Paradox of Self-Amendment by Peter Suber Preface to the Printed Version Why bother? Logical paradoxes in the strict sense produce statements like those of the Liar ("This very statement is false") that are false if true, and true if false. They resist rational solution or at least divide logicians for centuries of apparently irreconcilable wrangling. What happens when similar paradoxes arise in law? That is a difficult question, but part of the answer is that paradoxes come and go without much notice and are dealt with without much ado. This fact makes the question important as well as difficult. How law copes with strict paradox sheds light on the nature of legal reasoning and rationality, the nature of legal practicality, and the sense in which law can be reasonable, even "wise", while being illogical in the technical sense. I select one principal paradox —the paradox of self-amendment— and explore its variations in several Anglo-American jurisdictions and contexts, but mostly in American constitutional law. If a constitution has an amendment clause (a provision describing or prescribing how to amend that constitution), then can that clause be used to amend itself? The question may be widened to embrace the self-application of any legal rule that authorizes any legal change. Is self-amendment paradoxical? If so, can it be lawful? If so, can the logic of law be logical? Some legal rules govern the change of other legal rules. But even these "rules of change" (as Hart calls them) are changeable, usually by higher level rules of change. When a rule of change is supreme within its own system (as a constitutional rule of change probably is), then it is changeable, if at all, only under its very own authority. The paradox of self-amendment arises when a rule is used as the authority for its own amendment. It is sharper when the rule of change is supreme, sharper still when it is changed into a form that is inconsistent with its original form, and sharpest of all when the change purports to be irrevocable. Alf Ross has charged that self-amendment in this sense can be reduced to a formal self-contradiction. After considering some obvious and not so obvious objections and subterfuges, I conclude that this is correct. Moreover, Alf Ross and his followers have decided that this contradiction invalidates self-amendment. In one sense this book is an extended meditation on Ross's confidence that what is logically impossible must be legally impermissible. I conclude that this view is factually incorrect and philosophically arrogant. One task that earlier scholars have not undertaken is to show that self-amendment has frequently occurred despite all contradiction and paradox. It is commonplace and, in fact, not even controversial in legal circles. By all legal tests of lawfulness, self-amendment in the most illogical sense is lawful. Nor have past explorers of these themes asked what law, legal change, legal reason, and legal rules are such that they can tolerate this kind of contradiction while forbidding many other things in the name of consistency. The principle that what is logically impossible must be legally impossible may be philosophically arrogant and ignorant of legal history, but it is not a simple mistake. It is a new variation on the theme of natural law. Instead of finding that human law depends for its validity on an eternal moral law, this version makes it depend on an eternal logical law. One of the most persistent and persuasive objections to traditional natural law theory applies as well to the new version. If human law can be immoral without ceasing to be law, it seems it can be illogical without ceasing to be law. Law has its own tests of what is law, and those tests validate much that is immoral and illogical. To decide that a transcendent moral or logical test supersedes the legal tests, and can invalidate what is otherwise law and validate or enact what is otherwise utterly tacit, is to transform law into morality or logic and unduly diminish its historical and social dimensions. It is to assert that law reflects human thought and community only at their best and never also at their worst. It is to assert that somehow this human construction has had an immaculate conception, and is never touched by human unreasonableness, historical contingency, and interest. The present work is in part a critique of the new natural law theory that replaces the traditional moral test with a logical one. I conclude that logic does not legislate a priori for law any more than morality does, although of course law remains perpetually open to logical and moral criticism. To know what the laws are in a society, and what is lawful there, we must look only at legal evidence and social phenomena. We need not consult any moral or logical doctrine, even any that happen to be affirmed unanimously, except to the extent that they influence the concrete legal practices of that society. In particular, I conclude that self-amendment may well be a strict self-contradiction, but the evidence of legal history suffices to show its lawfulness even in the face of logical doctrine. I argue that the only alegal source of legal authority is social practice, not normative principles from morality or logic. To philosophers of Alf Ross's persuasion it may appear that I am begging the question. By rejecting the a priori principle that (standard) formal logic is the higher law of law, and undertaking a more empirical investigation of the relations of formal logic and law, I have already erred. I argue, however, on the contrary, that the a priori principle that (standard) formal logic governs law begs the question of the nature of law. Inquiry into legal practice, not presumptions in advance of inquiry, should answer the question. Between these two positions it seems that question begging is mutual and unavoidable. If so, then arguments across the gap of this incommensurability are futile. This suggests that cordial suspension of disbelief is the duty of scholarship, not hasty dismissal. This at least has been my attitude toward Alf Ross. In any case, my lapse from the very common philosophical confidence in the dominion of formal logic is motivated by evidence; it is not just another irrational failure to recognize the sovereignty of formalism that others assume a priori. In short, I invite readers who know more logic than law to attend to the peculiarities of law that fly off like sparks from virtually every case and especially from paradoxes like that of self-amendment. iii