Entrenchment Reconsidered (Part II)

This brings us to the second pillar of the anti-entrenchment position, which might be called the “quasi-constitutive” argument. Professors McGinnis and Rappaport, who pioneered this argument, explain:

If a legislature could pass a measure that would prevent a successor from taking action, then that measure would function as a constitutional restraint on the future legislature. For example, if the Senate passed a rule prohibiting tax increases, that would function exactly as if it were a constitutional prohibition on tax increases. . . . Because the Constitution permits amendments only under the procedures in Article V and not simply through legislative enactments, the legislature does not have the power to bind itself in the future.

J. McGinnis & M. Rappaport, The Constitutionality of Legislative Supermajority Requirements: A Defense, 105 Yale L. J. 483, 505-06 (1995).

The McGinnis/Rappaport argument can be somewhat hard to understand because, as the title of their article indicates, they defend the authority of a legislative body to adopt rules that prevent a simple majority from exercising legislative power that it would otherwise have under the Constitution. This is acceptable, they maintain, so long as the majority retains the ultimate power to amend or waive the rules.

So, if I understand their argument correctly, it actually would be permissible for the Senate to adopt a rule prohibiting any tax increases, notwithstanding the language quoted above. The constitutional infirmity would exist if the Senate prohibited repeal of this rule (or required a supermajority to repeal it). So long as the majority has the ability to waive or amend the rule, however, McGinnis and Rappaport see no constitutional violation.

Still, it is an open question what this actually means. For example, suppose a measure to raise taxes is introduced in the Senate and a point of order is raised that the measure violates the (hypothetical) Senate rule. According to McGinnis and Rappaport, the point of order could not be overcome by the argument that the rule is unconstitutional. But it seems that supporters of the measure could argue that (1) they are entitled to a vote on whether or not to waive the rule and (2) in the absence of such a vote, the rule is unconstitutionally entrenched. If this is true, it is not clear why McGinnis and Rappaport would so vigorously defend the constitutionality of supermajority rules because it would seem that such rules would be largely meaningless.

Fortunately, McGinnis and Rappaport appear to recognize this problem in their 1995 article. They explicitly distinguish rules that prohibit a majority from changing or waiving an existing rule, on the one hand, and rules that prevent the majority from obtaining a final vote on changing or waiving the rule, on the other. The Senate rule allowing filibusters of proposed rule changes, they make clear, falls into the latter category:

If the Senate voted on whether to change the cloture rule, only a simple majority would be needed to change it. It is true that an attempt to change the cloture rule might be filibustered, but that is another matter. The historical and structural argument presented above does not demonstrate that a majority must at all times be able to obtain a vote on all measures that it desires.

105 Yale L. J. at 507 (emphasis added).

In a footnote, McGinnis and Rappaport note “[i]t is not clear what limitations, if any, the Constitution imposes on a chamber’s power to prevent a majority from obtaining a vote on a measure.” Id. at 507 n. 117. They suggest two possible options. Under the first option, a majority “must always be able to obtain a vote on all measures.” This would mean, it seems, that the majority must always have a readily available mechanism for circumventing the normal rules of legislative procedure that would otherwise prevent a matter from coming to a vote. They acknowledge that a problem with this option is that it would mean declaring unconstitutional in part aspects of legislative procedure, such as the committee system and the filibuster, despite their “venerable origins.”

The second option would be to recognize broad discretion in each chamber to channel the consideration and scheduling of votes on measures, including the use of mechanisms such as the filibuster and the committee system. A problem with this option, McGinnis and Rappaport observe, would be that a scheduling mechanism could in fact function “like extreme insulated repeal rules.” (This, of course, is essentially the argument made against filibustering of rules changes).

In their 1995 article, Professors McGinnis and Rappaport do not take a position on which of these options is preferable. Their subsequent writings, however, including this post recently by Professor Rappaport at The Originalism Blog, imply that they have come down on the side of the first option. (Professor Rappaport has confirmed to me that this is a correct interpretation of their position, which he stresses is based on their view of the best reading of the original meaning of the Constitution, as opposed to Senate practice or precedent).

It seems to me there are several problems with the first option, in addition to the one identified by McGinnis and Rappaport in their 1995 article. First, the majoritarian imperative would logically be applicable to all legislative rules, not just to supermajoritarian rules like the filibuster. In other words, any legislative rule would be unconstitutionally entrenched unless the majority had a readily available mechanism for changing or waiving it at any time. For the reasons I have discussed before, this seems not only contrary to longstanding legislative practice (particularly in the Senate), but inconsistent with the point of having rules of proceedings in the first place, which are designed precisely to restrain the discretion of the majority. Indeed, as Professor Bruhl observes, “[a] chamber’s rules of proceedings are, in a sense, its very own constitution.”

Second, as a practical matter, I am not sure how the first option would work. Clearly it is impractical to allow votes to change or waive rules whenever a senator demands one. Perhaps the only way to satisfy the requirements of the first option would for the Senate to adopt something akin to the discharge petition procedure in the House. If a certain number of senators (presumably 51, although some might argue it should be as few as 26, which could constitute a majority of a quorum) signed a petition, they would be entitled to a vote on changing or waiving any rule. Needless to say, this would be a major change in Senate procedure, and, as far as I know, no one has explicitly argued that such a change is constitutionally required.

Finally, even if one accepts that the first option is required by a plausible constitutional interpretation, McGinnis and Rappaport acknowledge that there are plausible constitutional arguments on the other side. If so, some attention needs to be given, not only to the Senate’s longstanding contrary practices, but to its constitutional precedents over the last century, which have consistently rejected the proposition that a majority has a constitutional right to a final vote on changing the rules. Determining whether these precedents should be discarded in favor of the first option requires considering other constitutional provisions and values, such as whether a legislature should follow its own precedent in interpreting the Constitution.

Despite all of this, I cannot deny that McGinnis and Rappaport have a point when they argue that an absolutely entrenched rule is constitutionally troubling and that in this regard it makes little difference whether it is substantively entrenched (eg, a direct prohibition on repealing the rule) or procedurally entrenched (eg, a prohibition on ever closing debate with respect to a motion to repeal the rule). The question then is whether the first option, for all its faults, is the best solution to the constitutional problem identified. I will turn to that in my next post.


Leave a Reply

Your email address will not be published. Required fields are marked *