of: We’ve been doing things that way for the last 60 years, and the situation has only gotten worse. Conscientious people have been elected, but they operate in a spending environment that renders it impossible for any but a handful to both (1) be constitutionally-scrupulous and (2) survive in office. Moreover, there is no realistic chance of the 17^th amendment being repealed. Even if there were, we would have to use the state application and convention process to force the Senate to act—the very method by which the states got the amendment passed in the first place. (Anyway, the research on the 17^th amendment’s actual effects on federalism has been ambiguous.) As for the 10^th amendment, states have been attempting for decades now to reclaim it, but without consistent success.

So the real question is whether state application and convention process is such a “risky gamble” that we should abandon it and thereby resign ourselves to leaving the constitutional system unbalanced and on the short road to bankruptcy.

The answer to this question is clearly “no.” Our fiscal problems are very real, but the amendments-convention approach to solving them is not particularly risky.

Like some other writers, Mr. Michaelis supports his claim that the process is risky by using an out-of-context quote from a constitutional scholar. In this case, the scholar is I, and the language he quotes is my acknowledgment that “abuses of the Article V amendment process are possible.” But that is no more than an acknowledgment that everything in politics is possible in at least a small degree. It is not an acknowledgment that the process itself is dangerous.

To understand why it is not, you must understand the meaning of the language in Article V as the courts and other actors are likely to interpret it. This, in turn, requires knowledge of (1) the historical, legal, and linguistic background behind the language, (2) two centuries of post-Founding usage and analysis, (3) governing principles of constitutional, international, and agency law, (4) a long line of Article V court decisions extending from 1798 into the 21^st century, and (5) certain modern political realities.

Runaway alarmists display almost none of this knowledge. For example, although Mr. Michaelis did cite the first part in my 2010-11 three-part study, but seemed unfamiliar with the rest of my work or with the writings of scholars such as Russell Caplan, Ann Stuart Diamond, and Professor Michael Rappaport. Self-restriction to the first part of a single study may be why he claimed I base my conclusions solely on original “intent.” But he must not have read even that first part carefully, or he would know that, strictly speaking, I do not base even originalist constitutional conclusions on original “intent.” And in fact my overall conclusions rest on all of the five factors set forth above, including post-Founding practice, standard legal rules, court decisions, and political realities.

Now, Mr. Michaelis’ next mistake lay in classifying a “convention for proposing amendments” as a “constitutional convention.”

As Professor Diamond has observed, the difference between the two is evident. A constitutional convention is charged with drafting, proposing, and sometimes adopting, a new basic charter. A convention for proposing amendments is charged with drafting and proposing one or more amendments to that charter. James Madison added that in our system the first is “plenipotentiary,” while the second operates subject to the “forms of the constitution.” This was well understood by the Founders and by subsequent generations: Widespread confusion between the two did not commence until more than a century after the Founding. No doubt it arose from ignorance, but it has been fostered by opponents of the process ever since.

Mr. Michaelis might respond that it is fair to refer to any gathering that addresses changes in constitutional rules as a “constitutional convention.”

The problem with this answer is that it renders the term far too broad. If we apply the term that way, the 1787 gathering was not, as he says, our “first constitutional convention.” One must also count the 1754 Albany Congress, which proposed a plan of colonial union; the First Continental Congress, which institutionalized interstate cooperation; the 1780 Hartford Convention, which formally recommended amending the Articles of Confederation; the 1786 Annapolis Convention, which also was called to recommend amendments; and perhaps the Second Continental Congress, which drafted and proposed the Articles.

Moreover, by that definition, we have had many constitutional conventions since: the 1814 Hartford Convention, which proposed constitutional amendments; the 1861 Washington Conference Convention, an assembly of 21 states that proposed a complicated constitutional amendment to avert the Civil War; and all the state conventions that ratified the Constitution or one of its amendments.

Of course it stretches the term to call any of these gatherings “constitutional conventions”—for the same reason it stretches the term to apply it to a convention for proposing amendments.

Our next issue consists of those gaps in the language of Article V that Mr. Michaelis warns about.

If you know the Founding Era record, you know the gaps mostly don’t exist. This is because the Framers employed the key terms in Article V in universally accepted ways. Everyone knew that a general convention would be a meeting of the states. Everyone knew that a “call” did not include authority to dictate the apportionment or selection of commissioners (delegates). Everyone knew—and the ratification record amply confirms—that the applying states would control the subject matter and that each state legislature would control its commissioners.  There was no need to restate the obvious.

But are the Founding Era convention customs and understandings part of Article V? Yes, they are.

This is where legal knowledge becomes helpful: The Supreme Court has held repeatedly that Article V consists of grants of enumerated powers to named assemblies (legislatures and conventions). As a host of judicial decisions make clear, historical customs and understandings largely define the scope of the Constitution’s words and its grants. An example outside of Article V is “trial by jury,” which is undefined in the Constitution but defined by Founding era practice and by judicial decisions following that practice. And while the courts have not always applied the Founders’ understandings to other parts of the Constitution, they have reliably done so in Article V cases.

So why does the language of Article V seem sparse? Because rather than re-state the obvious, the Framers focused on resolving uncertainties not resolved by existing convention practice. The words “application” and “call” sometimes were used interchangeably, so Article V distinguished them. In the Founding Era, moreover, a “call” could come from a state, Congress, or a prior convention—so Article V stipulated who did the calling. In Founding Era practice, a convention might be merely a proposing body or a deciding one. The Framers settled on the former. Article V also specified the ratification procedure and placed certain amendments off limits.

By the way, Founding Era convention protocols did not go away after the Constitution was ratified. They remain much the same even today. The second-to-last general (national) multi-state convention, the Washington Conference Convention of 1861, proved their viability among large bodies and in times of extreme stress. (The most recent general convention of states was in Phoenix, Arizona in 2017, and it followed the same protocols.) Similar procedures governed the convention of western states held in 1922 and the state conventions that ratified the 21^st amendment in the 1930s.

Next we come to the “any convention can do anything” claim. The general idea is that, as the direct representative of the people, no convention can be controlled by any outside force because each convention is sovereign.

You can marshal a few writers in support of that opinion, and in the middle of the American Revolution a few state conventions acted that way. But this view runs contrary to both prevailing practice and established law. When the Constitution was adopted nearly all interstate conventions had been limited by topic, and in the intervening years also this has been true of nearly all conventions.

Moreover, the “any convention can do anything” view directly contradicts established constitutional law. That law holds that when they act under Article V, all assemblies—both legislatures and conventions—derive all their authority exclusively from the Constitution. Their power is limited accordingly. To take one example: A state convention commissioned to consider only a particular amendment can be limited to that purpose.  In Re Opinions of the Justices, 204 N.C. 306, 172 S.E. 474 (1933); see also the relevant bibliography in The Law of Article V for citations to cases.

Some alarmists counter with a speculative essay written by Yale professor Akhil Amar when he was fresh out of law school. (The essay was written long ago and without the benefit of modern Article V scholarship.) In it, the youthful Amar argued that the people can, by convention, reform the political system extra-constitutionally. Now, to anyone familiar with the Declaration of Independence, this is an unsurprising thesis. Amar also suggested that some parts of the Constitution recognize this “popular sovereignty” power. But—and this is the important point—Amar explicitly distinguished the whole idea from Article V. Moreover, under an important rule of legal interpretation, the fact that Article V contains limitations on the amendment process tells us that it is the exclusive way of amending the Constitution.

We now turn to Mr. Michaelis’ assertion that the 1787 convention was called by Congress for the limited purpose of amending the Articles of Confederation, but instead “ran away” by drafting a new document. I address this common misconception briefly in the first part of my 2010-11 three-part study, for which Mr. Michaelis accuses me of “equivocation” and “a long and complex argument regarding the meanings of words.”

But the facts are neither equivocal nor complex. They are as follows:

• The Constitutional Convention was not called by Congress. It was called by Virginia in response to the recommendation of the Annapolis Convention. (During the Founding Era, most multi-state conventions were called by individual states.)
• During the Founding Era, the term “confederation” typically meant a league or treaty arrangement among sovereign states. It did not mean a federal government as we now use the term. The Articles of Confederation were, unlike the Constitution, essentially a treaty among sovereign states. The role of the Confederation Congress was much like the role of the UN among sovereign nations today. Signatories of treaties always have the power to reconsider the terms of their connections, even if their coordinating agent (such as the UN or the Confederation Congress) objects.
• Ten of the 12 states participating in the Constitutional Convention authorized their delegates (“commissioners”) to consider changes in the “federal constitution” without limiting them to amending the Articles of Confederation. The unanimous authority of 18^th century dictionaries (including the first American edition of Perry’s) tells us that “constitution” in this context meant the entire political system, not merely the Articles as such. This is why the Declaration of Independence could refer to “a jurisdiction foreign to our [American] Constitution” even though in 1776 there was no single document governing the colonies.
• It was understood that the convention’s power would derive from the state legislatures, not from Congress. John Jay, then serving as United States secretary of foreign affairs, pointed this out in a January 7, 1787 letter to George Washington. He also noted that such a convention would have only power to propose, not decide.
• The source of the convention’s authority, and its breadth, was well understood in Congress. That’s why the two states that disagreed—New York and Massachusetts—asked Congress to recommend that the convention be restricted to amending the Articles. The New York/Massachusetts proposal was rejected in favor of another resolution. That resolution was certainly not “strongly worded,” as Mr. Michaelis claims. It was about as weak-tea as possible: watered down even from a recommendation to the mere statement that “in the opinion of Congress it is expedient” that the convention be so limited. This is understandable, because Congress, as a mere agent of the participating parties, had no power to limit their decision, and it was presumptuous to try.
• In Philadelphia, only the seven commissioners from New York and Massachusetts lacked power to propose a new form of government. Of the seven, only three signed the Constitution, one of these in an individual capacity (Hamilton). Perhaps Massachusetts’ Nathaniel Gorham and Rufus King “ran away,” but no one else did.

Mr. Michaelis seems to argue—he is not quite clear on the matter—that because 16 commissioners failed to sign the Constitution, all 16 recognized that proposing the document was beyond their power. This is simply untrue. Elbridge Gerry and two New York delegates did refuse to subscribe for lack of authority. But most of the 16 non-signers failed to subscribe for very different reasons: Edmund Randolph wanted to maintain political flexibility (a good choice in retrospect). George Wythe went home early to tend a dying wife. Luther Martin, George Mason, and John Mercer all opposed the terms of the document. Alexander Martin favored the Constitution, but left (probably for health reasons) before the convention ended. And so forth.

As for the convention’s decision to “change” the ratifying process: It is true that the 1787 gathering adopted a process different from that in the Articles. But the 1787 convention was not called under, or empowered by, the Articles. By contrast, a convention for proposing amendments would be held under the “forms of the Constitution” and therefore would be bound by the very clear ratification procedures specified by the Constitution.

What of the fact that most of the state commissions appointing delegates referred to prior congressional approval of the result, but the constitutional text mentioned ratification only by states?

Actually, while many of the commissions contemplated congressional approval, none required that the new Constitution specifically mention congressional approval. The commissioners left the convention free to simply write a letter submitting the document to Congress. And that is exactly what the Convention did; the letter from Convention president George Washington specifically stated that the Constitution was being submitted “to the consideration of the United States in Congress assembled.”

Congress did consider the instrument, and then voted to duly transmit it to the states for their review. Congress refrained from making an additional recommendation, but that was not the fault of the Convention. Congress was unsure of its authority to recommend complete replacement of the Articles, so it signaled its approval by mere transmission to the states.

Suppose, however, that it were true that the 1787 convention “ran away?” Would this prove that a future convention would do so?

There were many multi-state conventions during the 18^th and 19^th centuries.  Why consider as evidence only one? The Providence Conventions of 1776-77 and 1781 did not run away. The 1777 Springfield and York Town Conventions did not run away. Neither the New Haven Price Convention of 1778 nor the Hartford Conventions of 1779 and 1780 ran away. The 1780 Philadelphia Price Convention and Boston Convention did not run away. And, more recently, the 1861 Washington Conference Convention did not run away.  Why is the 1787 convention “evidence” while nearly 20 others are not?

There also have been several multi-state conventions since the Founding: In 1814 in Hartford, Connecticut; in 1850 in Nashville, Tennessee; in 1861 in Washington, D.C.; in 1861 in Montgomery, Alabama; and in 1922, primarily in Santa Fe, New Mexico. All of those remained within their charges. Why is the 1787 convention “evidence” while five held since that time are not?

The answer is that for all their vaunted constitutional expertise, when the “runaway” theorists adopted their position they were ignorant of all or most of those other conventions. They jumped to their conclusion without sufficient knowledge of the historical record.

There is another aspect of this the “runaway” theorists overlook: modern communications. Even if the 1787 convention had run away, modern communications render the comparison an ill-fitting one. As Walter Phelps Hall and Robert Greenhalgh Albion pointed out in their History of England over 70 years ago, before modern communications diplomats were unable to consult home authorities quickly and sometimes had to make decisions that presented those authorities with a fait accompli. But today’s communications enable the authorities to control their diplomats to the point that the latter can be turned into “nothing but damned errand boys at the end of a wire.” At any convention for proposing amendments, the state commissioning authorities will be in constant contact with their commissioners, issuing instructions or implementing recalls, as necessary.

Now, let’s get back to reality: As a practical matter, there are redundant protections against a runaway convention for proposing amendments:

• Political factors: the damage that disregard of clear limits can do to a commissioner’s reputation;
• Popular opinion;
• State applications defining the scope;
• The limit on the scope of the call;
• The potential for lawsuits to enforce the foregoing;
•  State instruction of commissioners;
• State power to recall commissioners;
• The need to garner a majority of state committees (delegations) at the convention;
• Congress’s ability (and duty) to refuse to choose a mode of ratification for an ultra vires proposal;
• The requirement that proposals be ratified by 38 states;
• The potential for more judicial challenge, at every stage of the process.

You can argue against the efficacy of any one or two of these if you like. But combined together, they reduce the risks almost to the vanishing point. Consider, by contrast, the unrestrained reality of the runaway Congress.

One last observation: The Founders adopted the state application and convention procedure as an integral portion of the Constitution’s checks and balances. It is a way of preserving the state/federal balance, and its disuse has had predictable results.

In part this disuse may stem less from a desire to defend the Constitution as from dissatisfaction with the Constitution: Mr. Michaelis, for example, writes of a “flaw in the language of Article V” and that “[t]he language that needs to be there is simply missing.” Others in his same camp have suggested that Article V be ignored in perpetuity or that it be skipped in favor of extra-constitutional remedies such as nullification.

If their view is that parts of the Constitution are radically defective, then they should be careful what they ask for: Because if state legislatures do not step forward soon to establish their “ownership” of the state application and convention procedure, others—people hostile to the Founders’ design—certainly will. They are preparing to do so as I write.

The Founders inserted this procedure for the state legislatures to use, and to use particularly in times of federal overreaching. If James Madison and John Dickinson were to come among us today, and we were to tell them of our current predicament, what would they say?

No doubt, they would ask if we had resorted to the state-driven process in Article V to correct the problem. And when we admitted that we had not—that we had allowed ourselves to be gulled by alarmists and quacks—what would these Founders say then?

They would tell us that the whole mess was our own fault.

And they would be right.