Paradox of Self-Amendment by Peter Suber disclosure, we might say instead that Article I, §9.7 has been amended by custom to permit an exception for national security. The only obstacle to this reading of the facts, however, is that the C.I.A. budget has repeatedly been the target of civil action seeking disclosure. The rule forcing disclosure has not been beyond the reach of plaintiffs on account of obsolescence so much as, on the contrary, vigilant maneuvering. One more possible application of desuetude to constitutions may be noted. If a constitution or amendment is adopted in violation of a minor but clear rule of procedure, then many states and nations will allow it to become valid by acquiescence.[Note 15] But cure by acquiescence is a form of desuetude applied to the rule that had been violated. If the rule does not completely lose its validity as time passes and as a contrary custom arises, at least it loses its authority over a major violation.[Note 16] One may interpret cure by acquiescence in other ways, but the similarity to desuetude will not be erased. For example, my own preferred interpretation is that cure by acquiescence represents the triumph of social acceptance over the formal rules that would invalidate the constitution or amendment. Insofar as the violated rule is valid by the rule of recognition, acceptance overrules (or amends) the latter in order to invalidate the former. But any time acceptance overrules (or amends) the rule of recognition in order to invalidate a law, the "unacceptable" has lapsed according to a process with an ineradicable resemblance to desuetude. Nothing could be more foreign to legal formalism, or a Rossian inference model, than to suppose that either neglect or violation could amend any legal rule. By their favorite analogies, formalists would find desuetude as sensible as supposing that habitual bad reasoning could amend the principle of non-contradiction.[Note 17] Acceptance theorists like Hart, on the contrary, offer the readiest explanation for desuetude and amendment by violation. No rule is of present legal validity, under the acceptance theory, if it is not presently accepted as legally valid, either directly or through an accepted rule of recognition. A sophisticated acceptance theory must account for the validity of laws that have not been used for some time and which are therefore out of the actual consciousness of most citizens and officials of the system. No acceptance theory, however, need make this task more difficult than it is by equating acceptance with the contents of the actual consciousness of a bloc of citizens or of the constructive fiction of the collective consciousness. The presently accepted rule of recognition, not overruled on this particular by acceptance, may even deny desuetude by ascribing continuing validity to any statute or constitutional rule enacted in the proper way but not yet repealed in any of the official or non-desuetudinal ways. I suggest that the transition in America from an era in which desuetude was recognized to the present era in which it is not is due, not to desuetude itself, but to a transformation in our legal thinking from a variant of the acceptance theory in which popular sovereignty was a more concrete and compelling idea, to a more rigid formalism in which law became official, "scientific", "professional", and abstractly conceived as self-entrenched or amendable only by its own official means.[Note 18] Notes 1. Linda and William Rogers, "Desuetude as a Defense," Iowa Law Review, 52 (1966) 1-30; Arthur E. Bonfield, "The Abrogation of Penal Statutes by Nonenforcement," Iowa Law Review, 49 (1964) 389-440; John Chipman Gray, "Desuetude of Statutes in the United States," in his The Nature and Sources of Law, Macmillan, 2d. ed., 1921, pp. 329-34. 2. See J.R. Philip, "Some Reflections on Desuetude," Juridical Review, 43 (1931) 260-67. 3. For a detailed comparison of desuetude to different types of prosecutorial policies of non-enforcement, and a persuasive argument that each may justify an act of "violation" or rule-departure by citizens, see Mortimer and Sanford Kadish, Discretion to Disobey: A Study of Lawful Departures From Legal Rules, Stanford University Press, 1973, pp. 127-40. 4. The statute, 18 U.S.C. 956, was enacted in 1917. It prohibited conspiracies within the United States to destroy property in foreign nations with which the United States is at peace. 5. Lester Bernhardt Orfield, The Amending of the Federal Constitution, University of Michigan Press, 1942, at p. 81. Orfield asks whether the Fifteenth Amendment has lapsed through desuetude and non-enforcement, but eventually rejects the idea. Without using the word "desuetude", Arthur Machen believes that the Fifteenth Amendment is void (inter alia) because it has never been enforced. Arthur Machen Jr., "Is the Fifteenth Amendment Void?" Harvard Law Review, 23 (1910) 169-93, at p. 169. When Machen was writing, 40 years had elapsed since the amendment's adoption, including the reconstruction following the Civil War; and the Fifteenth Amendment had never once been used to invalidate a state or federal statute. However, as Orfield notes, two Supreme Court cases enforcing the Fifteenth Amendment occurred shortly after Machen wrote, viz. Guinn v. U.S., 238 U.S. 347 (1915), and Myers v. Anderson, 238 U.S. 368 (1915). Other aspects of Machen's thesis are discussed in Section 8.C. In addition, one may well decide that a constitutional provision is "used" and even "enforced" in many other situations than when it is used to invalidate a conflicting statute. 6. But see Section 12.C and accompanying notes 29 and 30. 7. Sir Carleton Kemp Allen, Law in the Making, Oxford University Press, 7th ed. 1964, at p. 478.n.2 (emphasis in original). 8. Philip, op. cit., at p. 261. 9. See Section 9.C. I explore the logic of amendment through violation in more detail, and in a different domain, in "The Reflexivity of Change: The Case of Language Norms," Journal of Speculative Philosophy, 3, 2 (1989) 100-129. 130