Deliberation governed by rules of procedure is a mainstay of constitution writing in the American states. Following the federal example, each of the fifty states has had at least one constitutional convention[1] in the course of its history, where delegates from a state gathered to deliberate and collectively author that state’s foundational text. As Jon Elster notes, constitution writing need not be democratic or deliberative in terms of both writing and adoption.[2] The Irish constitution, for example, was not written in a convention or assembly, but by a handful of individuals. On the other hand, constitution writing can be steeped in democracy and deliberation. For one, delegates to a convention can be directly elected by the people. Alternatively, the delegates can be elected by state legislatures, which are, in turn, elected by the people. At the convention itself, a procedure is established wherein proposals are discussed by the group and amended and adopted by the majority of that group. The constitutional text that results is subsequently put to the people. Democracy, then, prevails at multiple levels, from the appointment of the convention’s members, to the discussion and drafting at the convention, to the enactment of the resulting text. The people, then, have greater involvement in their governing text than in texts written by individuals.
There is also benefit in discussion prior to making political decisions. James Fearon outlines six of these, notable among which is that ‘[v]oting systems always require some prior statement of what the alternatives are, which is precisely what discussion might be intended to reveal…’[3] Additionally, discussion allows members a chance to offer their expertise on various points. To this end, conventions appoint committees comprised of subject matter experts to offer their specialised knowledge on certain topics. Different conventions approach this relationship differently, by restricting or leaving unregulated the ways in which the plenary committee can act upon the work of another committee.[4] The ways in which committees interact with one another and the Convention at large are prescribed in the convention’s rules of procedure.
The crux of the matter is that deliberative constitutional conventions are defined by an interplay of individual actors, groups, and sub-groups, often with different expertise as well as competing interests and opinions. It is the job of parliamentary procedure not only to direct the minute-by-minute behaviour of the convention but to manage the relationships among the component individuals and groups so that the text that results from these deliberations is the product of consensus.
This paper demonstrates, using an example from the 1910 Arizona constitutional convention, that the purpose of an established parliamentary procedure is not only to govern the practicalities of discussion, but to ensure democratic legitimacy of deliberative constitution-writing. The Arizona constitutional convention of 1910 provides an interesting example of the breakdown of procedure and its undermining effect on a constitutional text. That convention employed a procedure especially obsessed with consensus, but in its consideration of Substitute Proposition Number 94, subverted that procedure. This moment in the records, where procedure is cast aside during debate over a contentious section, illustrates procedure’s important role in ensuring democratic outcomes, and interrogates what is at stake when the mechanisms that attempt to ensure these democratic outcomes are not followed.
The Role of Procedure at the Arizona Constitutional Convention
Following the Arizona-New Mexico Enabling Act of 1910, fifty-two delegates from twelve of Arizona’s counties met in Phoenix to draft the state’s constitution. As the forty-eighth state to enter the Union, the Arizonan framers were well situated within a tradition of parliamentary procedure and state-level constitution writing. Though parliamentary manuals, such as Thomas Jefferson’s Manual of Parliamentary Practice for the Use of the Senate in the United States, Luther Cushing’s Lex Parliamentaria Americana, and Henry Martyn Robert’s Rules of Order, to name a few, attempted to codify guidelines of procedure in America, a survey of state constitutional conventions demonstrates that conventions did not take any one system, or even the U.S. federal example or other state convention procedures, as a definitive guide. Rather, they operated within a procedural scaffolding that was familiar, but they maintained the power to adapt procedure to their needs.[5] To this end, one of the Arizona delegates’ first acts upon their assembly was to appoint a Committee on Rules, who drafted a document outlining the practicalities governing the convention’s work. This document contained fifty-eight rules, and its purpose was to structure debate for this particular set of circumstances, rather than to define in detail every minute aspect of parliamentary procedure. That, the committee left to an existing manual. Rule 57 reads: ‘The rules of parliamentary practice comprised in Robert’s Rules shall govern the Convention in all cases in which they are not inconsistent with the standing rules and orders of the Convention.’[6] In other words, the committee thought of some rules as important enough to specifically enumerate and others as quotidian enough to defer to an existing manual. According to the wording of Rule 57, the rules enumerated in the committee’s rules document may even contradict those named in Robert’s Rules. For this reason, the very exercise of codifying this distinction implies something not just about the convention’s duties but also its values.
As outlined in their rules, the Arizona delegates approached constitution-writing in an open-proposition format, meaning that rather than basing discussion on a draft text submitted by an individual or delegation (such as the Virginia Plan in the U.S. federal example), individual delegates presented propositions on specific subjects to the Convention at large. The introduction of propositions is described by Rule 52, which the Convention approved as read.[7] It reads: ‘Rule 52. All matters intended to become a part of the Constitution shall be presented by a member of the Convention or one of the standing committees thereof, in the form of a proposition, and shall be in writing and in the following form […]’
The procedure for considering these propositions is defined by Rule 53: when a proposition was introduced, it was read the first time in full and sent for printing.[8] It would then be read a second time in full in the Convention and referred to the appropriate standing committee, comprised of the convention’s experts on that topic. The committee’s task was to consider the propositions referred to it and return its recommendation for each proposition for indefinite postponement (i.e. to effectively drop the proposition from consideration), adoption, amendment, or to offer a substitute proposition on the same subject. Though the standing committee’s recommendation was presented to the Convention, the next body to consider both the original proposition and the standing committee’s recommendations, including any amendments or substitute propositions, was the Committee of the Whole, whose task was to prepare a report recommending that the Convention either recommit the proposition or adopt the recommendation of the standing committee report, or offering its own recommendation. The proposition and all reports were then returned to the Convention, and the question was put on the Committee of the Whole report and standing committee report. If the proposition was not amended at this stage, the question was put on third reading. If it was amended, it was sent to the Committee on Style for engrossment. If the Convention approved the Committee on Style’s engrossment or revision of the language, the proposition was returned to the Committee on Style for incorporation into the draft Constitution. The Committee on Style would craft the approved propositions into a constitutional text, which would be referred to the Convention and then immediately to the Committee and of the Whole to be taken up section by section. The text as amended in the Committee of the Whole was next considered section by section by the Convention. Once adopted, the draft was referred again to the Committee on Style for enrollment and then reported to the Convention for the final vote.[9]
In short, the process of approving a proposition for inclusion in the constitution was careful and complex. From its introduction to its referral to the Committee on Style and Drafting for enrollment, a proposition was considered by four committees—Convention, standing committee, Committee of the Whole, and Committee on Style, Revision, and Compilation—for a total of seven times. This meticulous consideration of proposed constitutional text attempted to ensure two outcomes. First, the will of the convention regarding a certain proposition was confirmed by reiteration. Five of the seven stages of scrutiny posed opportunities to kill a proposition: first reading, second reading, the consideration of the standing committee and Committee of the Whole reports in the Convention, third reading, and the clause by clause consideration of the whole constitution in the Convention. It can hardly be argued that a provision which survived scrutiny seven times, five of which were potentially fatal, made its way into the constitution by accident. It was, very clearly, the will of the majority of the Convention.
The second of these outcomes is that no one group had undue influence on the process. Other constitutional conventions of the same time followed a much more streamlined approach to committee work than the Arizona convention did. For example, the 1889 Washington constitutional convention (from which Arizona borrowed some its constitutional text), also allowed members to introduce individual propositions, which were referred to standing committees. In Washington, however, each individual proposition did not require three readings in the Convention. The standing committees did not refer each proposition back to the Convention with a recommendation, but used the propositions referred to it as the basis for a draft article. That draft article was then considered in the Committee of the Whole and then the Convention. Once all the articles had passed third reading, they were enrolled and engrossed, and the Convention took one final vote on the constitution as a whole.[10] This is not to say that Washington’s convention was unconcerned with preventing undue influence over textual outcomes but to illustrate Arizona’s emphatic insistence upon it.
In both state examples, although only the Convention could make definitive decisions on amendments and passage, other groups, such as the procedurally-lax Committee of the Whole or a smaller subgroup of subject-matter experts could propose recommendations on amendment or indefinite postponement or, in the case of the standing committees, draft substitute articles. These sub-groups allowed members opportunities to have their say in smaller settings and to put forth recommendations on pieces of text that might have been swallowed among the suggestions in the larger group.[11] And importantly, each of a convention’s committees had a different chairman, meaning that, in theory, one person could not lord over the proceedings by, for example, selectively calling upon members to speak or disallowing amendments. Theoretically, if one chairman attempted to manipulate the discussion, the oppressed delegates had more opportunities, under different chairmen, to have their say.
Protection from undue influence at the Arizona convention was a resounding theme not only in procedure, but throughout a convention dominated by progressive delegates, in content. Articles IV and VI of the final constitution limited office holders from holding any other public office and prevented judges from practising law during their terms, respectively.[12] The convention set short term limits for state and local offices, and prohibited themselves from seeking any public office created by the constitution for five years following the convention.[13] In other words, at the forefront of both procedure and content of discussion at the convention was not just democracy, but an overt interest in preventing any one person or group of people from holding too much power or influence.
Procedure in Practice at the Arizona Constitutional Convention: Substitute Proposition Number 94, Section 29
The outcomes that the Arizona convention’s complicated procedure sought to accomplish—clarity of discussion, identifying consensus, and preventing undue influence—could only be achieved if the individuals involved in the convention and the people at large could trust in that procedure. Not following proper procedure calls not only the very idea of consensus into question but also the idea of a deliberative body free from overly powerful players who can act unchecked. In this way, procedure governs the practicalities of discussion, keeps all actors informed of exactly what is under discussion and what has been agreed, and importantly, lends a sense of legitimacy to the proceedings.
In failing to following procedure, then, deliberative bodies risk the legitimacy of their decisions being called into question. An example of a failure to follow proper procedure resulting in a failure of each of these intended outcomes can be seen in the Committee of the Whole discussion of Substitute Proposition Number 94. Proposition Number 94 was introduced by William Cooper of Pima County and borrowed heavily from the Declaration of Rights in the Washington state constitution of 1889, omitting only a provision on the right of the state to maintain a standing army and another on the enumeration of rights.[14] Like the Washington declaration, Cooper’s proposition did not contain a provision on the right of workers to unionize. Such a provision was introduced at the standing committee stage, when the Committee on Preamble and Declaration of Rights re-inserted the two provisions from the Washington constitution that Cooper had omitted, added seven new sections (not ‘four or five’ as Crutchfield suggested to the Committee of the Whole), and made a number of revisions to the original.[15] One of these new sections was Section 29, which was taken from Fred Ingraham’s Proposition Number 104, and read: ‘Section 29. The people shall at all times have the right to organize into unions, alliances, or otherwise, for the advancement of their interests to abstain either individually or collectively from work for any employer for any reason deemed sufficient by them; to abstain either individually or collectively from dealing with or patronizing any individual, firm, corporation, or association; and freely to speak, write, or publish their sentiments concerning such abstinence.’[16] The committee’s recommendation was that its Substitute Proposition Number 94 be taken in lieu of the original.[17]
Section 29 was hotly contested in the Committee of the Whole, so much so that the Arizona Republican described the debate as a ‘slaughter’.[18] A.C. Baker and W.T. Webb moved to strike out the section on the grounds that organizing into unions was a right the people held at all times, so specifically providing for it in the constitution was ‘unnecessary and uncalled for.’[19] Immediately following Baker’s motion, Thomas Feeney moved to ‘amend that motion by striking out all following the word “then” on line 20, to the end of the section.’[20] The effect of Feeney’s amendment to Baker’s amendment to strike out would be to reinsert the text reading, ‘The people shall at all times have the right to organize into unions, alliances, or otherwise, for the advancement of their interests to abstain either individually or collectively from work for any employer for any reason deemed sufficient by them.’ A debate on procedure followed. P.F. Conelley wished to move an amendment, which the Chairman Jacob Weinberger ruled out of order. Webb explained that ‘an amendment is susceptible of one amendment, and the gentleman from Cochise [Feeney] moved to amend that [Baker’s] amendment, and certainly no more amendments are in order.’[21] In other words, Webb summarised Robert’s Rules, which specifies, ‘This motion [to amend] takes precedence of nothing but the question which it is proposed to amend…It can be amended itself, but this “amendment of an amendment” cannot be amended.’[22] By invoking this rule, both Webb and the Chairman acknowledged Feeney’s motion as an amendment (rather than a substitute) to Baker’s amendment. As such, it had precedence of Baker’s motion, according to Robert’s Rules.[23]
The Chairman, however, attempted to put the question first on Baker’s motion. ‘A motion to strike out an entire section,’ he said, ‘has precedence over an amendment to strike out part.’[24] It is unclear why Weinberger believed this to be true, as it is a rule stated in neither Robert’s Rules nor the Convention’s rules. It is possible that he misunderstood Robert’s rule 23(c), which describes a motion to strike out and insert as being indivisible.[25] Regardless, a second objection to procedure was raised at this point. Mulford Windsor described the chair’s ruling as ‘manifestly unfair, as a section might be amended so it would be favorable to the majority, and by putting the motion to strike out the entire section at a time when it is in a form not favorable to the majority would defeat the true expression of the members.’[26] In other words, Windsor spoke to the important role of procedure in determining consensus, the will of the majority that was independent of whim, circumstance, or disproportionate influence of individuals or groups. The Chair heeded the objection, and Feeney’s motion passed.[27]
If the convention’s defined procedure were followed at this point, the text of Feeney’s motion would be taken into Baker’s motion. According to rule 23 of Robert’s Rules, a motion to amend takes precedence of the motion it seeks to amend.[28] This instance in the Arizona convention debate is unusual in that adopting Feeney’s amendment to Baker’s amendment would cause the text of Baker’s amendment to mirror the text of Feeney’s amendment exactly. Putting the question on Baker’s amendment as amended would require the Convention to vote on two verbatim motions in succession. Typically, an amendment to strike out some text would involve a vote on the amendment to strike out (A) followed by another vote on the text as amended or not amended (B). In other words, there would be a vote to remove the specified phrase and a vote on the state of the text once the phrase was either removed or not removed. Because passing Feeney’s motion would have resulted in these two subsequent verbatim texts, Feeney should have, per Robert’s Rules, offered his motion as a substitute to Baker’s motion rather than amendment to it. This is, perhaps, the root of the ensuing procedural confusion.
Immediately after the adoption of Feeney’s motion, Connelly moved, ‘an amendment to Judge Baker’s motion…that section 29 be adopted as amended.’[29] Connelly, then, mistook Feeney’s motion as a substitute for Baker’s, pertaining directly to Section 29, despite the fact that Connelly was denied the opportunity to move his earlier motion on the grounds that it was an amendment to an amendment to an amendment. The delegates themselves were clearly confused by this sequence of events, and that confusion is reflected in the record of them. The Official Record, Minutes, and the Arizona Republican all provide different accounts of this moment in the debates, but with the same overall conclusion: that Section 29 was stricken from the proposition. The Arizona Republican describes Feeney’s motion as an amendment to Baker’s that was lost from the start.[30] The Minutes’ account of Feeney’s amendment leaves ambiguous whether the amendment pertained to Baker’s amendment or Section 29 itself, but it records the amendment as passing.[31] The Official Record of this moment implies that Connelly’s ‘amendment’ to Baker’s motion was actually a motion to drop Baker’s motion and to adopt the section as amended by Feeney. Neither the Minutes nor the Arizona Republican describe Connelley’s motion at all. Connelly’s motion would have been a legitimate sequence of events had Feeney’s motion been introduced and treated as a substitute to Baker’s motion rather than an amendment to it.
Instead, the Chairman put the question on Baker’s original motion to strike out the entire section rather than on his motion as amended by Feeney. The question on the original motion passed, and Feeney’s amendment, which had been previously adopted, was completely disregarded. While Robert’s Rules does provide for amendments that contradict decisions made previously by a House, that rule applies to substitutes, not disregarding an amendment made to an open proposal.[32]
Discussion
This excerpt from the proceedings demonstrates the Arizona convention’s insistence upon procedure. Two objections to the procedure were raised. One was on general mechanics of amendment, but the other, raised by Windsor, addressed what he saw as procedure being used in a way that would not capture the will of the house. Despite this insistence on procedure, there was clearly a fundamental confusion over what proper parliamentary procedure, as defined by the convention, was. The result is that, in a convention where the delegates took great pains to objectively identity consensus and limit undue influence by individuals or groups by following an iteration of parliamentary procedure that required seven considerations of an article by four committees (and exceeding the same sort of process by most other contemporary conventions), both consensus and limiting undue influence are undermined. The Chair, in insisting on the precedence of a motion to strike out an entire section over a motion to strike out part of a section, treated Baker’s and Feeney’s motions as two separate amendments, each pertaining directly to Section 29, when earlier in the debate, both he and Webb acknowledged, on record, the latter as an amendment to the former. On those grounds, he denied Connelly the opportunity to move an amendment. In this way, Weinberger, whether through pure misunderstanding or with intent, made a ruling that casts doubt, at the very least, over the legitimacy of this section, and prompts the question of whether other important sections were the product of uncertain consensus or manipulation.
It is important to note that these events occurred in the Committee of the Whole, meaning that any decisions made at this point required the concurrence of the Convention. It is possible that in the Convention sitting, a delegate could have moved to reject the committee recommendation on Section 29 and thereby reinstate the section. No such motion was made. However, the membership of the Committee of the Whole and the Convention were identical, so any confusion that proliferated one body over the outcome of a particular piece of text would presumably proliferate the other. That the delegates, secretaries, and reporters were clearly confused by the sequence of events does not inspire confidence in the final decision.
It is also worth noting that the Convention was not given unlimited time to press the issue if delegates were unsatisfied with a particular text. Congress only authorised sixty days of payment for the delegates, and by 25 November 1910, the convention was running out of time.[33] Concern over whether the convention was on track to finish its work led the draft constitution to be presented to the Convention in a four part draft, rather than the full draft dictated by the convention’s rules.[34] This fractured mode of presentation did not allow the house to see all of the articles within context before commencing its consideration. Both Bernard Manin and Thomas Gilligan list time as restrictive in political, group decision-making and beg the question of whether consensus can truly be reached under time pressure.[35][36] Such questions are beyond the scope of this paper, but they add yet another layer of uncertainty to proceedings riddled with uncertainty.
Conclusion
The argument of this paper is not that the Arizona delegates necessarily wanted a section on unions in the constitution and that this transgression of procedure subverted the will of the house. Rather, it is that when procedure is followed haphazardly, whether willfully or through ignorance, the very outcomes intended by an established procedure fail: it becomes unclear to the delegates involved what text is being considered and what text has been adopted, there can be no trust that consensus has been reached, and individuals and subgroups are given disproportional power over textual outcomes.
Though it is just one moment in sixty-one days of discussion, the consideration of Substitute Proposition 94, Section 29 demonstrates how not abiding by procedure complicates the mechanics of discussion and challenges faith in both the process and the final product. To that end, this excerpt raises several counterfactual questions. Would the outcome of Substitute Proposition 94, Section 29 have been different if Weinberger had not been the chairman on 25 November 1910? It was Weinberger who insisted that the question on Baker’s amendment be put in its original form rather than its amended form as would have been proper procedure. In their summary on nineteenth-century precedents of recognition and calling upon speakers in the U.S. House of Representatives, Gilligan and Krebhiel note the influence a chairperson has over a negotiation and how that influence might result in unbalanced negotiation, where certain delegates’ views are privileged over others’.[37] Weinberger’s management of the Section 29 discussion illustrates this point. The role of the chairman in upholding a convention’s rules and procedure is vital in ensuring democratic results.
Further, would the outcome of Section 29 have been different if Weinberger had properly treated Feeney’s motion as an amendment to Baker’s? While it seems unlikely that two verbatim motions put in succession would have had different outcomes (e.g., to adopt Feeney’s and reject Baker’s), it is not impossible. Lastly, this moment in the negotiation raises a further question of whether the people would have had confidence in the final constitution—had they had the education, time, interest, and access to detailed records of the convention and an understanding of procedure themselves—knowing that the procedures which sought to ensure democracy and legitimacy had not been followed?
To conclude, what is at stake in the described scenario and others like it is not just whether a particular section ends up in the final constitution (or other legislation), but confidence in consensus and democracy. The answer to the question of what parliamentary bodies risk in not following their own rules is multifold. As this paper has reiterated, procedure attempts to facilitate orderly and informed debate, identify consensus, and prevent any individual or group from steamrolling the proceedings. All of these intended outcomes bear upon the legitimacy of the text produced. In extension of those intended outcomes, parliamentary procedure protects the people by ensuring that their interests—not just the interests of individual delegates to the convention—are being represented. When parliamentary bodies view the disregarding of well-placed boundaries as a viable option when those boundaries would not result in desired outcomes, democracy and confidence in democracy as an institution are threatened.
[1] In this paper, ‘convention’ refers to the setting of a constitutional convention, whereas ‘Convention’ refers to the plenary committee.
[2] Jon Elster, “Deliberation and Constitution Making,” in Deliberative Democracy, ed. J. Elster (Cambridge University Press, 1998), 98.
[3] James D. Fearon, “Deliberation as Discussion,” in Deliberative Democracy, ed. J. Elster (Cambridge University Press, 1998), 49.
[4] Thomas W. Gilligan and Keith Krehbiel, “Collective Decision-Making and Standing Committee: An Informational Rationale for Restrictive Amendment Procedures,” Journal of law, economics, & organization 3, no. 2 (1987): 287-290, https://www.jstor.org/stable/764831.
[5] Nicholas P.S. Cole, “Writing America’s Constitutions: Understanding the Drafting and Re-Drafting of America’s Foundational Texts,” Rutgers University Law Review 75, no. 4 (2023): 1070.
[6] Minutes of the Constitutional Convention of the Territory of Arizona (Phoenix Printing Company: 1910), 26-41, https://azmemory.azlibrary.gov/nodes/view/135959.
[7] Minutes of the Constitutional Convention of the Territory of Arizona, 41.
[8] Ibid., 33.
[9] Ibid., 33-6.
[10] Quentin Shipley Smith, The Journal of the Washington State Constitutional Convention 1889, ed. Beverly Paulik Rosenow (William S. Hein & Co., Inc., 1999), 21-33. https://digitalcommons.law.uw.edu/cgi/viewcontent.cgi?article=1003&context=selbks.
[11] Elster, “Deliberation and Constitution Making,” 109.
[12] John D. Leshy and Rebecca White Berch, eds., The Arizona State Constitution, 2nd ed. (Oxford University Press, 2013), 17.
[13] Ibid., 11, 16.
[14] John S. Goff, ed., The Records of the Arizona Constitutional Convention of 1910 (The Supreme Court of Arizona: 1991), 1232-6.
[15] Ibid., 658-9.
[16] Ibid., 1251-6.
[17] Ibid., 1236-43.
[18] “The Executioner Arrives,” Arizona Republican, November 26, 1910, 7, Library of Congress: Chronicling America.
[19] Goff, The Records of the Arizona Constitutional Convention of 1910, 676.
[20] Ibid., 675.
[21] Ibid., 675-6.
[22] Henry Martyn Robert, Robert’s Rules of Order (Skyhorse Publishing, 2017), 56.
[23] Ibid.
[24] Goff, The Records of the Arizona Constitutional Convention of 1910, 675-6.
[25] Robert, Robert’s Rules of Order, 56.
[26] Goff, The Records of the Arizona Constitutional Convention of 1910, 675-6.
[27] Ibid., 676.
[28] Robert, Robert’s Rules of Order, 56.
[29] Goff, The Records of the Arizona Constitutional Convention of 1910, 676.
[30] “The Executioner Arrives,” 7.
[31] Minutes of the Constitutional Convention of the Territory of Arizona, 295-6.
[32] Robert, Robert’s Rules of Order, 56-7.
[33] Leshy and White Berch, The Arizona State Constitution, 11.
[34] Goff, The Records of the Arizona Constitutional Convention of 1910, 890.
[35] Bernard Manin, Elly Stein and Jane Mansbridge, “On Legitimacy and Political Deliberation,” Political Theory 15, no. 3 (1987): 349, https://www.jstor.org/stable/191208.
[36] Gilligan and Krehbiel, “Collective Decision-Making and Standing Committee: An Informational Rationale for Restrictive Amendment Procedures,” 296.
[37] Gilligan and Krehbiel, “Collective Decision-Making and Standing Committee: An Informational Rationale for Restrictive Amendment Procedures,” 292.