Paradox of Self-Amendment by Peter Suber consider it a meter (106). The rule of recognition, then, is law only by acceptance, it seems, and not also by self-application. Acceptance itself, however, is not law but only the authority for law. Whether acceptance is the proper or "authoritative" authority for law only because it is accepted, in self-application, that is, whether we should take a reflexive or irreflexive view of acceptance itself will be discussed below (Section 7.B). Sometimes Hart suggests that acceptance in the sense intended is performed only by the officials of the system (106, 107, 108, 111, 113, 114, 118), sometimes by ordinary citizens (114, 121, 142), and sometimes he insists that citizens and officials are both needed (59, 98). Once he says that the practice of a "social group" suffices (106). To the extent that citizen participation is necessary, we must distinguish citizen acceptance from obedience. Obedience is at least necessary (113, 197), and possibly sufficient (111). Citizens contribute to acceptance primarily through "acquiescence" while officials contribute primarily by explicit acknowledgement of rules as law (59-60), doing their jobs (59-60), appealing to the tacit rule of recognition (99), and by using the law from the "internal" standpoint (99). At the same time, acceptance defines the internal standpoint (86, 99, 135), and the internal standpoint defines acceptance. Both citizens and officials display their internal standpoint or acceptance by accepting not merely individual legal rules, but rules that commit one in advance to accept future rules (229, 230). Citizens display acceptance by "acting in a certain way" (136) and justifying their action by citing rules and criticizing departures (136). Hart sometimes speaks as if acceptance of a rule of recognition can be displayed by "the general operation of the system" (105). In any event, the acceptance given must be voluntary (197, 198), even if it is induced by attitudes that are indirect products of law itself (172, 196, 199). It is the voluntariness of acceptance that creates authority, not just another social fact (196). Acceptance of law is not morally obligatory (198). Certainly citizens need not consent to the criterion of legal validity embodied in the rule of recognition, nor need they even agree among themselves on what it is. To require anything as express or deliberate as consent or knowledge is the "opposite error" from Austin's requirement that citizens habitually obey (111). Acceptance by officials presupposes some understanding (59), but citizen acceptance or obedience must be conceived consistently with the "realistic" fact that citizens are normally "passive" participants in the legal system, and frequently do not even know what the law is (47, 60, 110-11, 114).[Note 2] This shows how far Hart's acceptance theory is from a classical consent theory of law, and shows his intent to explain legality, not legitimacy. Hart leaves somewhat unclear exactly what function he believes acceptance serves in a legal system. It may indicate that rules exist, that they are valid, that they are efficacious, or that they form a system. As noted, acceptance clearly makes the rule of recognition law, but not "valid" law. Acceptance at least indicates a rule's existence (57, 142); widespread disobedience can indicate non-acceptance and hence, non-existence (142). For ordinary rules, existence and validity are probably not to be distinguished under the acceptance theory, at least if the rules exist only in the practice or usage of the people (106). But the efficacy and the validity of rules are distinct, unless the rule of recognition includes a rule of desuetude (100). But normal inefficacy does not suggest lack of acceptance, and therefore does not imply invalidity (100-01). However, "general disregard" may imply invalidity regardless of the presence or absence of a rule of desuetude, because it may imply lack of acceptance (100-01, 142; cf. 114f). Similarly, validity does not imply efficacy, even if it implies general acceptance (100). Acceptance at least functions to unify a set of rules into a system (59, 90, 91, 93, 106, 113, 118, 228f), but the acceptance need not extend either to consensus among the members of the requisite group of people nor to consensus on all the rules and criteria of validity of the unified system (118- 19). Not only is unanimity unnecessary, but the acceptance may be tacit rather than express, inferred from usage and practice (99, 106, 107) and acquiescence (60). Some practices constitute explicit acceptance (59). To achieve its law-making or recognizing effect, practice need only be "general" (98) or "normally concordant" (107), and the standards "unified or shared" (111). However, disunity may reach a point at which the system becomes "pathological" and one may question whether it still exists or is still the same system (100, 114-20, 141). The fundamental role of acceptance, and the participation of citizens in it, does not imply for Hart that "the people" are sovereign, even in their capacity as electors in a democratic system (70-76). Sovereigns are constituted by rules that are validated by acceptance; they are not the makers of those rules (75, 218). But whether acceptance is the officials' or the people's, active or passive, tacit or express, attitudinal or behavioral, knowing or ignorant, widespread or pathologically scarce, it must be "effective" (113). Effectiveness requires majority acceptance but not unanimity (89, 226). The effectiveness of acceptance is a generally undefined, probably a conclusory, term in Hart's work. B. Hartian acceptance self-applied One of the weaknesses of the inference model is that it cannot deal with the circularity and inconsistency of some legal processes. It must declare them invalid even when accepted by the people and the courts. The acceptance theory can cope with the self-contradiction and circularity of self- amendment, but it eventually becomes enmeshed in a circle concerning its own authority to authorize rules and practices. Is acceptance the proper source of authority only because it is accepted as such? Here I will ignore weaknesses of the acceptance theory not relating to the self-application of acceptance (but see Section 21.D). 36