Paradox of Self-Amendment by Peter Suber establishment of a legal system, under the acceptance theory, requires a creative type of acceptance that will always resemble a social contract on many points. Article VII is self-justifying in the sense that it states the acceptance-terms of the federal contract, and is law because it was accepted or ratified by its own terms.[Note 10] Hart gives contradictory suggestions on the question whether a rule is needed to authorize the ultimate rule of recognition. His general support for what I have called the normative practice doctrine supports the existence of a rule justifying acceptance, insofar as acceptance is used in practice to identify the rule of recognition. On the other hand, Hart just as clearly believes that the rule of recognition is neither valid nor invalid (105), nor justified by a rule (104). When Hart is thinking on these lines we should assume that he would say the same of acceptance a fortiori. His irreflexive theory of treaties and contracts is suggestive in this light. Treaties and contracts do not bind just because the parties agree to be bound; they bind only if made within a legal system containing a rule that states that mutual promises bind (40, 219, 220, 231). If we take the liberty of applying this contract theory to the establishment of a new regime, either because of the contractual elements in it or on the evidence of Hart's general disinclination to cite reflexive sources of authority, then Hart would not allow a self-justifying act of legal creation. He would not allow a contract-like constitution that decreed that it would be valid if ratified in a certain way, and actually was so ratified, unless the act of ratification could be construed to take place in an already legal context in which a valid rule stated that constitutions so decreeing and so ratified will be valid. He would probably deny the self-justification of the constitution and rest its authority on acceptance which neither needed nor received any further authorization. By analogy to his contract theory, Hart might say that acceptance would not operate to confer legal status on a rule of recognition or a constitution unless manifested within a system containing a rule that acceptance should so govern. The absurdity of requiring new systems to be made only in larger, preexisting systems is more than just another way of denying a right to revolt; it also denies the legal capacity to recover from discontinuity. It may have its own merits but Hart could not tolerate it, for it implies that a revolutionary regime could never become legal, even after 200 years of acceptance by the officials and the people. Like the strictest inference models, it implies that every lawful regime requires an infinite genealogy. Hence, we should abandon Hart's irreflexive theory of contracts and treaties insofar as it leads to these consequences in conflict with the acceptance theory, which is more central to his work and more defensible on its own merits. If contracts derive their binding force from the parties's agreement, not an external rule that agreement binds, then, as noted earlier, the AC of a contract could be amended without paradox, for the authority of the new rule would derive from the sufficient agreement of the parties, not the old clause or an external rule. Hart is clearly inconsistent on the question of the existence of a rule authorizing acceptance to authorize law. I believe the acceptance model of authority is separable from both the normative practice doctrine and Hart's irreflexive view of contracts. In fact I believe the most sensible way to render Hart consistent is to abandon his theory of contracts and allow agreements to bind ex proporio vigore, independently of any other legal rules. The self-justification involved is harmless as well as theoretically desirable to explain the initial establishment of legal systems or the "legalization" of a system that broke off from another. The normative practice doctrine does not affect this question significantly. If retained, it leads Hart to a circularity which he did not intend but which is harmless and in any case preferable to the alternatives; if abandoned, acceptance authorizes law as a matter of fact, not because it has a prior or higher justification for doing so. The views that acceptance is self-authorized and that it is beyond authorizing are equally acceptable as solutions to the paradox of self- amendment. Like agreements to contracts, acceptance of law is sufficient by itself, either because it needs no justification or because it justifies itself, to give authority to legal rules. Like agreements to contracts, acceptance of law is an anchor to legal rules in historical actuality; it avoids the need for an inference model that forbids self-justification and requires incomplete justification or infinite regress. Notes 1. H.L.A. Hart, The Concept of Law, Oxford University Press, 1961. All subsequent references to this work in this section will be kept in the text in parentheses. 2. My argument in Section 2 that the people's failure to use their AC could show their consent to be governed by their constitution, provided the "consent" not to use the AC was competent by some cognitive and volitional standards, obviously required some citizen understanding of their legal system. Therefore it is consistent with Hart's statement, op. cit., at p. 76: Failure to exercise an amending power as complex in its manner of exercise as that in the United States constitution, may be a poor sign of the wishes of the electorate, though often a reliable sign of its ignorance and indifference. If the people's role in acceptance must be compatible with their ignorance or indifference, much of the validating burden of acceptance is shifted to the role of the officials. But Hart has been criticized on this point for failing to clarify the relation between the contributions of the people and the officials of the system. See e.g. Hugh Williamson, "Some Implications of Acceptance of Law As A Rule-Structure," Adelaide Law Review, 3 (1967) 18-45 at p. 28; and Gabriel Mosonyi, "Legal Obligation, Social Acceptance and the Separation of Law and Morals," Connecticut Law Review 6 (1973) 36-48, at 37.n.14. If a reflexive theory of legal authority that permits some kinds of self-justification is rejected for its potential to validate tyranny, then the same potential in Hart's irreflexive blessing of ignorance should not be overlooked. 39