Constitutional Conventions


Roger Sherman Hoar


Jon Roland,
Constitution Society

This work remains today the canonical treatise on the subject, but there are a few points that deserve further discussion. The first concerns the relation between the "people", considered as the sovereign authority in a republic, a "convention" to represent them in that capacity, and a "referendum" by which they might ratify the work of a convention.

We begin by considering the similarities and differences between the "people", or "body politic", and a "convention" that represents them for some purpose.

Body Politic Convention
Most members cannot easily communicate proposals and preferences to all the others. All members can hear and be heard by all the other members.
It is difficult to communicate a proposal of any complexity greater than a call to meet in convention to solve some problem. Proposals of as much complexity as the members can read and comprehend can be discussed by all members.
Deliberation is not moderated. Discussion is continuous, but unending and unstructured. Deliberation is moderated by a chairperson, using formal rules of parliamentary procedure, allowing discussion to be ended so that decisions can be made.
Without a cooperative government, there is no way for the members to make common, specific decisions, such as by holding an election or referendum. Can make decisions, either under rules of existing constitution and laws, under rules adopted by the members of the convention, or both, with or without the assistance of government.
Without a cooperative government, has no effective way to control who is a member and who and how they can participate. Can control who will be recognized as delegates and permitted to participate and vote.
Difficult to form majority coalitions and achieve compromise. Easier to form majority coalitions and achieve compromise.
Tendency to divide into factions that don't communicate. More tendency for factions to communicate and come together.
No effective funding mechanism for larger expenses other than voluntary contributions, and get many free riders. Can more easily assess members for funds and cover expenses, which are lower.
Without a cooperative government, has no effective way to divide tasks among specialized subgroups. Must depend on self-appointing voluntary associations who may become factions and may not cover all subjects well. Can divide tasks among a system of officers, committees, and subcommittees, and make sure every task is assigned to someone and that nothing is neglected.
Without a cooperative government, cannot effectively enforce a decision without resorting to revolution or vigilantism. Can enforce its own decisions as long as it remains in session.
Difficult to act with any authority. Easier to act with some authority.

It is clear from this comparison that what we have is a spectrum defined across the domains of size, physical proximity and communications ability of the members. A very small body politic who lived close to one another could meet as a convention on a regular basis, or as needed. The town meeting form of direct popular government is an example of this, but it only works for small towns. On the other hand, as a convention becomes larger, and especially when it becomes too large to fit in any single meeting hall, it begins to take on the attributes of a large body politic, until it becomes effectively unmanageable. Improved communications, such as we can now get using computer networking and conferencing, can facilitate deliberation and decisionmaking by larger groups, but an assembly is still limited by the information processing capacity of its members, no matter how good communications might become. At some point the volume of subject matter requires the division of that subject matter into categories and the organization of the group into committees, subcommittees, or departments, to each of which is assigned responsibility for one such category, with perhaps some overlap in membership to facilitate coordination. This in turn requires some agent, such as a manager or steering committee, to make the assignment of tasks to each such subset of the whole.

For purposes of theoretical analysis, therefore, it makes sense to consider the body politic as a kind of convention, just a convention that may be too large and dispersed to be manageable. If this is done, then one can apply most of what is said by Hoar about conventions to the body politic as a whole, subject to considerations of manageability. Conversely, what Hoar and most others call a convention, to the extent it is supposed to represent the body politic as a whole, can be considered as a "subconvention" of the body politic, considered as a "superconvention".

This leads to the concept that the society has a "constitution", unwritten, for making decisions at the most fundamental level, which is distinct from a written constitution of government that it might adopt. The provisions of this social constitution are that decisions be made by convention, or a succession of conventions, called by due notice and conducted by established rules of procedure. Thus, to constitute a government constitution, there might be a convention called to draft a proposal, followed by conventions to deliberate on whether to ratify the proposal, and perhaps demand amendments, or a referendum which represents a decision by a convention of the whole body politic. In the United States, such ratifying conventions were held in each state, because at that point the states were each sovereign nations, and the proposed Constitution represented the transfer of many of the powers of government from each state to a common national government. Any amendment to the federal Constitution would normally represent another such transfer of powers of government, but the Constitution itself provided that ratifications of amendments could be by convention or by state legislatures, sitting as ratifying conventions, something that state legislatures did not have the authority to do prior to ratification of the Constitution. It also provided that ratification by 3/4 of the states is sufficient for ratification, which effectively transforms the federal union from a compact of states into a federal nation.

In republican theory, legitimacy is based on consent, but the social constitution is critical to the transition from unanimous consent to decision by less than unanimous vote. As John Locke explained in the Second Treatise on Government, each individual consents to the social contract, and by implication, the social constitution, by remaining within the territory which is the exclusive dominion of that society, past some established age of majority. That means, essentially, that for there to be consent, there must be somewhere for each person to go outside the dominion of the society, and if there is not, then either some such external territory must be created, or explicit consent must be obtained. If an individual rejects the social contract and social constitution, and acts in violation of it, the society may exile the individual from the dominion, either beyond national borders, or perhaps to an internal detention area, such as a reservation or prison.

This leads us in a natural way to discuss something that Hoar does not, which is the structured selection of delegates to a top-level convention by a succession of conventions, beginning at the local level, to select delegates to the next higher level, representing a larger territory. Thus, the delegates to a national convention could be selected by first holding conventions at the precinct level, who select delegates to a county convention, who select delegates to a state convention, who select delegates to the national convention. This is, in general, the procedure used in many states by the two major political parties, who have too many participants at the lowest level to effectively select delegates to the national convention without such a multi-level structured process. Of course, each state can hold a primary, and each candidate can appoint the delegates who will attend the national convention, based on the outcome of the primary, either in proportion to the votes each candidate gets, or on a winner-take-all basis. However, this primary-based system of selection requires the assistance of government, and will not work if government assistance is unavailable or unwanted.

Such a system of multi-level selection of a succession of subconventions of subconventions, with each level being a combined convention of the subconvention delegates selected at the previous level, representing some geographic area, until the entire territory to be represented is represented, depends on an established system of successive geographic subdivisions of the total territory. It is easy if one has a nation that is already divided into states or provinces, the states into counties, and the counties into precincts, but without such a structure it may be necessary to create one on an ad hoc basis, and that can introduce controversy about how well each subdivision is being represented as the process rises from one level to the next.

Of course, a body politic may use a multi-level selection process on another basis than territorial subdivisions. Each successive subconvention can represent tribal groups, occupations, religious sects, ideological groups, or other such subset of the body politic, and try to maintain some kind of balance among such subsets at each level. Political parties representing coalitions of special interest groups typically try to do this in their delegate selection processes, so that all constituencies are satisfied that they are represented in proportion to their voting strength. The same process can be seen in parliamentary governments, where several parties may need to unite in a governing coalition, and allocate key cabinet or other positions among each of the coalition partners, as a condition for participation in the coalition.

Hoar is focused on conventions called to propose and perhaps adopt new constitutions or amendments to existing constitutions under some established legislative system, which issues a call-up using what he calls a "convention act".[1] But he does not discuss what is the mechanism for issuing a call-up for what he calls a "spontaneous" convention, or an unsanctioned one. He considers a convention act being a popular referendum, but that still requires some assistance from government.

For purposes of theoretical analysis, it makes more sense to refer to the convention initiation procedure as a "call-up", which could be a constitutional provision, a legislative act, a popular referendum, or it could be a public notice issued by a non-governmental entity. All of these procedures involve some kind of "public notice", so for the sake of generality it makes more sense to use the term "public notice" where Hoar uses the term "convention act".

This change of terms provides a different perspective on the question of who may call a convention. The answer is that anyone may call a convention, by issuing a public notice. The issue then becomes whether enough people respond to the call-up by showing up, and whether they are representative of the body politic they need to represent for the purposes stated in the public notice. If enough do, then it makes no difference who puts out the public notice. It could be any credible individual. This means that Hoar is wrong when he says that a "spontaneous" convention does not represent the people.[2] It may or may not, depending on who shows up. Indeed, in the absence of functioning government, or a revolution or war of independence, there may be no alternative. Hoar seems to accept the ruling of the Rhode Island Supreme Court that Dorr's Rebellion was illegal. The only thing that can be said about it was that it lost in a contest of physical force. It was certainly not "treason". However, Hoar later makes this very point[3] and thus seems to correct his earlier misstatement.

Restated in this way, the issue of whether the convention can be bound by the call-up[4] then becomes more clear. The answer is that it is bound by the terms of the public notice that calls it, because if attendance is voluntary, people will be deciding whether to attend based on those terms. Only the party issuing the public notice may amend it. If the convention exceeds the bounds of the public notice, since it is composed of attendees who decided to attend based on the public notice, it can no longer consider itself representative, and therefore able to make authoritative decisions for the body politic, and it must issue another call-up for a referendum or another convention to ratify anything done or proposed that exceeds the bounds of the public notice that called the first convention. This second convention could be a continuation of the first convention, as Hoar discusses,[5] but with participation by a different, possibly wider, set of delegates, selected or self-selected based on the second public notice.

Many issues also become more clear if we extend the discussion beyond "constitutional" conventions, that is, those called to propose or adopt a new constitution or amendments to an existing one. We have already briefly discussed party conventions. Without getting into conventions of private associations of various kinds, each of which could be considered functional or factional subconventions of the body politic, we need to look at other conventions that have a legal status for the society. The U.S. Constitution refers to several of these: militia, legislatures, courts and juries.

A militia is a convention called up for the purpose of defense. It illustrates that a convention can be for other purposes than deliberation and legal decisionmaking. It can also be for organization, training, the distribution of equipment, and for operations, which for the militia include defense against invasion, suppression of insurrection, law enforcement, and, although the U.S. Constitution neglects to mention it, disaster response.

Because of the importance of participation, a subset of the body politic may be legislatively designated that is required to respond to a militia call-up, with penalties imposed on members of that subset who fail to respond and who lack an adequate excuse for such failure. This subset is properly called the mandatory militia, because participation is mandatory for that subset. In the United States, at the federal level, it includes "able-bodied" males aged 17 through 45 who do not have official duties that take precedence over their militia duties.[6] At the state or local level, other age ranges may be included, 16 through 60 being a typical range, and it may include women, and exclude different officials. Full- or part-time military, such as reserves or the National Guard, or law enforcement officials, are typically excluded as having official duties that take precedence over militia duties, but under some circumstances such persons may be subject to brief call-ups if they are off-duty at the time. While it would be a violation of the posse comitatus act[7] for military personnel to act under orders to enforce the law, they still have the militia duty to enforce the law if they are off-duty and have a crime occur in their presence.

Response to a militia call-up is not limited to members of the mandatory militia. Others may respond and participate on a voluntary basis. The combination of the mandatory militia and those who may volunteer might be called the general militia. If the situation requires it, and they are the only persons available, it may include women, children, and disabled persons. People have the duty to respond if they are the only ones available, even if they are not subject to penalties for failing to respond.

Since the threat or function, such as training, for which the militia is called up, may not require all the members of the mandatory militia in a jurisdiction, the call-up may be issued to a subset of them, and when this is done the result is called a select militia. It might be, say, for able-bodied males aged 18-22 born on the 14th day of any month who have or have not already received a certain course of training. The call-up, as for any convention call-up, can be issued by any credible person, and the "public notice" may consist of nothing more than yelling "Stop thief!", in which case it applies to those within earshot, who must determine whether a theft has apparently been committed and detain the accused until the matter can be more fully investigated. In early England such as call-up was called "raising the hue and cry". In those times an entire community could be held collectively liable for failing to investigate crimes and arrest criminals.[8]

There is no minimum size for the militia. It may consist of a single individual, if that is all that receive the call-up and respond, or need to respond, to it. If one person defends another person from criminal attack, what he is doing is issuing a militia call-up, to which he is the only respondent. If a third person assists, the militia unit consists of the two of you, and if the victim assists, of the three of you. If you at the victim, and defend yourself, then what you are doing is issuing a militia call-up to which you are the only respondent and the member of the society you are defending happens to be yourself.

There is an important moral implication to this. First, as members of society, we all have the duty, the militia duty, to not only obey the law, which is a low level of militia activity, but to help enforce it, which is a higher level of militia activity. The implication is that, just as we do not have the moral option to refuse to defend another person if we can do so without unreasonable risk, so we do not have the moral option to decline to defend ourselves if the person endangered happens to be oneself. Each of us is a member of society like any other, and the obligation is the same whether it be you or someone else.

Another moral implication is that if anyone finds himself in a situation in which there is an apparent conflict of laws or other official acts, then he has the duty to resolve the conflict and enforce the superior. If one of those is the Constitution, which is superior to all other official acts, then it is the Constitution that must be enforced, even if that means resistance to the conflicting official act. This is the theoretical basis for what is sometimes called "judicial review". It is really "constitutional review", a duty of everyone, which is called "judicial review" when done by judges in deciding a case before them. This duty is inalienable, in that it cannot be assigned to another. Each of us must resolve such legal conflicts individually, and we may not relinquish the decision to judges, superiors, or legal advisors. This is the basis for the "principle of Nuremberg" that it is not a defense to a charge of wrongdoing that one was "just following orders".

A further implication is that before it can act, a militia unit has the duty to determine the legality of the proposed action, including of any orders from an official who has apparent command authority over them, and if any significant number of them disagree about whether such proposed action or order is illegal, they must either deliberate to reach a common decision on the issue, or agree to divide, and perhaps to fight one another, something that should provide a powerful incentive to reach agreement. No one, including a majority of the militia assembled, can justly compel anyone to enforce an act he may consider, after due deliberation, to be illegal. Ultimately, each individual must make that decision individually. But it is always to be hoped that agreement is attainable.

A court or jury may be considered a specialized form of select militia, and therefore as a kind of convention. Likewise a legislature may be considered a convention to make statutes under the authority of a government constitution. Again, the importance of participation may make response to the call-up for such specialized conventions mandatory, but it is possible in principle to also make response to a call up for a constitutional convention mandatory. This has seldom if ever been done, but if it seemed likely that too few persons would respond to a call-up to make the work of the convention authoritative, then penalties or disincentives might need to be imposed to insure an adequate turnout. Normally, jury selection is an official function, typically by random selection, or sortition, from a pool of citizens, such as those registered to vote or to drive, but in some jurisdictions it is not uncommon to just grab whoever is in the hallway or on the street.

There are circumstances, however, in which the official jury selection process, or even the courts, may be unavailable or compromised. Under such circumstances, it may be necessary to call a convention to remedy the situation, to appoint a pro tempore judge and other officials, and to select juries, in accordance with law, even though not by established officials. Such extraordinary circumstances may arise due to invasion, insurrection, criminal violence, natural disaster, or to gross corruption or dereliction of duty on the part of the officials to be replaced. The usual mode for replacing them would be to call a convention to conduct a special election, but if an election cannot, as a practical matter, be conducted, or conducted in time, then the convention may have to appoint temporary officials on its own authority, especially if it is representative of the body politic, or at least the politically active members of it. Special care must be taken, if this is done, to make sure that the convention is not factional. Nothing will do more to discredit a judicial or other legal proceeding than the suspicion that it, or the convention that establishes it, is partisan.

When a court is viewed as a kind of convention, then it can also be understood that what we call "due process" is the social constitution, especially as applied to judicial proceedings, and that the social constitution applied to other kinds of conventions is "due process" for such other conventions. Thus, provisions of the written government constitution which call for "due process" incorporate the unwritten social constitution into the written government constitution, although such "due process" is more fundamental. Ultimately, due process is the only absolute right, in that all other rights, including life, limb, liberty, and property, are subject to it.

Hoar does make one erroneous statement, however: "A constitution is a social compact, by which the whole people covenants with each citizen, and each citizen with the whole people, that all shall be governed by certain laws for the common good. In other words, it is the Anglo-Saxon theory that government is in some way based upon a contract between the people and the State."[9] That is a common error which even some legal scholars make, but it is inconsistent with other statements of Hoar elsewhere in his treatise, as well as with sound constitutional theory. In fact, a constitution is a supreme law, legislated by the people in convention assembled, including the body politic considered to be an extended convention. It is not a compact. The underlying compact is the unwritten social contract,[10] which creates the society that functions as a convention. The only way the social contract can be broken is for the society to break up into several separate societies or its constituent individuals. There is no contract between the people and the State. The government is created by the constitution, and does not exist before the constitution is adopted and executed. The only contract is between the people and each government official individually, under which that official agrees to uphold the constitution and perform the duties of his office.

More needs to be said about whether what Hoar calls the "convention act" can be amended by the party issuing it.[11] When we recast it as a "call-up", then it is clear that the party issuing it can amend it, providing that all interested parties get the word in time to change their plans, and the convention does not try to make any decisions on points in the amended but not in the original notice, without submitting them to a referendum or second convention for ratification. The interested parties include those who planned to attend the convention, or who weren't planning to do so for the purposes of the original notice but would attend for the purposes of the amended notice. It also covers such things as changes in the date, time, and place of the convention, and if the date and site are amended too close to the date the convention is to begin, there should probably be persons or signs stationed at the previously announced meeting site and date to let people know who didn't get the amended notice, and direct them to the revised date, time, and site. Decisions by the convention should be delayed to give persons who might have gone to the original site time to get to the new one.

The "doctrine of acquiescence" requires more discussion.[12] Hoar focuses on acquiescence to decisions made by constitutional conventions that are not submitted to popular ratification, but does not examine acquiescence to other kinds of official acts, or define what acquiescence is, something that needs to be done if we are to extend the concept of a "convention" to include collective public decisionmaking of any kind at any level. The problem is that acquiescence may be temporary or illusory. It is not acquiescence as long as there is an ongoing course of resistance that the rest of the people might be watching, giving it time to succeed before resorting to other courses of resistance. This doctrine is sometimes invoked to legitimize the rule of a conqueror over a conquered people, but their failure to resist may be nothing more than biding their time until they can mount a successful revolt. The same might be said about court rulings, legislative acts, or executive actions, as well as constitutional changes. Acquiescence may be temporary, and the people may wait for hundreds of years before abandoning their apparent acquiescence and offering resistance. It might be said that acquiescence is never complete so long as some kind of resistance is being offered by someone, and this is likely so long as the official act is susceptible to being questioned due to defects in the way it was done.

Ultimately, legitimacy rests on consent to due process, the social constitution, and a government constitution adopted by such due process, and fails for acts not based on such due process, and for conventions not conducted by due process, which limits the powers any convention can exercise, and requires a minimum standard of deliberation. The critical elements of due process are due notice and fairness, and while due process can tolerate a measure of human error, it cannot tolerate bias or undue influence on the decisionmaking that results in decisions not based on merit.

In this electronic edition, although we indicate the points at which pages in the printed edition began by inserting the page numbers in braces {}, we have divided the chapters into sections, and added section titles. We have inserted chapter and section numbers into the endnotes, and created links to the section headings rather than to the page numbers, and expect any hyperlinks hereafter to be made to these sections. We have left the page number references in place for historical reasons, but will not support links to them.

Jon Roland
July 4, 1999.
Modified July 24, 2003.

1. Ch. V.

2. Ch. II § 6.

3. Ch. 3 § 3.

4. Ch. VIII.

5. Ch. XIV §10.

6. 10 USC 311.

7. 18 USC 1385.

8. Alice Stopford Green "The Centralization of Norman Justice under Henry II", from Henry II, London: Macmillan, 1888; Frederick Pollock, "The King's Peace in the Middle Ages", Harvard Law Review, vol. XIII, pp. 177-189 (1900).

9. Ch. I § 1, para. 3.

10. Roland, "The Social Contract and Constitutional Republics".

11. Ch. VIII.

12. Ch. XVII.

Also see:

Introduction to The American Republic of O. A. Brownson (1866), by Jon Roland

A Constitutional View of the Late War between the States, Alexander Stephens (1868)

Next | Previous | Text Version | Contents