«CONSTITUTIONAL AMENDMENTS AND THE CONSTITUTIONAL COMMON LAW Adrian Vermeule THE LAW SCHOOL THE UNIVERSITY OF CHICAGO September 2004 This paper can be ...»
PUBLIC LAW AND LEGAL THEORY WORKING PAPER NO. 73
CONSTITUTIONAL AMENDMENTS AND
THE CONSTITUTIONAL COMMON LAW
THE LAW SCHOOL
THE UNIVERSITY OF CHICAGO
September 2004 This paper can be downloaded without charge at the Public Law and Legal Theory Working Paper Series: http://www.law.uchicago.edu/academics/publiclaw/index.html and
The Social Science Research Network Electronic Paper Collection:
CONSTITUTIONAL AMENDMENTS AND THE CONSTITUTIONAL COMMON LAW
To structure the discussion, I advance two subsidiary theses about constitutional amendment.3 The first is that constitutional amendments can and do change constitutional law, including the law in action as well as the formal constitutional text. The second is that there is no good general reason to prefer common-law updating to the amendment * Professor of Law, The University of Chicago. Thanks to Beth Garrett, Jacob Levy, John Manning, Eric Posner, Geof Stone, Lior Strahilevitz, Cass Sunstein, and participants at a University of Chicago faculty workshop for helpful comments. Special thanks to David Strauss for providing both detailed comments and a model of collegiality in the face of disagreement. Justin Rubin and Carli Spina provided excellent research assistance, and the Russell J. Parsons Fund provided generous support.
See U.S. Const. Art. V (establishing rules for enacting formal amendments to the constitutional text).
I do not use “constitutional common law” in Henry Monaghan’s sense of constitutionally inspired doctrine that might be overridden by legislation. See Henry Paul Monaghan, The Supreme Court, 1974 Term--Foreword: Constitutional Common Law, 89 HARV. L. REV. 1 (1975). Rather, I use the term to denote constitutional rules, not defeasible by ordinary legislation, that are elaborated by judges through precedent-based reasoning. See generally David A. Strauss, Common Law Constitutional Interpretation, 63 U. CHI. L. REV. 877 (1996)[hereinafter Strauss, Common Law]. So constitutional common law is short for something like “judge-made constitutional law” or “common-law constitutional exegesis.” Moreover, Parts I and II equate constitutional common law with judge-made law, temporarily bracketing questions about whether constitutional change outside Article V is primarily driven by judges or by nonjudicial actors. Part III examines the issue in detail.
For simplicity, I shall reserve the term “amendment” to mean “formal constitutional amendment through the Article V process.” I shall describe other changes to constitutional rules, especially through judicial interpretation, as “constitutional common law,” “constitutional change” or “constitutional updating.” On the semantic problems surrounding “amendment,” see Sanford Levinson, How Many Times Has the United States Constitution Been Amended? (A)26; (B) 26; (C) 27; (D) 27: Accounting for Constitutional Change, in RESPONDING TO IMPERFECTION (Sanford Levinson, ed. 1995).
Constitutional Amendments process; to the contrary, an evenhanded institutional comparison suggests that each process shows to best advantage under particular conditions, and in particular domains.
These theses sound banal, but each has been denied in important recent work. As to the first thesis, David Strauss has argued that constitutional amendments are “irrelevant” because an amendment is neither a necessary nor a sufficient condition for legal change. The second thesis contradicts a standard academic view that I shall call the generic case against constitutional amendment. On this view, there are good general reasons to reject, or to indulge a presumption against, any proposed amendment. Among these reasons are the following claims: it is bad to “tamper” with the Constitution; the Constitution should not be “cluttered up” with amendments that will “trivialize” its majesty; constitutional amendments are “divisive” or “polarizing”; constitutional amendments may have bad unanticipated consequences; and constitutional amendments diminish the coherence of the constitutional text or of judicially-developed constitutional doctrine. Something like this view has become the conventional wisdom in the legal academy, following explicit arguments by Kathleen Sullivan and others.
My critical aims are to question the view that amendments are irrelevant, contra Strauss, and to question the widespread view that amendments are generically or presumptively suspect. The two positions I critique jointly privilege constitutional updating by judges, the first by suggesting that constitutional amendment is generically futile, the second by suggesting that constitutional amendment is generically harmful.
The implicit alternative is a practice that entrusts all constitutional change to commonlaw constitutionalism—that is, to an ongoing constitutional convention whose delegates are all judges (and hence all lawyers).
As against those views, I will attempt an evenhanded institutional comparison of the amendment process and common-law constitutionalism, as alternative means of constitutional updating. I consider the strengths and weaknesses of each process: relative to common-law constitutionalism, the amendment process is less focused on the facts of particular cases (both for good and ill); puts less weight on the views of past judges (both for good and ill); allows for the participation of decisionmakers from a broader range of professions and backgrounds (both for good and ill); produces more enduring
constitutional settlements, albeit at higher initial cost; and trades the benefits of flexibility for the benefits of rigidity. By identifying these variables, I hope to make some tentative progress towards identifying the empirical conditions under which, or the domain in which, either process is most likely to produce valuable constitutional change.
The discussion is organized as follows. Part I critiques Strauss’ thesis that constitutional amendments are systematically irrelevant. First, even if constitutional amendments are neither necessary nor sufficient to produce legal change, they may nonetheless be causally efficacious in producing legal change. Strauss has overlooked that the causal force of amendments is probabilistic: even if an amendment neither guarantees a desired legal change nor is indispensable to it, the amendment may nonetheless make the change more likely than it would have been in the amendment’s absence. Second, the irrelevance view slips imperceptibly from the denial that any particular amendment is relevant to the denial that the total set of amendments is relevant.
But if the reason that particular amendments are irrelevant is that other amendments would have been interpreted to produce the same effect—and this is the form of Strauss’ argument in many cases --- then the irrelevance claim cannot hold true of all amendments at once. Even if every particular constitutional amendment can be shown to be irrelevant through a seriatim procedure, we cannot generalize the conclusion that all constitutional amendments might be irrelevant simultaneously.
In Part II, the core of the paper, I argue that the generic case against constitutional amendment fails. The generic case rests on a nirvana fallacy that implicitly contrasts a jaundiced view of the amendment process with a romanticized view of constitutional common law. The real alternative to constitutional amendment is flexible judicial interpretation that updates the Constitution over time—a practice that can also be seen as tampering with or trivializing the Constitution, that is at least as polarizing or divisive as constitutional amendment, that equally risks bad unintended consequences, and so on.
Moreover, a public norm of the kind embodied in the generic case against amendment would produce either a suboptimal rate of constitutional amendments, or an optimal rate at excessive cost. The generic case can be reconstructed as a weak presumption, but in that form it loses its distinctive force.
Once we have dispelled the nirvana fallacy underlying the generic case against amendment, constitutional updating is seen to pose a comparative institutional question.
Part III compares constitutional amendment, on the one hand, and constitutional common law, on the other, as institutional alternatives for managing the inevitable updating of constitutional law over time. Under what circumstances might one process or the other prove superior? What institutional considerations, or variables, determine their relative performance? I suggest that amendments show to best advantage, relative to common-law constitutionalism, where the constitutional changes in question involve large value choices as opposed to technical improvements in the law, where constitutional change must be systemic and simultaneous rather than piecemeal, and where irreversible change is more valuable than reversible change. A brief conclusion follows.
I. The Relevance of Constitutional Amendments
The irrelevance thesis holds as follows:
[S]ubject to only a few qualifications, our system would look the same today if Article V of the Constitution had never been adopted and the Constitution contained no provision for formal amendment.... 4 The basic thesis is an extended counterfactual: in the absence of an amendment mechanism, the constitutional order would look roughly the same as it actually does today. (An important qualification to the thesis is that it does not apply to the original or “unamended” Constitution or to the Bill of Rights; I return to this domain-restriction on the thesis in I.B.). For want of historical expertise, I shall not attempt to engage the irrelevance thesis to the extent that it happens to incorporate historical claims about how actual amendments actually worked out, rather than counterfactual claims about the “relevance” of the formal amendment process. I will instead confine myself to some comments on the conceptual foundations of the irrelevance thesis.
In Section A, I examine some problems that inhere in the counterfactual character of the thesis. Counterfactual claims are not the same as causal claims. The irrelevance thesis makes a general counterfactual claim, but the thesis must not be interpreted to make, nor can it support, a general causal claim. The critical point is that constitutional David A. Strauss, The Irrelevance of Constitutional Amendments, 114 HARV. LAW REV. 1457, 1459 (2001)[hereinafter Strauss, Irrelevance].
amendments may be causally efficacious in producing legal change, even if (as the irrelevance thesis claims) they are neither necessary nor sufficient for producing legal change. In Section B, I examine some conceptual problems that inhere in the application of the irrelevance thesis to the interaction between and among different constitutional provisions. Here the main burden of the discussion is that, in many cases, the thesis establishes the irrelevance of particular amendments only by pointing to other amendments which would otherwise have subsumed the functions of the irrelevant amendment. This mode of analysis might work for any particular amendment, but it cannot be generalized across a whole set of amendments simultaneously.
The overall conclusion is not that the irrelevance thesis is wrong. If its conceptual foundations are infirm, much of the structure erected on those foundations retains great value. Construed in its best light—as a series of retail-level historical claims about particular amendments—Strauss’ thesis makes an important contribution to constitutional legal history. But nothing more ambitious follows. In particular, we cannot, in my view, subscribe to any more general claim that amendments are intrinsically inefficacious.
A. Counterfactuals, Causation and Constitutional Amendment The irrelevance thesis, although striking, says less than the casual reader might assume. There is a fundamental ambiguity in the claim that the formal amendment process is “irrelevant.” Rightly understood, the thesis claims only that amendments are neither necessary nor sufficient for legal change; but the thesis is easily confused with a very different and far more ambitious claim, to the effect that amendments do not cause legal change. The slippage between these two claims gives the irrelevance thesis a hard rhetorical punch, but the first claim does not entail the second, or so I shall argue.
The subsidiary propositions of the irrelevance thesis are these:
[The irrelevance thesis may be proved by] establishing four propositions. First—a relatively familiar point—sometimes matters addressed by the Constitution change even though the text of the Constitution is unchanged. Second, and more dramatically, some constitutional changes occur even though amendments that would have brought about those very changes are explicitly rejected. Third, when amendments are adopted, they often do no more than ratify changes that have already taken place in society without the help of an amendment. The changes produce the amendment, rather than the other way around. Fourth, when amendments are adopted even though society has not changed, the amendments