1. Caligula’s pillar

Caligula was an emperor who made a ruse of the laws. He made his laws known, whilst at the same time making it as difficult as possible for the people to know of them. Seeking to replenish the imperial coffers he issued tax laws that specified hefty fines for their violation, and had them inscribed in small characters on a tablet hung high upon a pillar.Footnote ¹ The emperor thus circumvented the Roman practice of giving notice of the laws, posting them publicly for all to see, whilst obstructing the people’s ability actually to learn of them. From Dio we have the conclusion of the story.Footnote ² The people, indignant at this leonine ruse, shouted their protests to the emperor in the Circus Maximus, whereupon Caligula had the protesters brought to the centre of the amphitheatre and summarily dispatched, so all Rome could observe how his will was to be done. The appealing intuition captured in the story of Caligula’s pillar is that law really should not be like this. The entire episode, embellishments and later accretions aside, seems more like fiat than the authoritative kind of enactment that we associate with the making of law. What appears particularly objectionable is the ruse itself, which delivers up the unwitting to the caprices of a tyrant who (ab)uses the law. Caligula’s pillar trick is an insincere command, and parasitic on ‘normal commands.’Footnote ³ Of course it might be dismissively said, ‘what else than extemporary decree do you expect from a madman, who wanted to make his horse a Consul?’ But we should note in this connection the salutary words of Daube (Reference Daube1967, p. 12), who counsels us not to dismiss Caligulan law as a species of ‘mad prank’. Of course, the days of Caligula and his horse were numbered, but however moribund they were, this does not settle the question of Caligulan ‘law’.

The intuition that law should not be like this, after all, says nothing about what law properly is, in any but a negative sense (‘law is not like that’). And so there remains the question of whether these laws are so unlawlike that they are not-law, or are just bad law. We might say with Blackstone that laws, qua directives, must perforce be promulgated, and believe that lays to rest the question of Caligulan ‘law’, negatively.Footnote ⁴ But that would not suffice for, via the pillar, Caligula’s laws were indeed promulgated. Then we might say that, due to its gross procedural irregularity, not even a dimly Austinian conception can rescue these commands as law.Footnote ⁵ But this would require an argument about procedure, artefactuality and access to be made out. In this respect an analogy to the Twelve Tables, occasioned as they were by the plebeian demand for generally accessible laws, will not do.Footnote ⁶ The legal provenance and nature of the different forms of imperial law has been much controverted, and likewise the debate over Caligula’s laws is not readily settled.

And so the story of Caligula’s pillar might lead us to wonder about law, and its knowability. We might ponder what connection there is between promulgation and the nature of law. What makes law exist as a thing that can be known by us? Is there something about the making of (positive) law that renders it a thing that we can know? Consider the following proposition:

P.1 Laws are things that can be known.

The argument, in brief, will go like this: laws are things that, qua artefacts (things humanly made), can be known.Footnote ⁷ The paper sets out to defend P1, which will be called the condition of access, or ‘access-condition’, against refutation. The truth of P1 will be maintained contra the supposed counterexample of the ‘secret law.’ Secondly, the paper considers some possible consequences of accepting the proposition, namely whether P1 commits us to either of the following:

P.2 Making people know of the law is desirable.

P.3 Ignorance of the law is no excuse.

2. Need laws be knowable?

An argument for the proposition that laws are things that can be known can be distinguished from two other responses to the promulgation question. These may be attributed to Hart and to Fuller, the latter of whom saw in Caligulan law a clear outrage to the demands of legality.Footnote ⁸ Hart, in The Concept of Law, distinguishes the rational communication by which laws are authoritative reasons for action from Austin’s concept of the command. Arguably, Hart’s conceptualisation of law suggests that law is a kind of thing that is taken to be knowable. That there are standing rules, and that by their indicia they are knowable, we might say, is part of law’s claim to be authoritative, and not merely or principally coercive. As directives, laws are not like Austinian commands, for the latter presuppose direct communication between giver and recipient. On this basis Hart contends that laws are general and thus unlike orders, for orders are standardly communications that are directly and personally addressed. Notably, in distinguishing laws and orders he subscribed to a particular notion of ‘addressing’:

With laws there need be no ‘Hey you. . .do this’, that is, there is no need for an interpellation preceding the directive, communicated directly from giver to intended recipient of the message. Hart is saying that the act of making laws differs from that of ordering people to do things. Central to his distinction between the giving of orders and the genesis of laws is the necessity of address to ordering, and its non-necessity to the making of law. This attracting of the addressee’s attention need not be a part of the genesis of laws, he says, but then comes the caveat:

It may be desirable in certain respects that laws are brought to the notice of the populace. However, the onus rests squarely upon the subjects of law to discover for themselves what laws there are, and how they may be affected by them. Hart observes a distinction between the bare existence of law, and the function of law as a mode of social control. This corresponds with a distinction between promulgation and publication. It is correct in the respect that it precludes publication from being a necessary condition for law. However, Hart’s position requires elaboration in the light of his broader conceptualisation of law. Particularly the importance of promulgation and that it imparts knowability ought to be made clearer.

Where Hart’s account is particularly promising is concerning the connection between laws not (necessarily) being addressed to the people, and the presumption of cognitive autonomy:

Laws standardly are not individualised in scope or delivery, and they are directives given without further direction. This is not to say that no legal system ever attempts further direction, but that may be left to section 4, below. Now the fact that general laws are made to meet this social control problematic means that it must be possible to understand them as rendering conduct non-optional, without additional informative or educative instruction on the part of the lawmaker. Secondly, the communication of general standards by general forms of language means that the subjects of law must be capable of recognising instances in which the case falls under the rule. The possibility of social control through law therefore presupposes a basic cognitive autonomy of the legal subject. This presupposes the legal subject’s unsupported recogitional capacity, which concerns discerning whether and when the rule applies. Hart adds that there may be uncertainties in particular concrete cases, and says:

Notably, Hart is saying that particular fact-situations are not of themselves differentiated from each other as this kind of situation or that. Fact-situations are not, outwith our recognition of what they are, ‘marked off’ as, say, an emergency or something more normal. We may surmise from this that the class of instances that are regulated by law are not generically vested with telltale signs indicating their regulability by law in a way obvious to us. The unlawful does not somehow raise a red flag to us, saying ‘Prohibited: unlawful’, or even ‘Proceed with caution: there may be law here’. Furthermore, the rule cannot ‘step forward’ to claim the instances in which the conduct it regulates is non-optional, simply in virtue of its existence. Precisely because law is artefactual, we should maintain a conceptual gap between a legally regulated fact-situation and the actual knowledge of its legal regulation. Thus far an argument in two steps has been adduced, namely that:

1. 1. something about the genesis of law is also a thing that makes it knowable, so laws are things that can be known;

2. 2. the kind of artefactual entity that law is presupposes a cognitive autonomy on the part of those to whom the law applies.

Step 1 gives the access-condition. Step 2 harks back to something like Hart’s puzzled man, as the practical reasoner upon whom our concept of law might settle. It is a matter of curiosity that legal positivism has not accorded greater significance to the question about promulgation. If there is a recognitional element, this deflates the assumption that the bare existence of posited law somehow of itself identifies law to us.Footnote ⁹ Consider the following proposition:

To recognise that the law prescribes Φ is not merely to know in the abstract that the law prescribes a certain kind of action. It is of no avail to know of law if one cannot or does not recognise in medias res the existence of an actual legal reason for action. This is to be able to recognise, when one is in the ‘fact-situational’ circumstances in which the legal reason does indeed obtain, that the law renders this particular conduct non-optional. Knowledge of the rule and its operative facts are hence necessary parts of an account of a complete legal reason.Footnote ¹⁰ Can recognising a legal reason for action, and therefore being able to take it as specially weighty, or even as conclusive if one is so inclined, simply rest on the fact of its positedness? Is the ‘authoritative mark’ (Hart, Reference Hart1961/1994, p. 95) by which a person identifies law as law sufficient to allow practical reasoners to recognise legal reasons for action? We might contend that its ‘pedigree’ is necessary both to law and its knowability. The authoritative mark that makes law distinguishes law and not-law, and is hence necessary to both law and access. We may instance the view of Pufendorf, who maintained that to know law we need to know not only its content, but also that it issues from one vested with the authority of lawgiver.Footnote ¹¹ And so we may inquire into what identifying marks their genesis impresses upon laws, that may also make legal reasons for action things that can be known.

I have ventured the proposition that laws are things that can be known. The idea is that the concept of law includes a necessary element that I have called the ‘access-condition.’ This is merely a convenient term of art to convey the idea that laws are things that are accessible, in the strictly limited sense that they are things that can be known. There is an important difference between this basic access-condition and a requirement of publication, publication being the activity of taking steps to attract people’s attention to laws. I concur with Hart that this may be desirable, but however desirable it can hardly be an element of law per se.

This brings us to the competing position, which holds that to exist, law must be publicised. This line of thinking is seen in Fuller’s argument for desiderata of law. These he presented as both basic conditions for the existence of law and standards of success in the ‘enterprise’ of law. Fuller holds that a complete failure in any one of these eight conditions does not only issue in a bad system of law, but in something that is not properly called a legal system. Fuller thus asserted that these were not mere desiderata for efficiency, but built into the existence conditions for law. He further claims that they constitute an ‘internal morality’ because law that has these criteria present conduces to a state of social order, while also respecting human autonomy in using rules to guide behaviour. The second of Fuller’s (Reference Fuller1969, p. 39) desiderata is publicity (the rules must at least be made available).Footnote ¹² Note that Fuller’s threshold condition of availability hardly exceeds my access-condition. Nevertheless, when he goes on to explain this desideratum, it is plain that his requirement is considerably more exacting than mine.Footnote ¹³ It is also the case that he disregards the connection between artefactuality and access, but this will be discussed a little later in the paper. Fuller says about publicity that a system of rules that fails on that score cannot guide behaviour because people will not be able to determine what the rules require. Fuller (Reference Fuller1969, p. 98) says that ‘laws ought to be published so that those subject to them can know what they are’ and thus his main ratio for publicity is knowability and not, say, check on the content of laws. Interestingly, Fuller (Reference Fuller1969, p. 44) says that a formalised standard of promulgation tells the lawmaker where to publish his laws and lets the citizen know where to go to learn what the law is. But his concern seems to remain with holding the lawmaker to stringent requirements and he raises little discussion about the duties and difficulties on the part of the citizen. His (Reference Fuller1969, p. 54) desideratum that ‘the laws should be made known to those affected by them’ is employed to demand publication. He says (Reference Fuller1969, p. 51) that the citizen is ‘entitled to know’ of the law, but at no point does he mention any kind of corresponding duty to take reasonable steps to know about law on the part of the citizen. He chides the positivist tradition for supposedly conceptualising an inert legal subject, and counterposes to this his allegedly superior interactional-contractarian view of law. But he then says nothing about the exercise (and indeed the sometime difficulties) of the legal subject’s cognitive autonomy qua her being actively informed and so responsible and indeed, according to the ignorantia maxim, prima facie liable.

Unlike Hart, who recognised a distinction between the two, Fuller confounds promulgation with publication. On that basis he falls foul of provincialism and derides Austin’s approval of the English completion of promulgation at the Royal Assent (if this had been said by Fuller merely about desirable methods of promulgation it would be less objectionable, although jingoistic even then). We need be clearer than this about what promulgation is. Promulgation is the declaration of a law, the formal statement or proclamation of new legislation. It is the act by which the authorised official (for example the Head of State) attests that a given text has been approved as a law. Promulgation hence attests the creation and validity of a law. Promulgation is all that is required by P1, and promulgation via the Royal Assent meets this basic requirement. The Royal Assent marks the culmination of the legislative process, as the institutional act by which a bill receives the final and formal approval required for it to become law. This final act that makes a bill a law also makes that law public, in the sense of imparting a basic availability to the generality of persons whose conduct it governs, who thereby may discover what it requires of them. No more than promulgation is required by P1; under P1 publication is not necessary to law. Of course in some countries promulgation is concomitant with publication, when promulgation is deemed done by publishing laws in an official gazette, bulletin or journal, and a law is not in force until it is promulgated in this particular way. In that case, an extra publicatory activity is being done at the same time as formal promulgation. Yet other countries might recognise a governmental responsibility to publish laws, but not deem publication necessary for the creation of law. This might accompany vacatio legis, say of one month, during which time the text of the law can be published, and people can avail themselves of the purport of an Act and its commencement provisions.

I mentioned above that Fuller disregards the connection between artefactuality and access. Fuller says that a coincidence of moral and legal demands relaxes the stringency of the desiderata of publicity, clarity and retroactivity. Consider a case in which law is made to apply only retroactively. In this case the law does not exist at the time when a person acts, and thus the fact-situation is only regulated by law afterward. This means that a person does something he or she could not possibly have known to be illegal, since it was not illegal when he or she did it.Footnote ¹⁴ Here, nonretroactivity is redundant on executability:

Where law is a reflection and expression of extra-legal morality it merely confirms widely held views and the requirements are less exacting. So, according to Fuller, if law is ‘congruent’ with morality, the desiderata of publicity, clarity and retroactivity diminish in importance. It might be replied that a coincidence of law and mundane moral notions may arguably exist, but cannot be assumed however.Footnote ¹⁵ Now remember Hart’s assertion that particular fact-situations are not inherently discrete and do not await us flagged up as regulated by law. Well, one aspect of this may just be that law and personal moral views do not always coincide, so a fact-situation may not correctly appear to us as morally salient and thus possibly legally regulated. Others might adduce a line of argument about law being known naturally. They might appeal to per se nota practical reasons, which are self-evident. For example, they might contend that acts which are mala in se can be identified without positive laws, for reason can know per se nota truths. But much of criminal law is mala prohibita, and crimes that were once mala in se can be decriminalised altogether. So the per se nota does not reliably tell us what law is. We should remember the artefactual character of the law, and that even criminalisation is a social practice. The access-condition provided by P1 requires no presumption that everyone can know the content of legal directives directly (that is, without knowing of the laws themselves).

For all these reasons, Fuller’s arguments cannot be accepted, even in attenuated form. Indeed we can concur with Shklar (Reference Shklar1964, p. 108) that Fuller falls prey to an unnecessary ‘definitional self-insurance’. As concerns the nature of law, Hart was more right and Fuller was more wrong. But Fuller was onto something, and something to which Hart directed little attention. What Fuller was onto is the value of raising questions concerning whether certain kinds of publicising law are desirable. In Hartian theory, the law needs must take the form of rules which can be applied by private individuals to themselves without further official guidance or weighing up of social issues (Hart, Reference Hart1961/1994, p. 130). Questions remain concerning the considerable onus that such an approach countenances resting on the citizen, who must take reasonable steps to know the law and to recognise instances in which to guide her conduct by it. We will come to these in section 4, but first we must deal with a counterexample proposed by Fuller.

3. ‘Secret laws’

Fuller says that if there are secret laws, then there must be implicit laws of lawmaking. This however is counterfactual, since there are not secret laws. Nevertheless, it is instructive to trace his faux pas. Fuller (Reference Fuller1968/1981) tells a tale about a newly created nation that adopts a written constitution. This clarifies by whom and how binding laws must be made. The first act of the new legislature is to pass a resolution that all laws enacted by it shall be kept secret from the citizenry. The constitution had made no requirement as to the publication of laws. Fuller remarks:

Here Fuller is asserting that if there is a principle of access, there must be ‘implicit laws of lawmaking’. This is an untenable position, guilty of non sequitur. He then turns to reductio. He comments that the view that secret law is bad law, but law nonetheless, relies on a view that there are no implicit laws of lawmaking and says:

Here he seeks to establish his contention that there are ‘implicit laws of lawmaking’ by deriving an absurdity (the inability to distinguish concealed laws from extensively publicised laws and to account for what he takes to be the obvious superiority of the latter) from its denial. However, the obvious reply is that we can distinguish concealed from publicised laws, and so the reductio founders.

It is relatively straightforward to defend P1 against the supposed counterexample of secret laws. If P1 is true, there is no such thing as a secret law. The possibility of a secret law stands then as a certain kind of test of P1. But P1 is true. There is no such thing as an inaccessible law. There cannot be secret laws, but there can be hidden laws, or laws that to all intents and purposes are hidden. The point is that laws can be created, that are then kept confidential, within a certain closed circle. Let us return to the story of Caligula. The point of this story is that the lawgiver should not actively conceal existing laws, or otherwise take steps to obstruct the people’s knowledge of laws. States do seek to keep some lawlike things confidential. For example, a person might challenge the government’s issue and maintenance of binding regulations that they keep secret from her. The government could reply that the keeping secret of these regulations is justifiable because they constitute ‘security sensitive information’. In such a situation the individual is in a position of being required to comply with legally binding regulations both unknown to her and that indeed she is being forbidden to know. The government might further add that the court can be told of the regulations in question, under seal, but these must not be made publicly available. So there are questions about justifiable restrictions on access, but the access-condition itself is not damaged thereby. Under the access condition law need not be actually known to everyone, so long as it is in principle available.

4. Consequences of P1

I have ventured the proposition that laws proper can be known. We now proceed to questions about the consequences of P1. Does either P2 or P3 follow from P1?

P2. Making people know of the law is always desirable.

This statement is untrue. In rejecting it we should reject the suggestion that procedurally proper (and in our case knowable) law tends toward being substantively good law. We should not presume the necessary goodness of knowable law. This is not to say that unknowable laws cannot be abused, and indeed that a basic knowability of law may not exclude some forms of abuse of law.Footnote ¹⁶ We must be careful to avoid the suggestion of any necessary connection between the ability to stand publicity and the rightness of a position. And so we may ask how far the state may go in making people know of the law. Consider ‘public legal education’, whereby the state helps citizens to be better able to recognise their legal rights and responsibilities? Is this desirable? Some might seek to posit from P1 the necessity of public legal education, as indeed Bentham did. In ‘Of Promulgation. . .’ (Reference Bentham and Bowring1843) Bentham wrote that even once a law has attained the royal assent, a work of promulgation remains. This is no mere matter of publicity, as though one might trumpet the passing into the corpus legis of a new law, and be done with it. Law must not only be publicised, but also explained in order that people may understand why it has been established. Rather, effective promulgation must be of the nature to make laws memorable to those who are to be governed by them. Indeed and relatedly Bentham advocated promulgation of the reasons for laws. He envisaged understandable and memorable laws as accompanied by a ‘commentary of reasons’, a sort of dicta legis for the ordinary person. The commentary would not be inherent to the text of the law itself which, Bentham observed, ‘ought only to present a pure and simple expression of the will of the legislator’. If laws were accompanied by an exposition of justificatory reasons, they would be more easily known, more readily memorable and more cordially approved. Notably, Bentham relates promulgation to publication and a relatively ambitious educative theory of law. He contends that publicised laws, if accompanied by dicta legis, would infuse themselves into people’s minds, form part of their reasoning, and influence their moral nature and the cast of public opnion, with the upshot that ‘obedience to the laws would come to be hardly distinguishable from the feeling of liberty’. This is highly contentious. A thorough appreciation of Bentham’s argument is beyond the scope of this present essay. Nevertheless, some indication of my concern about its uses may be advanced. Recall step 2 of my argument (above, section 2). This provided that, given the nature of law, individuals must exercise certain cognitive capacities of knowing the law, and thus play a part within the ‘can be known’ formulation. One feature of this necessary exercise of cognitive capacities relates to the artefactual gap between a rule and its operative facts. This does not, of itself, justify heavy-handed ‘making known’ of the laws by the state. It is not a licence for intrusive and obtrusive legislation and regulations. Publication means taking reasonable steps to draw the people’s attention to the legislation. What is reasonable is open to debate. Imagine the introduction of new legislation, requiring all persons to refrain from doing a certain commonly done thing in public places, smoking for instance. The government, as part of this legislation, requires the prominent display of compliant signage. On the one hand, we might say that the signage conforms to a certain principle of fair notice, conforming with a requirement that people be placed on fair notice of their potential subjection to criminal sanctions (that specific contemplated conduct is prohibited and subject to a penalty). The signs tell people that there is a (new) law, what it is, and that it applies in this instance. On the other hand, we might not want to see obtrusive reminders that and what the law is when going about our everyday life. We might contend that law should not be undesirably noticeable. Imagine that the signs have a legally required form, which includes displaying the name of the person to whom a complaint may be made by anyone who observes someone violating the new law. Here there is a kind of recruiting private citizens into criminal law enforcement; perhaps soon we will have a revival of qui tam actions!

In section 2 above I discussed Hart’s distinction that orders but not laws are personally given. From P1 we might infer a normative principle relating to modes of address and cognitive autonomy. Hart says that criminal law designates by rules general norms, or standards of conduct for the guidance of ordinary citizens in their everyday life, but that those individuals to whom the law is applied are ‘expected without the aid or intervention of officials to understand the rules and to see that the rules apply to them and to conform to them’ (Hart, Reference Hart1961/1994, p. 38–9). The criminal law, Hart observes, is not like a policeman on traffic duty, directly issuing individuated orders, visibly and personally present directing each to stop or go:

When conceptualising laws as directives, we should make it clear that we reject the policeman-writ-large. For laws are not only unlike the gunman’s orders; they are also unlike the policeman’s traffic directions. Laws should not be that obtrusive. There is some difference between making people know the law, and permitting them to know of it. There should be limits to the coercive and the educative efforts of a state. Too many policemen-laws and the cognitive autonomy of a reasoner is under threat.

P3. Ignorance of law is no excuse.

Ignorance of law is a lack of knowledge that a law exists. From the ‘can be known’ formulation in P1 may be derived a ‘could have known’ formulation that goes to culpability. Is there a duty to ascertain our legal obligations? Step 2 of my argument can be observed to found a principle of responsibility on the part of the individual, of which knowledge of law is a part. Sometimes ignorance of law can be culpable, for example in cases where there is wilful blindness. Perhaps there are even kinds of epistemic akrasia, where I believe against my own better judgement as to the weight of the evidence. I knowingly accept as true while also accepting that it is epistemically wrong to so believe. I engage in a kind of self-deception. The reasons are actual but I fail to see them. I cannot take them as reasons for action, but this is my own fault. Perhaps this occurs when I disagree with the law, or desire so much to do the unlawful that I neutralise the prohibition. When should ignorance of law exculpate? Mere offence seriousness cannot generate a negative answer; for that we need to know about when one’s ignorance of law is at least negligent. An argument about P1 and what follows from it concerning criminal responsibility might appeal to a notion of fair opportunity.Footnote ¹⁷ Remember my proposition about the knowledge-element of a complete legal reason for action (section 2, above):

Cases of excusable mistakes of law are those in which the law has not been reasonably accessible to one. Consider three cases. In case α, a person lays in a coma for a lengthy period, during which time many new laws are passed. Upon waking she inadvertently breaks one of the laws passed during her repose. In case β, a person is doubtful about a law. She consults the government department responsible for administering it, and is misadvised as to the law by the relevant official, who tells her that her planned course of action is lawful when it is not. The two examples differ in that in only α could the individual not have known of the law, for purely agent-relative reasons (she was incapacitated). In β the individual could have known of the law. Consider now case γ. In γ a person is doubtful of the law, and proceeds with her chosen course of action without taking any consultation. This course of action is unlawful. β differs from γ because in only β has the individual taken steps to avail herself of the law. From P1 then we may derive a principle about fair opportunity, but this should be understood to cut both ways. The state should make the law available, but then the individual must take the opportunity to avail herself of it.