Hart's Concept of Law Lecture Three 2001 [OHP 3]

I turn now to an all too brief resume of Austin's Command Theory of Law. It may be simply stated in his own words: 39. "Every positive law ... is set, directly or circuitously, by a sovereign individual or body, to a member or members of the independent political society wherein its author is supreme ... The party or parties to whom a law is set, or the party or parties on whom a duty is laid, are necessarily obnoxious to the sanction which enforces the law and the duty ... It is set by a party armed with might, to a party or parties to whom that might can reach". (John Austin, op.cit., p.350). Austin's first pre-occupation was to differentiate "law" in the juristically relevant sense from other usages which are not or only indirectly the concern of the jurist. Austin regarded law as the command of the sovereign; positive law is a general rule of conduct laid down to a political inferior. A command requires a determinate body or individual to issue it and the threat of an evil in case it is not obeyed.

Austin realised that "law" in the juristically relevant sense might be confused with

1. Laws of God ("God's law"

2. Rules of Positive Morality ("the moral law")

3. Rules of etiquette, games, etc ("the laws of chess")

4. Empirical Generalizations, the so-called laws of nature and of physics ("the law of gravity" "McCavity, McCavity, there'e no one like McCavity/ He's broken every human law; He breaks the law of gravity")

The notion of "command" excludes this last. The law of gravity is not commanded. The idea of a determinate superior excludes the rules of positive morality and of etiquette; or, where one can point to a moral rule being posited by a determinate person it is still excluded as not being a command; it does not involve threat. The so-called laws of God are not positive; they do not issue from a human source. Positive law is human law; law made by men and women; posited by men and women for men and women, and not created by God or Nature.

For Austin ... 40. "Every law ... is a command". (John Austin, op.cit., p.13). Thus the concept of "command" is the "key to the science of jurisprudence". As the key concept it must be explained : 36 "A command, then, is a signification of desire. But a command is distinguished from other significations of desire by this peculiarity: that the party to whom it is directed is liable to evil from the other, in case he comply not with the desire". (John Austin, op.cit., p.14). This leads directly to Austin's conception of duty: 41. "Being liable to an evil from you if I comply not with a wish which you signify, I am bound or obliged by your command, or I lie under a duty to obey it". (John Austin, op.cit., p.14). Command and duty are therefore correlative terms. Thus wherever a duty lies, a command has been signified; wherever a command has been signified, a duty is imposed. Indeed, for Austin, Command, Duty, and Sanction are inseparable terms ... 42 "... the evil which will probably be incurred in case a command be disobeyed or in case a duty be broken is called a sanction". (John Austin, op.cit., p.15). For Austin, sanctions are necessarily evils and though reward is a motive to action, it is not a sanction.

Advantages of Austin's Conception of Duty:-

1. Demystification: we are not faced with transcendental entities; or irrefutable intuitions. Duty is not her logically tied to conscience, or moral sense, or human happiness , or benefit. This conception of duty is explained solely in terms of facts. This demystification is, of course, highly desirable since the concept of duty, and of legal duty is frequently surrounded by mystery and obscurity. Yet there may be other ways of demystifying the concept which do not run counter to the logical gap between "is" and "ought".

2. Generality: the sanction theory of duty seems to provide a truly general theory of duty. The same account can be given for legal, moral, and religious duties - the difference merely being in their sources, i.e. (1) sovereign; (2) men and women, not as political superiors; (3) god. As against which it is by no means self-evidently true that the various forms of duty can be or must be analysed in terms of one common, central aspect. It may be that the different forms of duty have nothing in common and that a general theory of duty therefore distorts.

3. Sanctions: the important connection between duty and sanction is emphasised. However one analyses "duty", the idea of sanctions is closely related ... there clearly is some relationship between duty and sanction. None the less it need not be the case that "duty" simply means "the possibility of a sanction".

4 Systemic import: this analysis does provide a basis for the explanation of other fundamental legal concepts. Thus we start with "command" as the key concept; this relates to "duty" which connects up with "sanction" and the concept of "punishment". Further duty implies "right", "liberty", "claim", privilege", "immunity" etc. Thus from the sanction theory of duty flows the whole gamut of fundamental legal concepts - what Austin would call the concepts of General Jurisprudence. Thus the conception of duty adopted by Austin provides a systemic basis for the explanation of legal concepts.

The following points can be made against the identification of duty and the possibility of a sanction:

1. The likelihood of suffering a sanction is not a sufficient condition of an act being a duty. Take the case of vicarious liability. I can be liable to a sanction without having a duty. If there is a duty in such a case it has a different content and refers to a different person than the act which is the condition of a sanction. Thus one person may be liable to a sanction while the duty lies on another (e.g. road hauliers, publicans).

2. The likelihood of suffering a sanction is not a necessary condition of an act being a duty. Given the certain knowledge that a man has 24 hours to live, it would follow on this account that he had no duty not to kill, since there is zero probability of his being visited with an evil for his wrongful killing before he ceases to exist. Yet one might still want to say that he has a duty not to kill.

3. Further we can conjoin statements about the likelihood of a sanction with statements about the obligatoriness of the act without any sense of redundancy or tautology:- e.g. "If you steal you are almost certain to be punished and, furthermore, there is an obligation not to steal. If "almost certain to be punished" actually meant "there is an obligation" the above statement would be tautologous - a pleonasm merely adding to the Left Hand Side a Right Hand Side restating in different words that which has already been said. The type of linguistic redundancy involved is obvious from the following example: "You had better not start dating him because he is married and, furthermore, he is not a bachelor" ! This seems to show that "Duty" does not mean "Probability of a sanction"

4 Indeed, to assert an action to be a "duty" is not to assert that anything will probably or certainly happen in the future. To assert an act to be a duty is to provide a reason for the imposition of a sanction; it is not to assert that something will occur. "Duty" involves justification, not prediction and "justification" involves relating behaviour to a rule or norm. The behaviour which is a duty is that which ought to be according to a rule or norm and not that which will be according to a prediction. This, the most telling criticism of Austin's concept reveals that duty is a normative concept whereas the imposition of a sanction is a factual circumstance. But one cannot derive an "ought" from an "is"; or a norm from a fact. An "ought" cannot have the same meaning as an "is" and therefore "duty" cannot mean the probability or the actuality of the imposition of an evil [Louvre OHP photos]

Austin's concept of law rests upon his notion of the sovereign. The sovereign is characterised by two criteria: the one "negative" and the other, "positive": 43. "If a determinate human superior, not in a habit of obedience to a like superior, receive habitual obedience from the bulk of a given society, that determinate superior is sovereign in that society, and the society (including the superior) is a society political and independent". (John Austin, op.cit., p.194). And, again, 44. "... the power of a monarch properly so called or the power of a sovereign number in its collegiate and sovereign capacity, is incapable of legal limitation". (John Austin, op.cit., p.254). And compare with Tom Paine's remark: 45 "... governments arise either out of the people or over the people. The English Government is one of those which arose out of a conquest, and not out of society, and consequently it arose over the people; and though it has been much modified from the opportunity of circumstances since the time of William the Conqueror, the country has never yet regenerated itself, and is therefore without a constitution" (Rights of Man, p 94).

The concept of "independent political society" which is closely tied up with sovereignty is an assumption of Austin's which he does not analyse. Further, the classification of "political superiors" and "political inferiors" without reference to the law is either metaphysical, relying upon some such idea as the divine right of some to rule or some notion of "natural rulers"; or it rests upon the idea of power in its crude physical sense. For Austin law is the command either of a specially qualified group or individual; qualified by virtue of metaphysical factors, or by crude power. This seems highly anti-democratic; it seems that Austin's account is elitist or is the voice of power; the command of a gang of robbers which many might think of as the very negation of law. 46 "The right of a Parliament is only a right in trust, a right by delegation, and that but from a very small part of the Nation; and one of its Houses has not even this. But the right of the Nation is an original right ..." (Paine, op cit p 142)

Although Austin sought a general definition of law in terms of the direct or circuitous commands of the sovereign, he was forced to acknowledge a number of exceptions to his thesis. HANDOUT ("technical stuff"): Exceptions to the thesis that Laws are Commands

I. Laws which are not commands (conceded, perhaps too easily, by Austin)

(1) Declaratory Laws

(2) Laws Repealing Laws

(3) Imperfect Laws

(a) Roman Jurists' sense of "laws without sanctions"

(b) Rules of Positive Morality

II. Laws which may seem not imperative (not conceded by Austin)

(1) Laws which merely create rights

(2) Custom

III. Express exclusions (Rightly conceded by Austin)

(1) Constitutional Law

(2) International Law

ad I (3) Coercion? Laws without sanctions?

Roman Jurists' Imperfect Law, e.g. "Steal Not"; Austinian Law: "If you steal you will be punished".

For Austin, (a) is not a law unless the courts and executive organs infer a sanction and apply it. But that simply means that an Imperfect Law is an incomplete law in the sense that a sanction has not yet been annexed to it. But that means that it is only a part of a complete (Austinian) law; it is a fragment of a law, not a complete law. But not being a law it cannot be an example of a law which is not a command, and therefore it cannot be a counter-instance to Austin's thesis. As to (b) Rules of Positive Morality are never Laws and therefore are not exceptions to Austin's thesis either.

ad I (1) Declaratory Laws - Commands?

A so-called declaratory law; that is, a "law" to explain existing law (such as an Interpretation Act) is admitted by Austin as an exception to his thesis. But this may be too swift since it can readily be shown than an interpretation section or even parts of the Interpretation Act, are parts of the very meaning of other laws. Thus the import of the "declaratory law" can be included as an element in the content of the pre-existing law. Declaratory Laws are thus Fragments of Laws and not complete laws. They are not, therefore, counter instances to Austin's thesis. (Note that "declaratory" is a misleading term and may be used as a cloak for law-making activity, where new law is created. In England, "The Declaratory Theory of the Common Law" holds that common law judge merely declare pre-existing common law and do not make it, whereas in Scotland the "Declaratory Power of the High Court" refers not to a declaration of what the law already is, but to an actual law-making power, a power to declare new offences).

ad I (2) Laws Repealing Laws are slightly problematic: they cannot be parts of something which no longer exists. Austin styles such laws "permissions"; but since he also holds that all permissive laws are at bottom imperative, such laws cannot be counter-instances to his thesis. One can only be "permitted" or be "free" if others are "prohibited" or "bound"; every person's legal freedom is another person's legal subjection - rights and duties are correlatives. Thus any law which "permits", indirectly "commands"; Thus a Law Repealing Law permits and therefore indirectly commands. Thus a law Repealing Law is not an exception to Austin's thesis: The negation of a command must also be a command.

Austin seeks to show that II(1) and II(2) are not really exceptions to his thesis;

ad II (1) Austin argues that rights and duties are correlative and that it is therefore impossible to create rights without imposing duties. Thus there can be no laws merely creating rights. Thus II (1) is not a counter-instance.

ad II (2) Austin assimilates Custom to Command by holding that, although Custom as such is positive morality at inception, when it is "adopted" or "recognized" by the courts it becomes the circuitous command of the Sovereign, under the rubric "What the Sovereign permits, he commands". Thus for Austin, all judge-made law is the circuitous command of the sovereign, and therefore no exception to his thesis.

ad III Austin allows, as must we, that International Law and Constitutional Law are exceptions to his thesis.

[N.B. Much of the argument turns upon the obscure notion of "a complete law". But what is a complete law? What is an individual law?]

For Austin [II(1)] there can be no laws merely creating rights. But we must ask: Do some laws merely create rights? Hart distinguishes duty-imposing and power-conferring rules. The latter do not impose duties or confer rights directly; they provide facilities. His example is a rule about witnessing a will. It does not impose duties, it merely provides a standard or legally proper way of doing things. So, if we go along with Hart, then Austin's glib answer to this question is unacceptable. But, as ever, there are possible counter arguments:

(1) Hart's power-conferring rules pre-suppose duty-imposing rules and are parts of them;

(2) Power-conferring rules create duties indirectly.

Customs are sometimes represented as binding legally because the citizens have observed them "from time immemorial", as the saying goes, and they cannot themselves be creations of the sovereign. However, as Austin claims, and as classical English constitutional theory maintains the sovereign can abolish them. If not abolished by the sovereign they subsist as positive law by spontaneous adoption or "reception" by the sovereign. They are not established by a political superior. Therefore they are not commands, and so are not laws properly so called. Custom is positive morality merely. Yet Austin holds too that customary law is imperative in the proper sense; and that all judge made law is the creature of the sovereign. Custom is mere positive morality at inception; i.e. originally its only force derives from the general disapprobation falling upon those who transgress it. Custom is transmuted into positive law when it is adopted as such by the courts. When judges adopt a customary rule and when they make a rule not suggested by custom, the legal rule which is established is that of the sovereign legislator. A subordinate or subject judge is merely a minister - an agent, or representative of the sovereign; the sovereign power is delegated. The rules the judges make derive their legal force from the power of the sovereign, conferred expressly or acquiesced in by the sovereign. But all this seems the baldest of fictions - like that according to which the King is always present in the Court. 47. "By many of the admirers of customary laws ... they are thought to oblige legally (independently of sovereign or state), because the citizens or subjects have observed or kept them. Agreeably to this opinion, they are not the creatures of the sovereign or state...". (John Austin, op.cit., p.30). 48."The admirers of customary law love to trick out their idol with mysterious and imposing attributes. But to those who can see the difference between positive law and morality, there is nothing of mystery about it. Considered as rules of positive morality customary laws arise from the consent of the governed, and not from the position or establishment of political superiors. But, considered as moral rules turned into positive laws, customary laws are established by the state: established by the state directly, when the customs are promulgated in its statutes; established by the state circuitously when the customs are adopted by its tribunals". (John Austin, op.cit., p.32). 49. "Customary laws are positive laws fashioned by judicial legislation upon pre-existing customs. Now, till they become the grounds of judicial decisions upon cases, and are clothed with legal sanctions by the sovereign one or number, the customs are merely rules set by opinions of the governed, and sanctioned or enforced morally: Though, when they become the reasons of judicial decisions upon cases, and are clothed with legal sanctions by the sovereign one or number, the customs are rules of positive law as well as of positive morality. But, because the customs were observed by the governed before they were clothed with sanctions by the sanction one or number, it is fancied that customary laws exist as positive laws by the institution of the private persons with whom the customs originated". (John Austin, op.cit., pp.163-164).

The implication is that the sovereign could reverse the rules made by judges and so these rules are his because he permits them. But remember Prohibitions del Roy 1607 12 Co Rep 603 in which it was decided by all the common law judges, headed by Coke that the right of the King to administer justice no longer existed. Coke declared, 50 "... that the King in his own person cannot adjudge any case, either criminal, as treason, felony etc., or betwixt party and party, concerning his inheritance, chattels or goods, etc. but this ought to be determined and adjudged in some Court of Justice, according to the law and custom of England". Look to the Calder v Bull handout page for tywo views of "sovereignty" or "legislative supremacy".

Commands may be express or tacit - where customs are turned into legal rules by decisions of subject-judges the legal rules which emerge from the customs are tacit commands of the Sovereign; i.e. "What the Sovereign permits, the Sovereign commands". Bear in mind the distinction between positive law and positive morality and Austin's account of custom will present no problems of comprehension although one may or may not agree with him) As positive morality, custom arises from the acceptance of the subjects, not from the position of the sovereign. These moral rules are established as laws by the activity of the sovereign's agents.

Austin's objective in all this is to show that customary law is "imperative". He is not claiming that "judicial law" is the same as statute law. Nor does he allow the judge to hide behind the legislature: 51. "[There is a view that] to impute judge-made law to the sovereign, to suppose judges speak with the sovereign's will is one of foolish or knavish fictions with which lawyers darken the clearest truth ... [This is] groundless ... all judge made law is the creature of the sovereign or state". (John Austin, op cit p 31). Thus judges are permitted or authorised to make law by the sovereign. Their office is by virtue of the sovereign will. That is why their decisions are law; but they are to be held responsible for what law they make, at least in so far as the Sovereign does not try to constrain judicial decisions by prior rules. Thus for Austin, as for positivism in general judges are to be understood as enjoying or suffering strong discretion.

There are a number of different assessments of the command theory of law depending on whether one accepts Austin's arguments relating to "laws which seem not imperative" and "custom", or the counter arguments thereto. Austin's case can be strengthened as I have suggested, but if these arguments are not accepted, the command theory is less comprehensive and more vulnerable to criticism. Depending on one's assessment of the command theory a "fresh start" - as Hart calls it - will be felt to be more or less necessary. One, rather positive, assessment is that 52. "Austin by establishing the distinction between law and morals.... laid the foundation for a science of law". (William Markby, Elements of Law, 1905, para. 12). A rather less appreciative view is implicit in the following observation: 53. "What he [Austin] really had in mind was contemporary English law and primarily an English criminal statute". (Buckland, Some Reflections on Jurisprudence, p.4).

6. Chapter III: Variety, Nullity, Fragmentation In Chapter III, under the rubric "the Variety of Laws", Hart unleashes a full-scale attack on the Austinian position built up in the previous chapter out of the gunman situation. Hart asserts that not all "laws" order people to do things. [OHP 4] There are four main classes of phenomena which do not square with Austin's position, viz.:-

1. Power-conferring rules or laws giving powers to officials; e.g. judges to try cases etc., and powers to individuals to make wills, contracts etc.; i.e. public and private power-conferring rules.

2. Not all "laws" are enacted, nor are they all the expression of someone's desire, e.g. Custom

3. Not all "laws" are directed to others; self-binding laws seems excluded by Austin.

4. Nor is all law the expressions of the legislator's desire; because often the legislature (i.e. those voting) do not know the content of the Bill and therefore cannot actually desire it or will it.

It is important to notice that Hart is not here concerned with the relatively trivial exceptions to the imperative theory which Austin admitted. Hart is arguing that a whole area of the law - specifically power-conferring laws, are simply not accounted for by the Austinian position. In order to accommodate these, Hart believes that we have to modify the Austin's theory out of all recognition. Hart therefore deals with various defences which Austin's apologists might offer. His claim is, of course, that these defences are inadequate. However one weakness here is that Hart tends to run together the Austinian and the Kelsenian positions and considers or gives us to believe that what counts against Austin counts also against Kelsen. I hope to convince you that while Hart succeeds in his attack on the Austinian position, he fails to make out an adequate case against Kelsen.

One point which should be stressed is that Hart asserts, but does not show, that there are laws of different types. I grant that Hart distinguishes uses or functions of law which are different, and indeed different in important ways, but I do not accept that we therefore have a case of "different kinds" [p 26] or "various sorts" [p 5] or "different types" [p 8] or "classes" [p 27] of law and certainly not different logical types; i.e there is a different sociological or functional difference but not a difference of logical form.

Hart, however, classifies law according to its "social function" [p 27]: 54. "... the social function [of criminal law]... setting up and defining certain kinds of conduct, as something to be avoided or done by those to whom it applies irrespective of their wishes". (Hart, op.cit., p.27). Thus, on the basis of social function, Hart admits that thee is ... 55. "... some analogy... between .... general orders and the law of torts". (ibid). Yet, says Hart, 56. "... there are important classes of law where this analogy with orders backed by threats altogether fails, since they perform a quite different social function". (ibid).

Chief among such laws are those defining the ways in which valid wills or contracts or marriages are made, which do not require persons to act in certain ways whether they wish it or not. Such laws do not impose duties; they ... 57. "... provide individuals with facilities for realizing their wishes, by conferring legal powers upon them to create, by certain specified procedures and subject to certain conditions, structures of rights and duties within the coercive framework of the law". (ibid). The power of individuals to mould their legal relations [58. "The power thus conferred ]... is one of the great contributions of law to social life". (Hart, op.cit., p.28). And this very important characteristic of law is obscured by [mis]representing all laws as orders backed by threats. 59. "The radical difference in function between laws that confer such powers and the criminal statute is reflected in much of our normal way of speaking about this class of laws". (ibid).

This argument is strengthened by reference to laws which confer "public" as against "private" powers; i.e. rules of jurisdiction, of judicial appointment, and of judicial procedure. Such rules, too, are very different from orders backed by threats. Further, rules defining the competence of subordinate legislatures cannot be fitted into the Austinian model: 60 "... there is a radical difference between rules conferring and defining the manner of exercise of legislative powers and the rules of criminal law which at least resemble orders backed by threats ... it would be grotesque to assimilate these two broad types of rule" (Hart, op.cit., p.31). This last point may be a bit unfair as against Austin who specifically excluded constitutional law from his theory. None the less a general theory of law should be able to accommodate such obvious instances of "law".

Hart hints that a much fuller taxonomy than his D-rules and P-rules dichotomy is possible once we are freed from ... 61 "... the prejudice that all must be reducible to a single simple type ...[n]ever the less the itch for uniformity in jurisprudence is strong: and ... by no means disreputable ..." (Hart, op.cit., p.32). It is hard to see why Hart stigmatises the "itch for uniformity" [my itch] as "a prejudice" if it is "by no means [intellectually] disreputable" In resisting the itch for uniformity Hart is opposing a dominant and recurring trend in the history of jurisprudence and he therefore turns attention to two theories sponsored by great jurists because they contain an element of truth, in particular 62 "... it is important to realise that rules of the power-conferring sort, though different from rules which impose duties and so have some analogy to orders backed by threats are always related to such rules: for the powers which they confer are powers to make general rules of the latter sort or to impose duties on particular persons who would otherwise not be subject to them". (Hart, op.cit., p.32-33). The question is: Can all laws be assimilated to "Orders backed by threats"? [Handout] Two arguments or theories for the fundamental uniformity of all legal rules are:-

I Nullity as a Sanction - adopted but not developed by Austin in lecture 28

II Power-conferring rules as fragments of laws - Bentham, Ross, and Kelsen. Hart needs to refute both lines of argument to succeed in his criticism of Austin's theory of law. If law is a union of different types of rules, I and II must be false. Hart argues:- ad I (a) Nullity is not a Sanction because nullity is not an evil Because 63 "... in many cases, nullity may not be an 'evil' to the person who has failed to satisfy some condition required for legal validity. A judge may have no material interest in and may be indifferent to the validity of his order ..." (Hart, op.cit., p.33). But this is too swift. It assumes that a sanction is necessarily perceived as an evil by the individual against whom it is directed. If a sanction is not always an evil, then it is no different from nullity which may also be, according to circumstances, subjectively perceived as an evil. But there are cases in which a "sanction" is not an evil to the individual who does not obey a coercive order, e.g.

(1) the martyr or the masochist

(2) a rational criminal who seek imprisonment to protect himself from vengeance

(3) vicarious liability - "evil" not inflicted upon the individual whose behaviour does not conform to the coercive order, but is inflicted upon another; the delinquent may not see this as an "evil"

(4) collective liability: the "evil" is inflicted upon a number of individuals which may or may not include the delinquent; again the delinquent may not perceive this as an evil [college staircase damage rule]

Furthermore, it is non-juristic to define sanction as an "evil" because what is an "evil" is a matter for individual psychology. All that a juristic definition can say is that a sanction is a legally determined consequence of particular behaviour, characterized as a "delict", "tort", crime" or "wrong". Hart's argument does not work and he is rightly diffident about it because a sanction may not be an evil to the person who failed to act in accordance with a coercive order. Thus Hart requires "nullity" to fulfill a test that even "sanction" does not fulfill, viz that be an "evil" to the person who fails to satisfy some legal condition.

Hart's second argument is, however, much more persuasive: [Handout] (b) Nullity is not a Sanction because we cannot subtract nullity from a rule and leave anything intelligible, whereas the removal of the sanction from the rule still leaves an intelligible standard of behaviour. This argument, although sound, does not disprove the "uniformity thesis".

A coercive order is the expression of a command plus a threatened sanction. The sanction can be removed leaving a simple standard of behaviour But a rule constituting a facility is of the following form: "To make a will you must do a, b , c, and d". But that is synonymous with: "In the absence of a, or b, or c, or d there is no valid will". Thus the expression of a facility creating rule is simultaneously the expression, or at least the implication, of the conditions of nullity. Nullity is implicated in the notion of validity. Thus to subtract "nullity" is to leave nothing. Thus Hart has shown that "nullity" functions in a logically different way to "sanction" Therefore Hart has shown that "nullity" is not or is not equivalent to "sanction". But this argument - sound though it is - does not prove that rules constituting facilities are logically different from rules imposing duties. The argument that nullity is a sanction is not a necessary part of the thesis that all laws can be reduced to one logical form but only part of the thesis that all laws are commands or orders backed by threats, i.e the "coercive" thesis.

Let us [Handout] Compare "making a will" with "committing an assault": A valid will is made if certain legally determined conditions (a, b, c, d) are satisfied; an assault is committed if certain legally determined conditions (p, q, r, s) are satisfied. Unless the conditions are satisfied no legally determined consequence follows. The "will" is invalid; the alleged "assault", a "non-assault"

Valid will if and only if a, b, c, d

If not a or b or c or d then no valid will

If there is a valid will, then the legally determined consequences ought to be implemented But here the legally determined consequences do not include a sanction. A valid will is an incomplete condition of a sanction, i.e. if the will is valid, it justifies, together with the satisfaction of other conditions, the application of a coercive act; these further conditions include interference with the property transferred under the will, and the initiation of legal action by the beneficiary.

Thus, the beneficiary can by his or her own motion bring about, as an ultimate legal consequence a coercive act directed against any individuals who interfere with his rights under a will. In practice, the sanction is still a long way off. In the first instance, the court will direct that the interference cease, that the goods be restored to the beneficiary, or that damages be paid. Only if the interference continues, or if the goods are not restored, or the damages are not paid will forcible reaction be brought about.

Assault if and only if p, q, r, s

If not p or q or r or s then no assault

If an assault is proved to have occurred the legally determined consequence ought to be implemented. Here the legally determined consequence is or includes a sanction. Assault is a complete condition of a sanction, i.e. if an assault is proved to have occurred, it justifies, by itself, the application of a coercive act.

Thus the analogy of "nullity" in a criminal law is the non-satisfaction or non-existence of the legally determined conditions of the criminal act. The analogy of "sanction" in a facility creating rule is the establishment of some but not all of the conditions under which a coercive act is stipulated as a legal consequence.

Thus of Hart's two arguments against nullity as a sanction, only the second is cogent. The cogency of that argument however does not prove that there are different logical types of laws and it certainly does not prove that P-rules are not necessarily related to D-rules (see cited passage 44). I will turn next to Fragmentation of Law and Dependent Norms:-