Paradox of Self-Amendment by Peter Suber themselves" the way rhubarb washes out the digestive system and itself, the doctrine of desuetude resembles the skeptics' "proof that proofs are impossible" —even when it succeeds, it displays the continuing possibility of what it aimed to undo. Amendment by frequent, successful violation is a variant of desuetude. Desuetude is classically triggered by neglect or non-enforcement, both of which may be responses to frequent violation. Desuetude without violation is possible for rules whose original purpose or sphere of application fades away, making violation as obsolete as compliance. But where desuetude is recognized, a custom or practice contrary to the rule of law must usually have arisen,[Note 8] a custom that is itself a function of frequent violation. In this sense violation may become repeal even though the acts of violation do not and ordinarily must not have law-making effect.[Note 9] Amendment by violation is officially rejected at the constitutional level in the United States. State ex. inf. Barrett ex rel. Bradshaw v. Hedrick, 241 S.W. 402, 421 (Mo.Sup.Ct. 1922) (the legislature cannot amend the constitution merely by violating it frequently). Does desuetude apply in any way to the American state and federal constitutions? The negative answer of Hedrick is common. In Walz v. Tax Commission of the City of New York, 397 U.S. 664, 678 (1970) the Supreme Court asserted that It is obviously correct that no one acquires a vested or protected right in violation of the Constitution by long use, even when that span of time covers our entire national existence and indeed predates it. In Walz the "custom" was tax exemption for churches, which the court upheld after so dramatically declaring it open to question. Softening its bold position on desuetude, it cited the statement of Justice Holmes in Jackman v. Rosenbaum Co., 260 U.S. 22, 31 (1922): If a thing has been practiced for two hundred years by common consent, it will need a strong case for the Fourteenth Amendment to affect it. The Holmsean qualification almost reintroduces desuetude, like the Elliott case, in the guise of fairness or due process. However it is the bold, unqualified rejection of desuetude that has been most quoted from Walz.[Note 10] A similar principle often found is that "[c]hanged economic conditions or developments do not amend the Constitution." Burns v. Alderson, 51 Wash.2d 810, 322 P.2d 359, 361 (1958) (contains many citations to other cases). One may suspect that these unqualified rejections of desuetude are, like the exclusivity of the AC that they support, merely an official view repeated for form's sake and compromised in practice as needed. Desuetude may be conceived as the repeal of law through custom. One may expect to find it, at least unofficially, wherever custom is recognized as a source of law. Custom becomes a rule of change for constitutions primarily through the agency of judicial interpretation and reinterpretation, a quasi-official rule of change for constitutions with a strong claim to de facto validity (Section 15). Changes in practice, usage, custom, or popular values commonly affect the interpretation of the constitution, sometimes through judicial "activism" that incorporates such changes, and sometimes by judicial "restraint" that defers to legislative, executive, or administrative departures from firm rules or prior interpretations of the constitution.[Note 11] An example from our own discussion in Section 17 is the rise of executive agreements as alternatives to treaties that dispense with the consent of the Senate. The constitution does not recognize them, but they have become customary. An anonymous English scholar observes that[Note 12] By such and kindred devices our kin beyond the sea have ingeniously contrived to modify in various directions their much venerated Constitution which was fondly imagined by its authors to be immutable save by the orthodox method of formal amendment as prescribed in Article V. When we look for something more direct we face many difficulties. Is the convention method of amendment desuetudinal just because long non- use has created uncertainties that form practical and political barriers to its use? Until 1933 the same might have been said of the method of ratifying amendments in state conventions. Provided by the constitution from the beginning, it had never been used. But that method was dusted off and used for the first time, one might say with relish, in ratifying the Twenty-First Amendment, the repeal of Prohibition (see Section 16). There are other sections of the constitution to which we might look as clauses that may have lapsed through desuetude. One lies in §2 of the Fourteenth Amendment. It says that if a state improperly restricts the right to vote, then its basis of representation "shall" be reduced in the proportion which the number of victims bears to the number of eligible voters. Despite the strong language and frequent violations, this mandatory penalty has never been imposed. By one estimate, the British North America Act of 1867, which functions as the constitution of Canada, contains 50 sections that have "no remaining constitutional force although they have formal validity."[Note 13] One of the most litigated examples in our constitution is Article I, §9.7, which requires that "expenditures of all public money shall be published from time to time." This clause still lives for most public money, but the budget of the Central Intelligence Agency is notoriously, and officially, secret. Repeated attempts to compel publication of its budget under Article I, §9.7 have failed, not through any express appeal to desuetude, but through an impenetrable web of objections based (for the most part) on standing.[Note 14] Of course the official reasons for the secrecy of the C.I.A. budget do not include the desuetude of Article I, §9.7. They are all appeals in some way to national security. But there is no official national security exception to Article I, §9.7 (or to virtually any clause of the constitution, either), which is one reason why lack of standing must be invoked by a government anxious to preserve secrecy. Hence, while it appears that national security and standing problems protect the C.I.A. budget from 129