by Iain Stewart¹

(2002) Oxford U Comparative L Forum 4 at ouclf.law.ox.ac.uk | How to cite this article

Summary | Sommaire

The discipline ‘comparative law’ seems still to be distinguishing itself from lawmaking. In this paper I investigate theoretical reasons for this and propose a new type of concept of law, which should secure that distinction. This proposal is also made for further theoretical reasons, as well as for empirical reasons drawn from the structure (or rather lack of structure) of the Australian legal system, being both a common-law system and a system with a peculiar history.

La discipline ‘droit comparé’ paraît être encore en train de se distinguer de la création du droit. Dans cette intervention j’en recherche les raisons d’ordre théorique et propose un nouveau type de concept du droit, qui devrait assurer cette distinction. Ce propos s’appuie également à d’autres raisons du même ordre, et en plus aux raisons d’ordre empirique qui proviennent de la structure (ou bien du manque de structure) du système juridique australien, étant à la fois un système de common law et un système marqué d’une histoire particulière.

1. Approaches in Comparative Law

‘If I were you, I wouldn’t start from here’
(alleged Irish saying)

1.1. Le Droit Comparé n’est pas un Droit

‘Comparative law’, Pfersmann has recently observed, ‘is an odd discipline. Long considered to be exotic and irrelevant to real issues, today it is a required component of university curricula and legal practice’. It has, he says, become indispensable for the researcher, the judge and the legislator, especially as globalisation turns legal idiosyncrasies into obstacles. It

has given rise to great hopes and ambitious projects, yet it still relies upon the weakest of epistemologies.

Judging by what is undertaken under the label ‘comparative law’, the following beliefs can be found to predominate, more or less implicitly, in discussion about it: (1) comparative law is a transnational legal system; (2) it is a science facilitating the unification of different legal systems or anticipating the unification (or, failing that, the homogenisation) inherent in the evolution of legal systems on a global scale; (3) it is the science of foreign legal systems; and (4) it is a science that assists in resolving individual cases before the courts.²

Pfersmann finds all four of these ‘beliefs’ inadequate. Their primary fault, in his view, is that they proceed from ‘the illusion of a legislator-jurist (l’illusion du juriste-législateur), whether through projection of an ideal order or through claiming capacity to produce one’. They involve ‘the naturalistic sophism that induces the jurist’s habitual fantasy of believing oneself to be a producer of ideal rules inasmuch as one is an expert on positive rules’.³ Pfersmann identifies the problem as epistemological and ontological, and tries to locate its source in the nature of a positive legal order. I will investigate this identification.
The ‘illusion of a legislator-jurist’, he is saying, is on one level an illusion as to achievement. There is a ‘projection of an ideal legal order’, so that ‘comparative law is a transnational legal system’. His proposition may seem overstated. Yet it does appear to fit the most prominent founders of the discipline ‘comparative law’.⁴ And, even though such an outlook is no longer espoused, it might not have been effectively superseded. For Pfersmann is only one of several writers who still find it necessary to stress that ‘comparative law’ is not a type of law but a discipline.⁵ The illusion is implicit in the very label ‘comparative law’ or (which the former translates) ‘droit comparé’. The illusion persists as an implication of the label. Even if nobody now follows that implication, it may remain necessary to warn against doing so.
The other level that he identifies is an illusion as to capacity. On this level, however, it seems to me that there is not a complete illusion. For there is a sense in which a jurist can make law: the formulation of legal doctrine (la doctrine).⁶ If a positive legal order is understood in terms of ‘sources’ and if legal doctrine is elevated to being the primary source, then doctrine written by the comparatist could be seen as transnational positive law.
That vision, however, would be doubly flawed. First, the norms so produced might not clearly be legal. They would neither fit into the traditional distinction between the natural and the positive nor include the attribute of coercion, through which a positive legal order is ordinarily shown to exist. Second, even if the norms so produced would be legal, they would not have a specifically comparative character. For the act of comparison itself does not create norms. It is, at most, one element in the production of a norm. Once a norm is produced, any prior act of comparison is history.⁷
So, comparative law is not a body of law: le droit comparé n’est pas un droit. There is no ‘comparative’ legal order. Comparative law is a discipline. It consists of the use of comparative methods where the comparison is not confined to a single legal order. The act of comparison does not, itself, produce norms. At the most, it contributes to the production of norms within an existing national or transnational legal order. And it does indeed perform that function.
How, then, did the illusion that comparative law is a body of law arise? Nobody confuses a proposed legal norm with an actual one; everybody knows the moments of constitutional adoption, statutory enactment and pronouncing of judgement. Nor does it seem plausible to suppose that any comparatist has seen a divine will, Zeitgeist or other telic impulse refracted in the sort of technical detail through which legal comparison may be most constructive. Except—taking a hint from Pfersmann’s reference to ‘evolution’—for the evolutionist ideology, a background assumption of the era in which the discipline of comparative law was founded and sometimes tendentiously labelled ‘Social Darwinism’. Its high priest was actually not Darwin but the now almost forgotten Spencer, whose combination of philosophy, religion and unified science included a doctrine of natural law.⁸ What may have happened in comparative law is that an idealisation of positive legal order was projected as the content of that natural law,⁹ so that the comparatist’s statement of the ideal positive law would function as a statement of that law as natural. The habits of idealist iusnaturalism would have been transferred into a primarily positivist outlook.

1.2. Legal Orders, Closure and Alterity

More interesting now is the possibility that Pfersmann may be correct so far as he identifies a continuing risk in a combination of transnationalist legal doctrine with what he terms an epistemological and ontological ‘closure’ of legal orders.¹⁰
The closure that he identifies is, he says, formal. Legal systems are ‘formally closed’, in that a norm is legal solely in virtue of a validity with which it is endowed by a pre-existing legal norm. At the same time, legal systems are ‘substantively open’ in that ‘any content of any possible legal norm can become that of a norm in an existing legal system, unless it conflicts with a norm already in force in that system’. If there are incorporation rules, the foreign rule that is incorporated in accordance with them ‘will be valid within the system that incorporates it, as a rule of that system, by reason of the rules of that system’ and ‘with no effect upon the closure’ of the latter. Pfersmann thus prefers a ‘presupposed ontology of the closure of chains of validity’, which would be a ‘constitutive ontology’.¹¹
In this constitutive capacity, nonetheless, he finds ‘an extremely serious difficulty’: that, for a particular legal system, such an ontology would not ‘constitute the sole conceivable model’; ‘obviously, if one is allowed to determine the constitutive properties of any object, nothing will prevent one from doing so in another way than that proposed here’. His response to the difficulty is elliptical yet illuminating: ‘However,’ he says, ‘the problem is then merely inverted and it is the expressions “national laws (droits nationaux)” or “legal systems” that no longer make sense’.¹² He does not explain why such an inversion would take place, but appears to say simply that to admit alterity into legal ontology would render incomprehensible these basic notions of legal science. He might better have asked: how could one validly object, on the criteria of a modern science, that ontology (or epistemology) has been opened up to alterity? It would not follow that all alternatives would be of equal value. If such an objector were to share Dostoyevsky’s old moral fear—if God did not exist, everything would be permitted¹³—one could reply that in science, at least at this basic level, one should be avoiding moral censure anyway. At any rate, it is a while since the Sovereign King lost his head and the God-State vanished away. How is it, even so, that alterity is still to be introduced and that, with its introduction, such basic categories of legal science would cease to make sense?
It is still to be introduced into legal science. Behind the problem of one legal order being closed to another lies the problem that a legal order is closed to everything but itself. On a prescriptive plane, that might not be a problem: after all, one could say, the very aim of law, with regard to a particular sphere of behaviour, is to secure conformity to a particular norm and not to any conflicting norm. The problem is rather that this exclusivity also appears on the descriptive plane. I will not argue for the standpoint of a detached observer, which is not feasible where what is to be observed is social meaning. Rather, I shall note (as I have put it elsewhere) that in contrast with legal science

every other science of meaning (with the possible exception of theology) accepts, so thoroughly for it to be banal, this principle: that the meanings composing the perspective, the framework of investigation, shall be, at least logically, independent of the meanings that are the subject matter of the investigation; which has the corollary that the meanings that are the subject matter will be, in the first place, somebody else’s.¹⁴

I would contradict this principle if I were to claim that it could not be formulated differently. Nonetheless, adherence to this sort of formula seems to me to be a commitment to the independence of thought that is a necessary qualification for a position in the academy.
The plight of legal science is that, when this principle has been adhered to, it has been adhered to in two ways and neither of them has been successful. The difference between those two modes of adherence to the principle is marked by whether the scholar also attempts adequacy to the phenomenon that is socially called ‘law’—that naming being an element of the phenomenon. This phenomenon (I shall say by way of empirical intuition, and there seems no other way to recommence) consists centrally of norms that appear as simply to be obeyed—to be obeyed unconditionally and for no other reason than that they are ‘law’.
The most rigorous attempt to achieve both adequacy to this phenomenon and descriptive alterity, Kelsen’s ‘pure theory of law’, foundered in self-contradiction.¹⁵ Others who have been committed to descriptive alterity have abandoned adequacy to the phenomenon and dissolved the concept of law into a broader normativity or into some even wider category—sometimes pursuing the fallacy that breadth is intrinsically a desirable quality of a definition.¹⁶ The intermediate path of ‘contextualisation’ only brackets and postpones the issue: so long as the principle is not applied in constructing the very concept of ‘law’, attempts to amplify that concept with historical or social ‘context’ will be plagued with conflict between effects of that principle’s absence in the core of the model and effects of its presence in the disciplines that are summoned to fill the model out.¹⁷ Alternatively to using the principle at all, adequacy to the phenomenon is desired while descriptive alterity is evaded in preference for a regression to apologetics.¹⁸
Pfersmann has a separate objection to the third belief that he lists, that it is merely incoherent. Every ‘other’ legal system, he observes, is ‘foreign’ in relation to the one that is the point of departure, without there being any scientific reason for so privileging the latter. The apparent foreignness of any other system ‘is merely the consequence of the fact that different legal orders exist in different places and at different moments’.¹⁹ This seems to me unfair to the extent that it is surely legitimate to undertake legal comparison with a view to reforming the legal system under which one lives and in relation to which every other legal system is ‘foreign’. Yet the criticism does seem fair to the extent that it questions the apparent obviousness of a commitment to the home legal system. Is that commitment, in its obviousness rather than in being the outcome of a decision, a consequence of a closure of the home legal order to alterity, whether foreign or domestic? If so, the comparatist exercise intrinsically involves the identification of alterity.
The closure of law can then be maintained only by a subsumption: the conflict between the closure of any legal order and the comparatist recognition of alterity might be resolved by supposing a subsumption of both within a higher legal order. In this way, the temptation to suppose a body of comparative law (un droit comparé), actual or at least possible, may arise from the very exercise of legal comparison.²⁰ Either one recognises an already existing higher legal order, under which the products of comparative law can be subsumed, or the exercise of legal comparison may be assumed to produce a higher legal order, consisting of very general norms, that would accommodate the more particular products of legal comparison.
When the temptation to subsume is resisted, however, the comparatist exercise remains intrinsically critical of legal closure: even though the closure is at first sensed only as parochialism, the critique commences just as a feeling of inadequation, and only through development of the critique does either the closure become visible as closure or the critique understand itself as critique.
By that point, however, the stakes have become very high. Intuitively, closure is a bad thing while critique is a good thing, even if all that it does is to reveal closure. But then, if law is characterised by closure, to that extent law is a bad thing. At best, it is a necessary evil. Critique might identify both the evil and its necessity. Now, that can readily be done with regard to the coercive aspect of law, portraying it as social self-defence. But closure is in a different league: it seems to involve belief. And how can a belief be a necessary evil?
Pfersmann does not follow through from his identification of closure and preference for a constitutive ontology. In the end, the comparison of legal orders is for him their ‘differentiated conceptual interpretation (l’interprétation conceptuelle différenciée)’.²¹ The practical contribution of legal comparison, which should not be political, can nonetheless be technological: by examining a variety of legal systems, the comparatist becomes able to advise how, in any one of them, a particular reform may have a particular outcome. The value of that outcome will be a separate, moral issue.²² Yet differential interpretation may be just what is excluded by closure. And I suspect that one avenue that legal closure closes off is that of a technological neutrality subject to moral choice. That which is simply to be obeyed by all is also simply to be served by the specialist.
Moreover, these issues of comparative law appear also, and if anything more starkly, in the construction of general theory of law—for comparison is a necessary element of generalisation. General theorisation of positive legal orders is thus possible in either of two ways. Either through subsumption within a higher legal order—be that an order of natural law or the international legal order in a monist perspective. Or through a recognition of alterity and, eventually, of closure—where that recognition takes place within a theoretical framework that is independent of its subject matter. In these respects, general theory of law and issues of the identity of comparative law are mutually relevant, both through analogy between the two disciplines and so far as the exercise of legal comparison is a necessary element in generalisation.
The curricular status of comparative law may become clearer in this light. If the curriculum is dedicated to producing lesser servants of the home legal order, a study of comparative law may be irrelevant and even threatening. It can become relevant when the student is encouraged to adopt a perspective of law reform or to consider that they might become a legislator or a judge, more relevant when the home legal order is in fact being subsumed under a higher legal order,²³ and vital when the science of law is presented as grounded in general theory. In addition, students of general theory within a curriculum devoted to a common-law system must be introduced to the western legal tradition’s romanist mainstream, to which most general theory refers.
If, then, the game is ‘closure versus alterity’, Pfersmann’s indication is that the closure is located in the relations between legal norms. However, the case of the common-law systems indicates that this is not generally correct. I will take the Australian legal system as an example. Afterwards I will suggest that, in any legal order, the closure is to be found not in the relations between norms but in the norms themselves. Then I will examine some of the implications of that view for the discipline of comparative law.

2. A Lack of Structure

It is conventional to inquire into the ‘structure’ of any ‘legal system’, assuming that a legal ‘system’ is the behavioural embodiment of a legal ‘order’ which is structured primarily either through deduction or through authorisation.²⁴ But common-law legal orders in general, and the Australian legal order in particular ways, are remarkable for their lack of structure; it is even difficult to identify them as ‘systems’.
Certainly, there are legal ‘systems’ in an administrative sense: there are the expectable persons of law (legislators, ministers, judges, lawyers, police officers), places of law (parliaments, courts, prisons), and things of law (books, guns, robes, uniforms and cars that go ‘da-da, da-da’ at any time of the day or night). But it is hard indeed to discover a structure of normative meaning. Superficially, there is the Stufenbau common throughout the western legal tradition: a constitution (usually), statutes and regulations, and a hierarchy of judicial authority. But the constitutions, statutes and regulations are understood to mean what the judges say (or probably would say) that they mean. And, when that is seen, the picture of a hierarchy breaks up. Firstly, the judiciary insists that it is an autonomous source of law. Secondly, what common-law judges say is difficult to understand as distinct ‘norms’ or, consequently, as involving distinct relations between norms.
In what follows, I shall mean by ‘legal order’ a set (ensemble) of legal norms and by ‘legal system’ the social behaviour that is the application of those norms by persons designated in terms of the legal order as officials.

2.1. Common Law

2.1.1. The Common-Law Mind

Common-law thought, at its core, is a process neither of deduction nor of induction but of manipulated analogy.²⁵ It is still basically exegetical, as throughout the western legal tradition,²⁶ but analogy has come to predominate.
The model is the curial process, in which the business of counsel is to persuade the bench of an analogy, being an earlier decision of that court or of a higher court, that ‘fits’. The judgement then has to provide reasons for accepting one analogy and rejecting others. Since the lines of analogy will be primarily factual, the judgement has to set out and examine the facts of the particular case in detail. In a major case before an appellate court, each judge can and frequently will provide a separate judgement, giving their own assessment of the facts and interpretation of the applicable law. It is possible for the whole bench to agree on the order to be made, even though no single line of reasoning commands majority support. For counsel to take an earlier case as an analogy is thus usually to select a particular judgement in the case. Opposing counsel might accept that the case is ‘in point’ but argue that there is a closer analogy with another judgement in that case. Judgements are very long, by European standards: where most or all judges give separate judgements, the whole decision can run to several hundred pages.²⁷ Today in Australia, decisions of the superior courts are immediately put online. Also available online, as soon as possible after each day’s argument, are transcripts of the proceedings of the High Court (the federal supreme court). The transcripts can turn out to be more important than the decision itself, as a pointer to future decisions, if the decision avoids or downplays an issue that had been argued extensively.²⁸
My point, briefly, is that in a common-law system it is generally difficult to identify distinct ‘norms’. Both in case law and in the judicial interpretation of constitutional and statutory provisions, it would be more accurate to identify ‘normative patterns’. This is not to say that common-lawyers literally ‘think in pictures’,²⁹ although they may sometimes do so when making factual analogies.³⁰ Rather, it is to say that they think in patterns rather than propositions, that they match more than infer. The skills of matching (and of denying a match—‘distinguishing’) are largely tacit and are acquired by imitation.³¹
To the extent that a common law system does not consist of distinct norms, neither can it consist of distinct relations between norms. Hence it is difficult to determine a sense in which one can speak of a common-law legal ‘order’—a sense in which one can say that a common-law legal system has a ‘structure’ in the sense of being an order of normative meaning.
From a practical point of view, the reliance on analogy makes the law highly flexible, to a degree that a civilian can envy.³² On the other hand, it leads to several types of obscurity:

• Lack of rational clarity: analogy has a limited capacity to state a general rule or principle, which would belong to deduction.³³
• Literal obscurity: a judgement of a superior court is incomprehensible to anybody without a legal qualification; and, since the law is not what was decided yesterday but what is likely to be decided tomorrow, one needs to be able to predict the analogies that will be entertained—and old cases never die.³⁴
• Lack of moral or political transparency: it is hard to perceive, under the analogies drawn, by what moral or political values the judges are guided.
• Unpredictability: this is a defect in terms of organisational efficiency³⁵; it is also a defect in being anti-democratic—so far as it amounts to ‘dog-law’, as Bentham complained so long ago.³⁶

These obscurities, however, are a goldmine for the legal profession. It is rarely possible to discover one’s legal rights or duties reliably without consulting, hence feeing, a lawyer. This compounds the anti-democratic bias: partly by privileging those who can afford legal advice and representation³⁷; and partly because the judge-made law, including constitutional and statutory interpretation, is then moulded upon the interests of those who can afford to sue.³⁸
The effect on legal scholarship has been to produce either doubtfully academic certitude or academic confusion. Either the legal academic accepts and pursues the centrality of analogy, which conflicts with scholarly criteria; the judicial process then tends to be seen as the research process and the legal academic as a research assistant to the bar. Or the legal academic attempts to convert matching into inference, which may be scholarly but is doomed to befuddlement.

2.1.2. The English Aberration

The historical root of this is what may be called the English Aberration from the mainstream western legal tradition.
Before the Norman Conquest of England in 1066, English laws had been substantially codified—both by Anglo-Saxons themselves and, in the north, by Norse invaders.³⁹ The Normans, although not ignorant of the romanist tradition, overlaid the existing laws with a system of ‘writs’ that privileged procedure over substance and a mixture of executive and judicial power over any form of legislation. This system privileged analogy in two ways: a writ would be issued if a matter were deemed either to fall clearly under that type of writ or to be sufficiently similar to a matter that had earlier been deemed to fall under it; and a new type of writ could be issued on analogy with an existing type that seemed not to cover a matter in which it was thought that a writ should be issued.
Corresponding to the privileging of procedure, there emerged a legal profession as masters of the procedural game. The core of the law came to reside in the royal archives of writs and in lawyer’s memories and notebooks, recording those forms and how they were handled. The romanist tradition was not forgotten, even by the lawyers.⁴⁰ The Norman judges were clerics (since few but they were literate) and acquainted with canon law, which continued in use for family matters. But, by the time that Justinian’s Digest had been generally revived in Continental Europe, the procedural emphasis of English law and a monopoly of the courts by a procedure-oriented legal profession were entrenched. The great game could not be played, because the small game was occupying the court.
The thirteenth-century work attributed to Bracton, which attempted to recast English law in Pandect mode, fell on stony ground.⁴¹ So did attempts to move English law back into the mainstream western legal tradition during the Renaissance.⁴² The influence of the new physical sciences, to which some lawyers were major contributors,⁴³ served only to encourage a systematisation of the procedurally oriented law. Chief Justice Coke’s seventeenth-century Institutes are still cited, as are the eighteenth-century Commentaries of Blackstone, which served as a handy compendium of English law for the American colonies and to a lesser extent in Australia. But Bentham’s efforts toward actual codification, in which substance would have been privileged over procedure, had far more success on the Continent than at home. One current result is English discomfort (and Scots amusement) in Brussels.

2.2. The Constitutional Structure of Australia⁴⁴

2.2.1. The Invasion

On 26 January 1788, eleven boatloads of illegal immigrants landed in a cove on the coast of a ‘great south land’. They unloaded their goods, raised a piece of coloured cloth on a pole and claimed the country for their king, who was mad.⁴⁵ They brought with them the law of England, which they made the general law of the new country. This is sometimes called the ‘reception’ of English law, but who ‘received’ it? It can be called a ‘reception’ only in a sense in which pugilism would illustrate the adage that it is better to give than to receive.
After some early fumbling, the colonisers gave no recognition to the normative institutions of the Indigenous peoples. That would not occur until 1992, when the High Court resurrected for Australia the old common-law notion of ‘native title’.⁴⁶ The colonial relationship of domination persisted, however, in that native title is ‘extinguished’ by any conflicting non-native title. Neither the courts nor the legislatures have been prepared to consider any form of Indigenous sovereignty.⁴⁷ Nor, unlike other parts of the former British Empire, has there ever been a treaty or similar agreement with the Indigenous peoples. They are not now mentioned in the Constitution.⁴⁸ Although the Indigenous peoples were recognised to have become British subjects, they were almost universally denied the rights of subjects. Following two centuries of genocide, some degrees of self-administration have been conceded through regional Lands Councils and the national Aboriginal and Torres Strait Islander Commission (ATSIC). Currently, however, the more that ATSIC, whose leadership is now elected by ATSI people, presents itself as an instrument of actual self-government, the more the federal government attempts to reduce its functions and funding. With that marginal exception, the constitutional structure of Australia consists of whitefella law.
That said, which whitefellas? Although the validity of the Australian legal order has clearly ceased to derive from British law, at least in that the High Court has found Australia to be an independent country and probably no Australian jurist would now make that derivation, the Australian constitutional structure is still confused by its imperial origins.

2.2.2. The Substantive Constitution

The substantive constitution (die Verfassung im materiellen Sinne⁴⁹) of Australia consists, at least, of: the Commonwealth Constitution, which is s 9 of the Commonwealth of Australia Constitution Act 1900 (UK); ss 1-8 (distinguished as the ‘covering clauses’) of the same Act; the Statute of Westminster 1931 (UK); the Statute of Westminster Adoption Act 1942 (Cth); the Australia Act 1986 (Cth) and the Australia Act 1986 (UK); the Royal Style and Titles Act 1973 (Cth); the constitutions of the States; and the Self-Government Acts of the Territories. As can be seen, this is an odd mixture of imperial and domestic legislation, composing the substantive constitution of a country that considers itself, and is considered by the former imperial power, to be independent.

2.2.3. The Formal Constitution

The formal constitution (die Verfassung im materiellen Sinne⁵⁰) of Australia, as has been mentioned, consists of a section of a British statute. However, the ‘covering clauses’ are also constitutionally important: in particular, cl 2 provides that the monarch of the United Kingdom shall also be monarch in Australia (although the Royal Style and Titles Act provides for a separate Australian title). The Statute of Westminster, which Australia reluctantly adopted after realising in 1942 that Britain could (or would) no longer provide a military umbrella, reduced Britain’s capacity to legislate with effect in Australia. That capacity (among other things) was finally abolished by the Australia Acts. The Australia Acts demonstrate the ambiguity of ultimate legal authority in Australia by being mirror legislation—almost identical Australian and British statutes. Nothing formally prevents either the British or the Australian parliament from repealing or amending its version of those statutes, or indeed the British parliament from repealing or amending the Commonwealth of Australia Constitution Act. However, the practical reality is that the British would not take such action and Australia would ignore it if they did.

2.2.4. Content of the Formal Constitution

More problematical in practical terms is the content of the Commonwealth Constitution.⁵¹ It sets out a federal structure, in which the federal legislative powers are listed⁵² and some of them are made exclusive, expressly or by implication, while most are concurrent with the States, whose legislative powers are plenary. In case of conflict, federal legislation prevails.⁵³ This document, however, is poorly designed. It is littered with expired clauses to the effect ‘until the Parliament otherwise provides’.⁵⁴ And its structure is obscure, especially after the first three chapters which appear to apply some version of a ‘separation of powers’.⁵⁵
Perhaps the greatest design difficulty, however, is one that compounds the other problems: this constitution is exceptionally difficult to amend. A proposed amendment must normally pass both houses of the federal parliament and be submitted to the people in a referendum, where it must receive a majority of all votes cast plus a majority in each of a majority of the six States.⁵⁶ Of the 44 proposals that have got as far as a referendum, only eight have succeeded and none since 1977.

2.2.5. The Formal Constitution and Australian Independence

A still deeper problem as to content is that this document was not designed to be the constitution of an independent country. What it initially achieved was a restructuring of British sovereignty in Australasia. New Zealand declined the option of joining (which remains open); Western Australia joined reluctantly and at the last minute, and in the 1930s attempted to secede. The original commitment to British sovereignty is marked in several ways. One of these is the position of the monarchy. The Australian monarch, the head of state, is chosen by a country that, as the same constitution is now interpreted, is a ‘foreign power’.⁵⁷ The monarch, who is rarely in Australia, is represented federally by a Governor-General⁵⁸ (and in each State by a Governor). The powers of the Governor-General, and powers that can be exercised personally by the monarch (which as of 1901 would effectively have been the British government), are substantial and are greater than those possessed by the monarch of the United Kingdom even in 1901. A ‘proposed law’ becomes statute on receiving the Queen’s⁵⁹ assent, which is given on her behalf by the Governor-General. However, the Governor-General may instead remit the bill to the parliament with suggestions for amendment or reserve the bill ‘for the Queen’s pleasure’. If a bill is so reserved, the Queen can take up to two years to assent. Even when a Governor-General has given the royal assent, the Queen may ‘disallow’ the legislation up to a year afterwards.⁶⁰ There is no reservation, in this regard, as to any particular type of bill—even a ‘money’ bill or a bill for an amendment referendum. These regal powers have almost never been used⁶¹: most constitutional experts are content to regard them as ‘dead letters’. But it seems to me a very bad practice to regard a constitutional provision as a dead letter, since then there is no limit to the range of letters that can be regarded as dead.
Nor are these letters entirely dead: they continue to enter into interpreting the federal constitution, so as to reinforce the monarchical principle against the democratic principle that the constitution also contains. The conflict between these two principles came to a head in 1975, on the issue of a Governor-General’s power to dismiss a Prime Minister. The then Prime Minister (whose position exists by way of constitutional convention) did not anticipate that a Governor-General would assume power to disregard a Prime Minister’s advice and dismiss a Prime Minister who still had the confidence of the lower and more democratic house of the parliament.⁶² The reverberations of that crisis have still not subsided: the immediately relevant constitutional provisions have not been amended, nor is there agreement on what could replace them.
Another mark of British sovereignty is that the Commonwealth Constitution contains no provision for an Australian citizenship: on the contrary, it assumes that citizenship in Australia will be British nationality. When in the 1940s Britain began to dilute the nationality ‘on which the sun never set’, along with other countries Australia did create its own citizenship.⁶³ But it is still difficult to see under what Commonwealth legislative power this statute can be valid. To find validity under the aliens power,⁶⁴ a power to legislate with respect to ‘Naturalization and aliens’, requires practically a reversal of the power’s originally intended meaning—a power to legislate with respect to the acquisition of British nationality through channels in Australia and with respect to the handling of people who were not British nationals.⁶⁵ Such a reversal would set an awkward precedent for interpretation of the Constitution in general—a self-issued licence for the High Court to reverse the meaning of any constitutional provision at all. To find validity under the occasionally implied ‘nationhood power’—a power to do things that are ordinarily within the legislative power of a sovereign nation—comes up against a severe problem of definition. It would be easy to understand the nationhood power so widely as to outflank the restriction of federal legislative power to certain topics and thereby almost destroy the federal structure to the benefit of central government. For this reason, the High Court has been reluctant to appeal to the nationhood power.⁶⁶ However, citizenship is not a States’ issue anyway.
The issue of the federal pattern also arises in interpretation of the power to legislate with respect to ‘External affairs’.⁶⁷ The word ‘external’ was preferred to the more usual ‘foreign’ so that the federal parliament could legislate on relations with other parts of the British Empire, which were not regarded as foreign. If Australia is now an independent country, that issue no longer arises—there are now, indeed, a department and a minister for ‘foreign affairs’. The constitutional word remains, however, ‘external’ and it has proved very difficult to define. For instance: if a matter is ‘external’ simply because it is the subject of a treaty to which Australia has become party, not only could the Australian government concoct a treaty with some friendly country in order to obtain power to legislate on a particular topic but, in any case, international conventions already cover a far greater range of topics than are listed among the heads of federal legislative power. Examples are such politically sensitive topics as the environment and racial or sexual discrimination.
To find that Britain is a ‘foreign power’ is, conversely, to assert that Australia is an independent country. But why is that so and, if it is so, when did it come about? The High Court found that it has been brought about, at the latest, by the Australia Act 1986 (Cth), but it argued from the provisions of the Act.⁶⁸ However, there may be a shorter way. The preamble to the Australian version of the Act refers to Australia as ‘a sovereign, independent and federal nation’. This is a statement made by the Australian federal parliament at the request and with the consent of the parliaments of all the States.⁶⁹ I think it can be read as a declaration of independence. Either a declaration (like that of the USA) by which independence was achieved on that date or, at least, a declaration that independence had been achieved by that date. That the statement appears only in a preamble would not matter in this regard. Nor is it a problem, but rather it is appropriate, that the British version of the Act has no counterpart. However, this Australian independence is hard to reconcile with continuing to permit a foreign power to appoint an absentee head of state.
The issue of independence or external sovereignty blurs into that of internal sovereignty. While there is judicial support for the proposition that, today, the authority of the Australian Constitution derives from the people,⁷⁰ that proposition is at odds with the historical evidence.⁷¹ The Constitution has never been endorsed by an actual majority of the Australian people: it was approved by referendum in 1899, although only 52% of those eligible to vote did so; in addition, few women or Indigenous people were able to vote and none had been involved in the drafting. Nor can it realistically be said that Australians now approve of their Constitution, since few of them are familiar with it. A 1987 survey found that 47 % of Australians were unaware of its existence. A 1994 survey found that only 18% had some understanding of what it contains; only 40% could correctly name the two houses of the federal parliament; younger Australians in particular appeared to have a better understanding of the US Constitution than of their own. The situation may have marginally improved following the Constitutional Convention of 1998, whose main business was to design a transition to a republic and in connection with which every household received a copy of the Constitution; but the focus of the Convention was narrow.⁷² Several reasons for this ignorance of the Constitution may be advanced: Australians’ general distrust of law and lawyers; the Australian etiquette of not discussing politics in a pub (the mainstay of grassroots democracy in Britain); the high level of immigration; and that, in the near total absence of constitutional rights, but with an efficient electoral system, the Constitution does not impinge on people’s daily lives.⁷³

2.2.6. Australia and a Bill of Rights

A different sort of constitutional confusion arises from the absence of a bill of rights from the Commonwealth and State constitutions as well as from the Self-Government Acts of the Territories. Indeed, these documents contain very few provisions as to individual rights. The restrictions that might have been in a bill of rights are found, instead, in explicit constitutional limitations (mainly on legislative powers), in implied constitutional principles, in statutes and in the common law. Except that any statutory provision must be constitutional, however, there is no limit to the repeal or amendment of a statute or to the extent to which statute can override the common law.
In the Commonwealth Constitution, the principal individual right is against deprivation of property by the Commonwealth without compensation on ‘just terms’.⁷⁴ There is no constitutional right to liberty or security of the person.⁷⁵ It has indeed been determined in the Federal Court (by a majority and with evident reluctance) that, in terms of the federal constitution, genocide would be legal.⁷⁶ There is some limit upon powers to restrict interstate travel or to discriminate on the ground of interstate residence, but it is unclear how far these extend.⁷⁷ The itinerant, non-white immigrant workers of 1901 are excluded when the rule against discrimination on the basis of interstate residence is confined to protecting a ‘subject of the Queen’,⁷⁸ whereas in today’s perspective the reasons for non-discrimination would extend to all legal residents whether citizens or not. The prohibition of legislation that would tend to ‘establish’ a religion or otherwise restrict the exercise of a religion,⁷⁹ largely copied from the Constitution of the USA, has never been successfully relied upon: it has not benefited conscientious objectors, it has been held to permit federal funding of religiously based schools, and it does not apply to the States. There is not even a constitutional right to vote.⁸⁰
These absences are not oversights: the framers of the federal constitution paid close attention to the Constitution of the USA, which does contain a bill of rights. An evident reason for this decision was a fear that such rights could be claimed by Indigenous peoples and by immigrant labour.⁸¹ For the same reason, while Australia was one of the earliest signatories to the Genocide Convention 1948 and even approved its ratification with a statute,⁸² it has yet to fulfil its Convention obligation to bring the Convention provisions into force in domestic law.⁸³ There have been several proposals for a federal bill of rights in Australia, but to incorporate one by constitutional amendment would be very difficult and it is unclear whether the federal constitution gives the federal parliament effective power to enact a statutory bill of rights.⁸⁴ No State has a bill of rights: the most recent State inquiry into the possibility of such a bill recommended against it, on the ground that it would give too much power to the interpreting judiciary.⁸⁵ Very little of the great English statutes on civil liberties, such as Magna Carta 1215 and the Bill of Rights 1689, is now regarded as law in Australia.⁸⁶ A judicial suggestion that common-law liberties could be read into the federal constitution was effectively met with the objection that one could not pick and choose or update: one would have to include all of the common-law provisions on civil liberties, as they stood in 1900 (with their gender bias and so on).

2.2.7. Constitutional Principles

More important in reality is the relatively recent move by the High Court to imply ‘principles’ into the ‘text and structure’ of the federal constitution.⁸⁷ The Court was impelled to this by the difficulty of developing the Constitution through amendment. The main principle that has been implied is a ‘freedom of political communication’: legislation or executive action will be invalid if it impedes communication, among individuals or between individuals and their political representatives, on a political matter; the governing criterion is the adequate functioning of the political system laid down in the federal and State constitutions.⁸⁸ It has also been judicially suggested that freedom of political communication can involve freedom of movement and of assembly.
The High Court has, however, been cautious about abandoning an originalist approach to constitutional interpretation. Both the Court as a whole and individual judges have wavered among originalist, literalist and progressivist approaches.⁸⁹ The Court’s problem is that the intentions of the framers form perhaps the only clearly legitimate interpretive source, yet those intentions are frequently unclear and, as time goes by, even where they are clear they may cease to match contemporary values. That is plainly the position as to gender and ‘race’.
The continuing weakness of civil liberties is however demonstrated by the continual failure of litigation to redress the oppression of Indigenous peoples. The High Court has rejected claims by refusing to consider any form of Indigenous sovereignty⁹⁰ or to question expressions of benign intent by legislators of decades past⁹¹ and by declining to decide that the legislative ‘race’ power can be used only to benefit the affected ‘race’.⁹² Although several hundred cases are being prepared, particularly with regard to compensation for the forcible or fraudulent removal of part-Indigenous children from Indigenous parents,⁹³ with the passing of time the victims face severe problems of evidence.⁹⁴

2.2.8. Proportionality

A different type of potentially protective implication is the High Court’s introduction, at both the constitutional and the administrative levels, of the concept of ‘proportionality’ or, in the alternative American language that the Court also uses, of being ‘appropriate and adapted’ to the purpose.
An example of this concept’s potential for protection is that in 2001 the Australian federal government introduced, and almost succeeded in getting passed, legislation that would have permitted duly authorised officials to tow a boat crowded with refugees out to sea and leave it there, even though it was obviously unseaworthy.⁹⁵ The legislation would have been entirely within the constitutional power to legislate with respect to ‘Immigration and emigration’,⁹⁶ unless the High Court had determined that to authorise such an action was disproportionate to the power. That ruling might been given after people had drowned and would then have left the Prime Minister and the officials purportedly authorised by him vulnerable to criminal and civil sanctions.
So far, at least, the concept of proportionality is applied only to those federal legislative powers that are classified as ‘purpose’ or ‘purposive’ powers and not as ‘subject matter’ powers, but that is itself a shaky distinction.⁹⁷ However, the Court has drawn the concept of proportionality from German law (Verhältnismässigkeit) without, it seems, much attention to the need always to be able to state clearly to what the action is proportionate or disproportionate.⁹⁸ It is one thing to introduce this concept where the constitution contains a bill of rights, as the German constitution does⁹⁹; but to introduce it where the relatum will be a judicial invention is a considerable arrogation of judicial power. More generally, it is a new dimension of obscurity.
There are other respects in which the Australian Constitution is out of date, particularly on the fiscal plane. My concern here, however, is the lack of constitutional structure. The longer that Australia has continued with a Constitution that is both increasingly out of date and very difficult to amend, the more the interpretation of that document has passed into an inward spiral of analogy—although that spiral may already (in High Court cases of 1997-98) have exhausted its capacity for innovation. The norms set out in the Constitution dissolve into the normative patterns offered in a plethora of interpreting judgements.

3. Reconceiving Law

There is, then, a lack of structure, in common-law systems generally and for particular additional reasons in Australian law. There is not, of course, a total lack, either administratively or in terms of legal order. Moreover, any legal system requires a degree of disorderliness if it is to cope with the hazards of social behaviour.¹⁰⁰ However, in terms of legal order, in common-law systems the degree of order is too low to permit a scientifically acceptable account of the order in its own terms. The case of Australian law intensifies this point: there can be such a low degree of order even in a country that has a formal constitution. The case of the United Kingdom might be even stronger.
It seems advisable to consider a concept of law that might dispense with assuming a high degree of order within each legal system. This concept will nonetheless have to permit a sufficiently high degree of order within legal science. For the case of a legal system that exhibits a low degree of order, this will need to be a concept that does not lead to an account of the legal order in its own terms. The path to such an external account will necessarily address the issue of closure.

3.1. The Meaning of ‘Law’ (1)

3.1.1. The Western Legal Tradition

The discipline of comparative law is a product of the western legal tradition. This tradition embraces the legal systems of western Europe, including English ‘common law’, as well as their imposed or voluntarily made copies elsewhere. It also contains the more global systems of law known as public international law and maritime law. Each of these systems, as well as the tradition as a whole, is accompanied by scholarly commentary, termed legal ‘doctrine’. It is within legal doctrine that the whole is defined as a tradition.
According to legal doctrine, the systems that are considered to compose this ‘tradition’, and especially their accumulations of doctrine, are said to derive their basic intellectual framework from ‘Roman law’.¹⁰¹ What is now known as ‘Roman’ law, however, was written after the fall of Rome. It consists almost entirely of compilations made in fifth and sixth century Byzantium: the fifth century Code of Theodosius and the works commissioned in the sixth century by Justinian which have been known since at least the sixteenth century as the Corpus Iuris Civilis. The extracts included in those compilations were edited—and we do not know to what extent, because almost all of the texts drawn upon have since been lost. In any case, some of those texts had themselves passed through several editions. What is now known as ‘Roman law’, then, is in substance Byzantine law of late antiquity. Especially, it is codified and, at least in commitment, it is Christian.

3.1.2. Basic Types of Law

This tradition divides ‘law’ into three basic types. In order to list these types comparably, I shall standardise expression through the concept ‘norm’. The tradition operates in numerous languages—I will provide English, Latin, French and German equivalents as typical—but as to the labels on basic concepts the translation-equivalences are now largely settled.¹⁰² The difficulties here lie much less on the level of the signifier than on that of the signified.
To list each basic type of law under its usual labels:

• Natural law (ius naturale, droit naturel, Naturrecht): binding norms discoverable in the structure of the world, particularly in ‘human nature’, either as simply present in the world or as laid down for it by ‘God’.
• Positive law (ius positivum, droit positif, positives Recht): binding norms laid down (‘posited’) by a ruling person or group.
• Customary law (mos or consuetudo, droit coutumier, Gewohnheitsrecht): binding norms arising in and sustained by customs (long-established patterns of social behaviour). The customs may be those of an occupation, an institution, a locality, a nation or the international community.

In relation to each of the above labels, the definitions given are of typical usages: there are other usages and some of those are common.¹⁰³ The important issue here is that the histories of the signifiers and the signifieds are different.

3.1.3. ‘Legal Positivism’

A westerner who speaks today of ‘law’ without qualification will be taken to refer to positive law. The signifier ‘positive’, however, is not very informative. It derives from Latin positivus,¹⁰⁴ an adjective derived from the verb ponere (to put, set, lay or lay down). The adjective, however is not a classical expression, but is first found in late antiquity (second to sixth centuries). In that period positivus had two meanings: (1) not belonging to nature but artificial; and (2) a grammatical root. During the Middle Ages it acquired two more meanings: (3) existing or affirmative, the opposite of ‘negative (negativus)’; and (4) materially real, as opposed to the merely thought or imputed. Its later career—in philosophy, legal theory, theology and the physical and social sciences—develops all of these meanings. The notion of positivity is first found in moral terminology around 400, in the expression ‘positive justice (iustitia positiva)’,¹⁰⁵ but the expression ‘positive law (ius positivum)’ is not found until the twelfth century. The first known use of that expression is by Abelard,¹⁰⁶ although he appears to treat it as familiar.¹⁰⁷ Later in the century it is found in more mainstream scholars, including a cardinal, and by about 1260 it became, as Kuttner puts it, ‘quasi-officially accepted’.¹⁰⁸ Later still, the character of human positive law as a product of will would be extended into the divine sphere. Grotius distinguishes from ‘natural law’ a category of ‘positive law’, characterised by its being ‘law which is immediately derived from the will of the legislator’. This category is then subdivided into ‘divine’ and ‘human’, according to whether the will in question is that of ‘God’ or of a human ruler.¹⁰⁹
Both the concept and the name ‘positive law’ thus originate and spend most of their career within the framework of Christianity. The modern expression ‘legal positivism’ consists of accepting both concept and name—although, beyond this, the label has been used to mean such a variety of things that it has been suggested we would be better off without it.¹¹⁰ Such dismissal, however, throws out the ambiguity as a potential resource: rather, acute and persistent ambiguity can be read as a symptom. The dismissal may also be too quick, for broadly one can say that ‘legal positivism’ is two things: (1) the view that there is no natural law, i.e. that all law is positive law (leaving customary law in the ‘too hard’ basket); (2) the conduct of legal science according to the tenets of ‘positivism’ as that word is understood in philosophy and social science.¹¹¹ The issue I wish to pose is whether these two views are actually compatible.
Outside legal science, to be a ‘positivist’ is, in the first place, to take as paradigmatic for any discipline the assumptions and methods of the physical sciences, queened by physics.¹¹² For the present purpose, I will call this outlook ‘philosophical positivism’.¹¹³ It belongs to the ‘positivist’ movement in philosophy and the social sciences that flourished during the nineteenth and twentieth centuries.¹¹⁴ However, whatever its success for physical science, philosophical positivism cannot be transferred wholesale into the social sciences: for one thing, the physical sciences deal with physical fact to the exclusion of meaning; for another, there are ethical limits to experimenting on human beings. Given these qualitative differences, the attempt to reproduce in social science as much as possible of the methods and assumptions of physical science places a premium upon those assumptions. When one’s focus is on social science alone, the label ‘positivism’ comes to refer just to those assumptions, as philosophical presuppositions for the possibility of social science.
Those assumptions are threefold:
(1) that reality is basically material, apprehended through sensuous experience; accordingly,
(2) that statements of what is and of what ought to be are different in kind; and, accordingly,
(3) that no statement of the one kind can follow from a statement of the other kind.
In this sequence, the first assumption entails the second and the second entails the third.
When legal theory proceeds upon the three assumptions of philosophical positivism:

• Natural Law. The concept of natural law is a quick victim on all three counts. It is found (1 above) to assume a non-sensuous (i.e. metaphysical) reality, in relation to which (2 above) statements of what is and of what ought to be are not entirely different, so that (3 above) a statement of what ought to be can be derived from a statement of what is.
• Positive Law.
  • Concepts of positive law that assume a suprahuman, personified state go the same way.
  • Yet the concept of positive law is allowed to survive. It is held (1 above) to satisfy the first assumption, in that the very ‘positivity’ of positive law—the human act of ‘positing’—belongs to sensuous experience, i.e. is ‘positive’ also in the philosophical sense. It then seems to be supposed (2-3 above) that the second and third assumptions are satisfied by a science of law in which description (‘is’) and evaluation (‘ought’) of positive law are at least logically segregated.
• Customary Law. The concept of customary law might satisfy (1 above) the assumption as to sensuous experience, yet it is condemned for purporting to (2-3 above) derive a statement of what ought to be from a statement of what is—that is, to derive a norm from a fact of observed behaviour.

3.1.4. Kelsen’s Problem

This survival of the concept of positive law is, however, unsustainable. It appears to have met its come-uppance in the fate of arguably the most sophisticated form of legal positivism, Kelsen’s ‘pure theory of law’.¹¹⁵ For Kelsen, positive law consists of norms and a ‘norm’ is ‘the meaning of an act of will’. The difference between a non-legal meaning of an act of will and a legal norm is that the former is a ‘subjective’ meaning, dependent on the individual’s recognition of it as binding, while the latter is an ‘objective’ meaning, binding independently of the individual’s recognition or even knowledge. What is the condition for this ‘objectivity’? Some have argued, for instance, that a norm that is the meaning of an act of will of a personified state enjoys the ‘objectivity’ of its suprahuman author. But Kelsen rejects such mundane metaphysics. Nonetheless, he does not deny the legal norm’s appearance of objectivity. Instead, he tries to construct within legal science a perspective in which that appearance can be accepted consistently with the tenets of philosophical positivism.
For half a century he maintained that one could simply presuppose, as a matter of transcendental logic, that it is so. Eventually, however, he accepted that this presupposition could at best be a ‘fiction’ in a tenuous sense. More plausible to my mind is a diagnosis that this dilution (at best) of Kelsen’s theory marks an overall failure of legal positivism: a point at which the concept of positive law fails the tests of philosophical positivism. That point is the second blow of the anti-metaphysical hammer, in which the concept of positive law is deprived of any metaphysical foundation. Once its appearance of objectivity cannot be sustained, it hardly matters whether one also takes the is/ought cleaver to it. That was only to be expected of a concept that survived from scholasticism. Nor is the concept saved, because it is not altered, by its amplification through the addition of historical or social ‘contextualisation’.¹¹⁶ In short: the concept of positive law is ultimately incompatible with the tenets of philosophical positivism, in the same ways as the concept of natural law.

3.1.5. Customary Law

The concept of customary law has never made sense from a philosophical-positivist standpoint and whatever sense it may have made in earlier times might now be undiscoverable.¹¹⁷
A clue to the difficulty, however, may lie in the history of the concept. The expression ‘customary law (droit coutumier, Gewohnheitsrecht)’ is relatively recent: it does not seem to have been in general juristic use before the nineteenth century. It appears to name an independent type of law. Historically, however, it is a renaming of the ‘custom’ element in the much older couple ‘law and custom’.¹¹⁸ Three characteristics of that element stand out.
First: the context tends to be imperial. A contrast between statute and custom (although not the very expression ‘law and custom’) first became salient on the level of theory—or, at least, of classification—in the Institutes of Justinian.¹¹⁹ In the history of the common law, it appears in England with the securing of Norman rule.¹²⁰
Second: the ‘custom’ element derives no content from within the imperial legal order. It is a null category, to which any content can be given that may be found in the metropolis, in the provinces or—especially—in territories that are to be conquered. The Other norms will be characterised as ‘custom’ because their bearers are subordinate, irrespective of how they would themselves characterise their norms.
Third: the universal components of the concept ‘custom’—as an element of this couple—tend to be contraries or negations of the salient characteristics attributed to ‘law’. Already with Justinian, there is a subtle and far from innocent mixing of contrariety and negation.¹²¹ In the Digest, custom (mores or consuetudo) appears as a popular creation, the contrary of creation by—mainly imperial—authority. Thus far, the former could be a rival to the latter.¹²² In the same breath, however, the creation of a custom is characterised as ‘tacit’. The creation of a law, evidently, is explicit. Formally, this is another contrariety and the tacit could be as valuable as the explicit. But—short of a deep irrationalism, which not a Justinianian option—that is obviously not so. This move in favour of authority is underpinned by negation: a law (lex) is ‘written law (ius scriptum)’ while custom is ‘unwritten law (ius non scriptum)’. Given the logical priority of the positive over its negation, this provides laws with a logical priority over customs. And that move is made in the Institutes, a textbook through which the student of law would enter the juristic world. The negation thus conditions the ‘young enthusiasts for law’,¹²³ training to become imperial officials, to assume the priority of norms created by authority, before they set eyes on the risky contrarieties. Later writers, including both jurists and anthropologists, would add still more emphatic negations into the ‘custom’ element: that it is involuntary, indeed unconscious and the following of custom even a form of automatism.¹²⁴
The nature of the ‘custom’ element may therefore be discoverable through a double reflection. On one side, by reference to imperial encounters and the rôle played in them by the couple ‘law and custom’. And, on the other, by reference to that of which it is the contrary or negation, the ‘law’ element. Closure in the ‘custom’ element is then presumed to derive from closure in the ‘law’ element, while the imperial context may provide reasons for the extension of legal closure into the idea of custom.
There are other uses, even of the couple ‘law and custom’, that do not fit this model: customary international law, the image of the common law as custom, commercial and professional customs, curial customs.¹²⁵ However, this is the model that appears to dominate when ‘custom’ is located as a type of law (ius, droit, Recht).
The inclusion of custom within a legal order may contribute to the element of disorder that any legal system requires. The claim that common law as such is customary amounts, however, to admitting a low degree of order in the system as a whole.

3.2. The Meaning of Law (2)

Back to square one, but let us change the board. Instead of proceeding from the idealist heritage, or in fondly believing that with ‘legal positivism’ one has escaped from it, one can take philosophical positivism seriously and attempt to understand law as a material reality. One can ask: what type of social behaviour is socially called ‘law’? The existing body of legal theory will stand as the most sophisticated expression of that nomenclature.¹²⁶

3.2.1. Encountering Law

A matter of ‘structure’ is first of all a matter of physical reality. To say this is to commit a banality, but it is of the first importance here. Also banal but important here: a physical structure is a process, composed of events. It is a patterned process: on the physical plane, a regularly occurring process is a system.
The physical reality of meanings is their (literal) embodiment in human social behaviour. Behaviour, as Weber says, is social when it is behaviour to which the actors collectively attach meanings.¹²⁷ Some of those meanings are descriptive, others prescriptive. With Kelsen, let us next distinguish the prescriptive meanings into those that appear as subjective and those that appear as objective. This brings us already to his central and frustrating issue: under what conditions does a prescriptive meaning appear as objective?
Kelsen’s starting point is the God-State: the prescriptive meaning that is a legal norm is binding because it is objective and it is objective because it is endowed with the objectivity of its creator. Once Kelsen rejected the idea of the God-State as metaphysical, his first option was to suppose that the illusory objectivity of the God-State was communicated to the legal norm. This would entail that the apparent bindingness of the legal norm was also illusory. Kelsen cursorily rejects this option and prefers to struggle, in the end unsuccessfully, to give an account of legal bindingness in which it will appear as a truth.
His mistake, I want to suggest, is that he failed to grasp meaning as something that is physically embodied, hence as something that is encountered. He neglects the phenomenon, the event, of encounter—the meaning embodied as actual speech-act (oral or written). It is an easy error, into which one can be seduced by the intellectual accessibility of the printed text.¹²⁸ The text, however, is also something physically encountered—as much as the person whose garb proclaims them to be a judge.¹²⁹ Or, to take a common but radical experience: at the line in front of an immigration officer’s desk, one physically encounters the national legal order. Thus one encounters the persons, places and things of law. That, indeed, is evident. Yet they are constituted as persons, places and things of law by legal meanings and a strange characteristic of legal meanings themselves is that they appear not to be encountered. Instead, they appear as always-already present and oneself as always-already in their presence. They appear not as specifically embodied but, rather, as environing. They efface the moment of the encounter with them.¹³⁰ And they appear to do this all by themselves, without external support. I will sketch a picture of the legal norm which will provide an account of this apparent erasure. This picture will be proposed as an empirical hypothesis: in the order of sociology or socio-linguistics.
The key factor, I think, is that this apparent erasure appears not only in the content of a legal norm, through which it constitutes a person, place or thing as ‘legal’, but also in its form, through which it purports to obligate. A norm that is socially named ‘legal’ has the distinguishing characteristic that it is assumed to obligate unconditionally and to do so simply because it is law. The obligation imposed by a legal norm is conditioned, if at all, only by another legal norm: at some point, there can be a final decision and the prospect of that decision is present from the beginning. To encounter one legal norm is to encounter it as an aspect of the whole legal order: the finality present in the whole is also, in that sense, present in the part. Nor does this finality come from outside the legal order: it does not derive from the particular norm or the whole order being (for example) just or rational or even effective. As to justice, indeed, the sole intrinsic relation between law and justice is that every independent legal order claims to provide the ultimate criteria of justice. Nor does this ‘legal’ character have a necessary linguistic form: it is sufficient that the meaning could be expressed as an (objective) ‘ought’; it might even, in fact, be expressed as an ‘is’.¹³¹
The starkest manifestation of this immediate finality is the police officer’s command: the officer simply does not take ‘no’ for an answer. Any possibility of the legitimacy of a contrary norm is always-already excluded. Yet nothing outside the legal norm seems to perform this exclusion.¹³² What is going on?

3.2.2. Six Tools for a Double Shift

What I am going to do, generally, is to execute a double shift in the focus of defining ‘legal’ character: (1) from the signified to the signifier; and (2), as to the signifier, from the task of nominal definition to that of socio-linguistic description.
To effect these shifts, I will need some special tools:

• Contrariety or Opposition. The first is to hand: it is called ‘contrariety (contrariété)’ or ‘opposition (opposition)’.¹³³ The figure operating when, for example, the officer does not take ‘no’ for an answer does not seem to be that of negation. For negation, as ordinarily understood, preserves the positive in negative form: not-P is always not-P. But the democrat who opposes a monarchy does not prefer just a non-monarchy. The preference is, rather, for an opposite sort of régime. Democracy is not the negation but the contrary or opposite of monarchy, as for example sunset is not the negation but the contrary or opposite of sunrise and retreat is not the negation but the contrary or opposite of advance.¹³⁴ Likewise, if there is a norm ‘All vehicles must be driven on the left-hand side of a road’ and I drive on the right-hand side, I will not simply be driving ‘not on the left-hand side’, because I could do that by driving in the middle. Rather, since ‘right’ is not the negation of ‘left’ but its contrary or opposite, I will be doing the contrary or opposite of what the norm prescribes. Or, if there is another norm ‘All vehicles must be driven on the right-hand side of a road’, the conflict between these two norms will not be one of contradiction but one of contrariety or opposition.¹³⁵ This figure is by no means mysterious: it is the theme of a child’s game of ‘opposites’; my point is that the figure of negation is inadequate here.
• Absentation. The second, I shall call ‘absentation (absentation)’. When the officer does not take ‘no’ for an answer, the possibility of a legitimate refusal is not rejected: it is simply not entertained at all. It is not preserved in any qualified form, but is rendered absent. As well as presence, there is absence; as well as presentation, absentation. Within the legal discourse, the alternative as an object of knowledge is simply not there.¹³⁶ And the encountering subject who might wish to present an alternative is, in that respect, silenced.¹³⁷
• Closure. Unconditional absentation, I shall call ‘closure (clôture)’. I shall distinguish two forms of closure: ‘weak closure (clôture faible)’ and ‘strong closure (clôture forte)’. By ‘weak closure’ I will understand simple exclusion: the Other as subject is simply absented; heard, of course, but not acknowledged, not listened to at all. By ‘strong closure’ I will understand a more subtle figure. The Other as subject is absented as such and, in the same moment, re-presented in terms of the closing discourse. The bearer of the Other meaning may then be shunned, condemned or subjected to therapy—even (totally conditionally) ‘loved’.¹³⁸
• Tacit knowledge. Knowledge can be explicit or tacit (connaissance/savoir tacite). The same knowledge can be sometimes tacit and sometimes made explicit: for example, I can discuss the rules of grammar explicitly, but if I tried to make them explicit all the time that I apply them I would be tongue-tied. Tacit knowledge is the ordinary form of an intellectual skill.¹³⁹ The knowledge involved in a closure, however, must remain tacit or what is absented would be made present in being made explicit. That is, in the case of a strong closure, what is re-presented as other than itself would then be revealed as itself. How, however, could knowledge that must remain tacit be communicated? Searle proposes that any tacit knowledge can be communicated through imitation of its application: there can be an osmosis in which the repetition conveys its own necessary presuppositions.¹⁴⁰
• Irrelation. In an encounter with closure, the relation between the encounterer and the encountered is far from equal. Some meaning that the closer attributes (let us presume correctly) to the Other is either entirely absent from the closer’s explicit knowledge or re-presented there in a form that the Other might not accept. I shall call this lopsided sort of relation an ‘irrelation (irrelation)’.
• Dark Performative. I will identify the mechanism of closure on the plane of the signifier, as a particular type of locution that I will call a ‘dark performative (performatif sombre)’. My model is Austin’s theory of ‘performatives’, as developed by Searle. Austin finds that certain locutions do not simply denote or connote, but actually constitute: for example, to say ‘I bet you sixpence it will rain tomorrow’ is to make a bet.¹⁴¹ Searle develops this concept.¹⁴² He calls a meaning that has this sort of objectivity an ‘institutional fact’, distinct from a physical or ‘brute’ fact. For Searle, all linguistic usage is ‘rule-governed’. The rules may be ‘regulative’ or ‘constitutive’: ‘regulative rules regulate antecedently or independently existing forms of behavior’ while ‘constitutive rules do not merely regulate, they create or define new forms of behavior’. The rules of a game, for example, do not merely regulate a game when it is played but ‘create the very possibility of playing’ the game.¹⁴³ A system of constitutive rules is an ‘institution’. An activity so constituted is not mere behaviour, a merely material or ‘brute’ fact, nor it is something merely subjective: it has an objective existence, as an ‘institutional’ fact.¹⁴⁴ Now, this sort of performative constituted by simply presenting. What I want to suggest is that there is another sort of performative, which constitutes by re-presenting, in the manner of a strong closure.¹⁴⁵

I am seeking a position beyond that reached in the literature on ideology and alienation, which tends to identify them as sets of effects, which are traced to causes without explaining the process by which those causes produce those effects. Greater attention to process is found in the literature on exclusion and silencing, but without identifying the mechanism by which these processes are carried out.

3.2.3. ‘Law’ as Dark Performative

I have some other tools that could tie up some loose ends that may emerge,¹⁴⁶ but the above will do the present job of introducing the idea of the word ‘law’ as darkly performative.
Perhaps the most powerful of dark performatives is a locution that includes the word ‘God’. I put a moral view to you. You oppose it. I then say ‘but God’, attributing that view to ‘God’ as its author. With that attribution, all possibility of your opposition being legitimate is excluded. More, since you are opposing the morality laid down by ‘God’, you are (re-presented as being) in a ‘state of sin’. This happens without my explaining the meaning of ‘God’. That might be done as supplement: more likely, however, I will declare that beyond a few pointers the nature of ‘God’ cannot be explained. The locution by itself already effects the closure. It is the same with locutions involving the word ‘faith’. I would say ‘Well, I disagree but for me this is a matter of faith’. Introducing the word ‘faith’ makes the difference between what Kelsen would call a ‘subjective’ and an ‘objective’ meaning. The view is now ‘objective’ as an ‘article of faith’. I do not have to explain what I mean by ‘faith’—indeed, I am unlikely to be able to do so and may well declare that it is inexplicable. The word ‘evolution’ seems often to have been used similarly. Probably, no word always operates as an element of a dark performative or of any other kind of performative. But it does seem that some words, such as ‘God’ or ‘faith’, usually do so.¹⁴⁷
One can discuss these locutions, without falling under their spell, by not using but mentioning them.¹⁴⁸ That discussion may lead to the construction of an alternative, non-performative or at least not darkly performative use.
I want to suggest – and the earlier comments on legal positivism ease the path—that the word ‘law’ operates in this way. A prescriptive statement that is, in Kelsen’s terms, a subjective meaning is converted into an objective meaning through characterising it as ‘law’ by way of dark performance. Not every use of the word ‘law’ will have this effect: I might use the word in the sense of the physical sciences. Nor, if there is this effect, will it always create obligation: I might, as historian or comparatist, refer to a norm as a norm of ‘Roman law’ or of ‘French law’; but in these cases I will be engaging in an oblique sort of dark performance, since I will be saying that the norm has been obligatory (in Justinian’s Byzantium) or is now obligatory for others (in Paris today). On the other hand, the same locution can be carried out with equivalent language: such as characterising a norm as ‘legal’, its site as a ‘statute’ or its author as a ‘judge’. Further, as Austin emphasised, performance is not a matter of single words or phrases, or even of language alone: it is a speech act, a social event with a context. That said, there seems no obvious barrier to the occurrence of performatives, light (clair) or dark, in artificial languages or through non-verbal symbols. I shall also leave aside here the possibility that physical symbols might perform directly and not only as symbolising a prior performative utterance.
Once legal character is located in dark performance, one loosens the assumptive connection between specific language and the signified. There is room for the ‘normative patterns’ that I identified in common-law thought.
With a dark performance, the excluded and the actual process of exclusion must remain tacit. Here may lie an explanation of the paradox that puzzled Hart, that most people seem to know what they mean by ‘law’ yet nobody seems able to say.¹⁴⁹ This explanation also has the advantage that the process it identifies is simple, albeit obscure. For the meaning of ‘law’ would have to be simple, comprehensible to the whole population, or no legal system could function.
The unconditionality of the legal norm would be subverted if the norm were accompanied by description that was open to registering the standpoint of someone who would prefer to follow a contrary norm. Instead, description has to be internal to the norm: the world has to be descriptively re-presented solely as contents of legal norms. This re-presentation would be a legal ideology. The first casualty of such a closure is the scientific principle of alterity.¹⁵⁰
These moves do not belong simply to the order of logic, since they do not entirely follow rules of thought that can be made explicit. Nor, however, do they belong to the order of rhetoric, since there is no attempt to persuade. Perhaps they belong to a third order or perhaps one needs to understand both logic and rhetoric as currently conditioned by closure.
In this account of the nature of ‘law’, legal norms or normative patterns have a double reality as materially embodied meanings: (1) as norms or normative patterns and (2) as illusive beliefs. The task of critique is not only to reveal and dismiss illusive beliefs but in addition to explain their origin and persistence.
This understanding of a norm or normative pattern applies to the concept of natural law as it does to that of positive law. Both natural law and positive law exist as human norms or normative patterns that are called ‘legal’ and with this sort of ideological accompaniment. Potentially it also applies to the concept of customary law, but there is the prior difficulty of understanding any type of ‘ought’ as customary. However, it seems likely that many bodies of ‘customary law’ would turn out, on closer examination, to be grounded in religious beliefs that colonial régimes have been unwilling or unable to recognise.
It could be argued that this includes too much. However, the criterion of effectiveness can be added in, not as a criterion for the existence of law but as a test of what it is relatively important to study. Then one would leave out of account the man with the sandwich board commanding that one ‘Abstain from Beans’.¹⁵¹ But one would still study natural legal orders, because the question of coercion is not about the intrinsic nature of a statement or action or about its potential effect upon oneself but about whether it is capable of having a coercive effect upon anybody. For the appropriate believer, there can be a real choice between facing either bullets or brimstone.
Finally, to show that legal character lies in the form of a norm is not to show that it does not also lie in the links between norms. Kelsen’s image of dynamic legal order, of a positive legal order as primarily a chain of authorisation, suggests strongly that it does.¹⁵² The answer, I think, is that authorisation does not necessarily involve closure. A sports club committee authorises its secretary to record membership, its treasurer to manage the funds and the coach to do just about whatever it takes to win the cup; none of that involves closure. If there is closure in links of authorisation between legal norms, it derives from the legal character of the norms themselves. The position is even clearer in the case of what Kelsen calls static legal order, in which the links between norms are deductive. Unless any deductive link at all is to be classified as closed (and I shall pass on that issue), closure in links of deduction between legal norms derives from the legal character of the norms themselves.

4. Conclusion

I shall conclude on two planes: (1) to specify the hypothesis about law further with reference to Australia; and (2) to reprise the initial issues in the light of the general hypothesis.

4.1. ‘Australians all’¹⁵³

I have located the key to legal character on the plane of the signifier and especially in the word ‘law’, which is to locate it within the form of a legal norm.¹⁵⁴ I have suggested that this is the mechanism by which legal closure is produced. The closure involved in a particular legal norm would not, however, be confined to the form of the norm but would be tailored to its content.
I will not examine individual norms here, but some categories of exclusion have already emerged. These are the closure of the Australian Constitution to the ‘voices’ of women, Indigenous peoples and immigrant workers. Those voices are simply absent. Women do not appear in the constitution at all. Indigenous people and immigrant workers are not only voiceless, absent as subjects, but present as objects of legitimated repression. They are singled out for special control through the power to legislate with respect to any ‘race’. Non-naturalised immigrants are excluded from protection given to a ‘subject of the Queen’. And a commitment to economic expropriation is manifest in the constitution’s protection of property but not of the person. Moreover, consistently with the general absence of alterity in legal scholarship, the voices of women and minorities are not heard nearly as loudly in scholarship on the constitution as are the voices of majority males.¹⁵⁵
If in the transition to independence the Australian legal order may have become multivocal by accident, registering British and specifically Australian voices, this multivocity has yet to extend to women and minorities. Given the difficulty of constitutional amendment, it is also very unlikely to do so unless the present federal constitution is replaced. How could it be replaced? One could take a tip from a certain manufacturer: just do it.¹⁵⁶

4.2. Approaches in Comparative Law (reprise)

In these terms I return to the initial issues. The four issues listed by Pfersmann, and his explanation of them in terms of an ‘illusion of the legislator-jurist’, were addressed in my first section; I add now that, in the approach I am proposing, these issues cannot arise.
In the succeeding sections I addressed his issue of ‘closure’, agreeing with him that law is marked by closure. However, I found the closure not in relations between norms but in norms themselves—not in their content but in their form as legal norms and not in the signified but in the signifier.
Those theoretical shifts, I suggested, had the particular benefit of facilitating an account of a common-law system. More broadly, they provided a different way of understanding the identity of ‘law’ as understood throughout the western legal tradition. For the global typology required in comparative law, this poses the question: are other traditions’ institutions that the comparatist considers to be ‘legal’ closed in the same way, closed in other ways or not closed at all? (In what ways does each legal ‘family’ seek its happiness?) This also becomes a question for general theory of law, both so far as comparison is involved in generalisation and so far as the question can also be asked about other normative institutions within western culture.

4.2.1. Epistemology and Ontology

There remain from Pfersmann’s analysis a claim and a preference. I accept his claim that comparative law has continued to adhere to a weak epistemology or ontology and have identified that epistemology or ontology as an idealism that pervades legal science. I also agree with his preference for a ‘constitutive’ alternative, although he is coy about how one would be ‘constitutive’. Perhaps, consistently with the claim, he is hinting that comparative law has yet to catch up with the Copernican revolution of Kant—that our knowledge does not conform to things but things to our knowledge.¹⁵⁷ If so, I think that the point could be made about legal science in general. In a more ‘history of ideas’ idiom, it might be said that legal science has passed through the Renaissance on the side of ‘ought’ but not on that of ‘is’—that it has humanised its values but not yet its construction of reality.
This perspective does not, however, require an immediate discussion of epistemological and ontological questions or, consequently, entail a scientific paralysis until those questions can be resolved. In scientific practice, propositions of epistemology and ontology are not preconditions but supplements. They are brought in after useful tools have been basically designed, to refine the design of each tool and of the set, and to address discrepancies of fit between those tools and tools of an earlier design. All the same, I acknowledge that the philosophical resonances of my little toolset are suspiciously eclectic; I have even drawn heavily upon one philosophical work whose resonances are themselves suspiciously eclectic.¹⁵⁸ I need to develop my own philosophical approach.¹⁵⁹

4.2.2. Is and Ought

Questions of epistemology and ontology are, at least to begin with, questions about what is. On every level of a science, however, there are also practical issues. I shall bracket here the question of the relation between ‘is’ and ‘ought’, except to say that at least a logical segregation of them seems to follow from treating meanings that are one’s subject matter as, in the first place, somebody else’s. If they are somebody else’s, it does not follow that they ought to become one’s own. That seems to be the case, whether the meanings themselves are descriptive or prescriptive.
That said, there are the questions of the practical value of legal science and, within it, of particular legal disciplines such as comparative law. To these questions one could give the limited answer that legal science describes law, that accurate describing is a benefit to humanity and that whatever humanity might do with a particular description—be that picture flattering, unflattering or nothing special to look at—is an extra-scientific issue. Then the expression ‘comparative law’ would have the same sort of meaning as ‘comparative religion’. One compares religions for the purpose of understanding them better, but not with the aim of adopting any of them, not necessarily with the aim of benefiting any of them and certainly not with a prospect of adopting all of them at once.
A philosophical positivist as radical as Kelsen would say that a science has to stop there because reason is only theoretical and not practical. But I do not share that view. For one thing, when one observes the operation of legal systems that view becomes counter-intuitive. For another, arguments that ‘ought’ statements cannot be rational may be equally applicable to descriptive statements, if those are understood (in a constructivist way) to have the meaning ‘ought to be considered the case’. Then no statement would be rational—including this one. So I shall prefer to say that legal science can be both descriptive and prescriptive. I shall also accept that it ought to be both. That is to say: legal science ought to be practical with regard to law.
But that is not to decide whether legal science should operate for or against law. This will be a subsequent decision: especially, it will no longer be obvious that law reform is the practical purpose of legal science, including the application of comparative methods¹⁶⁰ in law reform. And that decision will be made with regard to the description given.

4.2.3. Closure and Legal Science

I have offered a description of law in the western legal tradition: that it is marked by normative closure. I would go so far as to say that normative closure is its distinguishing mark. I have also implied and would suppose that, at least generally, closure is an undesirable thing. Would it be consistent with these views, to engage in law reform?
I shall assume that the question of law reform concerns legislation. It may be that particular reforms, or in a common-law system most reforms,¹⁶¹ can be achieved more readily by influencing judges than through the path of legislation. However, there are strong indications to the contrary: a separation of powers ordinarily excludes the judge from most major legal reform; adjudication, with its very limited research component and its focus on the individual situation, is intrinsically unsuited to major legal reform; and the judge’s rôle lies wholly within the framework of the law, with its closure, while the politician’s rôle does not.
It is tempting here, with regard to law in general, to develop an analogy between closure and coercion. The analogy would not be strained, since the two would often be found together—in law as elsewhere. Then one might say that, since it is generally agreed that coercion can be a necessary evil, so also with closure. Yet coercion and closure are substantially different. Coercion is an option, while closure is a fate. A fist can be clenched and unclenched, a gun picked up and put down; but closure is not something frankly chosen and not something that one can simply choose to discard. It hides itself, as an invisible cage. I have argued previously that closure is the mechanism of ideology and that ideology and law are the same thing seen from two different points of view.¹⁶² Ideology is a form of consciousness produced by social conflicts. Only secondarily is it a product of conspiracy, when people who perceive that there is a distortion promote that distortion for their own advantage. But that perception usually will not amount to actually seeing through the ideology.
It seems better to observe that, if normative closure is the distinguishing mark of law, nonetheless it is not law’s sole characteristic. Coercion is usually another. And there are others. Hence the negative value of the closure might be outweighed by a positive value of one or more other characteristics. That would be an evaluation to be made of a particular legal order at a particular moment, and usually of particular provisions or proposals, rather than as an assessment of law in general.
Legal closure, moreover, is not the only closure around. In addition, closure in law is abstract: a legal norm can have any content, including a reference to another closure. Accordingly, law can support or oppose another closure: that is, by protecting or penalising actions produced through the other closure. It can, for example, protect or penalise actions produced through closure towards people on the ground of their perceived membership of a ‘race’. In being abstract, however, legal closure has no reason to exist unless there is some other closure that it supports. One’s choice whether to assist law thus includes a choice whether to tolerate a closure that law supports, as the price of having law that opposes some other closure.
Legal closure, being abstract, is constant. But other closures may vary in their intensity. Thus it is possible to assist the introduction of a particular law, with its legal closure and together with a closure supported by that law or by the legal order as a whole, if worse closure is thereby prevented. For example: someone who considers that a particular legal order supports a closure involved in economic exploitation could readily advocate a combination of laws that provide for contracts of wage-labour and laws that prohibit slavery. Nor is closure obviously the worst thing that one can encounter. Thus I earlier noted with regret that Australia has failed to introduce domestic law against genocide.
If there is legal closure and the prospect of overcoming it reduces to the prospect of changing the social relations that give rise to and sustain it, the opponent of legal closure must address those particular relations. Moreover, social relations in general are already suffused with legal relations. The opponent of legal closure therefore needs in any case to contribute to processes of legal change. The legal comparatist would play a vital rôle in assisting such changes. One would advise how, in various legal systems, particular measures have reduced or sustained closures and what compromises have been made.

References

— Abelard 1136-9/1979. P Abelard Dialogus inter Philosophum, Judaeum et Christianum. Patrologia Latinae ed J-P Migne (Turnhout Brepols) 178: cols 1610-1683.
— Abelard 1136-9/1995. P Abelard Ethical Writings: His ‘Ethics’ or ‘Know Yourself’ and His Dialogue between a Philosopher, a Jew and a Christian tr PV Spade (Indianapolis Hackett).
— Allan 1993. TRS Allan Law, Liberty, and Justice: the Legal Foundations of British Constitutionalism (Oxford Clarendon P).
— Althusser and Balibar 1968/1970. L Althusser and É Balibar Reading Capital tr B Brewster (London New Left Books).
— Aristotle /1941. ‘On Interpretation’ in The Basic Works of Aristotle ed R McKeon (New York Random House) pp 38-61.
— Arnaud 1975. A-J Arnaud Les juristes face à la société: du XIX^e siècle à nos jours (Paris PU de France).
— Austin 1955/1975. JL Austin How To Do Things With Words (Oxford Clarendon P).
— Bentham 1792/1843. J Bentham ‘Truth versus Ashhurst’ Works ed J Bowring (Edinburgh Tait) 5: 231-237; also http://www.law.mq.edu.au/Units/law420/LAW203S/Ashhurst.htm.
— Benton 1998. T Benton ‘Naturalism in Social Science’ in Routledge Encyclopedia of Philosophy ed E Craig (London Routledge).
— Berger and Luckmann 1966/1971. PL Berger and T Luckmann The Social Construction of Reality: a Treatise in the Sociology of Knowledge (Harmondsworth Penguin).
— Berman 1977. HJ Berman ‘The Origins of Western Legal Science’ (1977) 90 Harvard Law Review 894-943.
— Blackshield and Williams 1996/2002. T Blackshield and G Williams Australian Constitutional Law and Theory: Commentary and Materials (3rd edn Sydney Federation P).
— Bracton c1256/1968-77. Bracton on the Laws and Customs of England ed and tr SE Thorne (Cambridge Mass Belknap P).
— Clark 2000. D Clark ‘The Icon of Liberty: the Status and Role of Magna Carta in Australian and New Zealand Law’ [2000] Melbourne University Law Review 34 http://www.austlii.edu.au/au/journals/MULR/2000/34.html
— Colletti 1974/1975. L Colletti ‘Marxism and the Dialectic’ tr J Matthews (1975) 93 New Left Review 3-29.
— David and Brierley 1964/1985. R David and JEC Brierley Major Legal Systems in the World Today: an Introduction to the Comparative Study of Law (3rd edn London Stevens).
— Davies 1996. M Davies Delimiting the Law: ‘Postmodernism’ and the Politics of Law (London Pluto P).
— Dostoyevsky 1880/1958. F Dostoyevsky The Brothers Karamazov tr D Magarshack (Harsmondsworth Penguin) in 2 vols.
— Dworkin 1977/1978. R Dworkin Taking Rights Seriously (London Duckworth).
— Dworkin 1986. R Dworkin Law’s Empire (London Fontana).
— Foucault 1961/1967. M Foucault Madness and Civilization: a History of Insanity in the Age of Reason tr R Howard (London Routledge).
— Foucault 1975/1979. M Foucault Discipline and Punish: the Birth of the Prison tr A Sheridan (Harmondsworth Peregrine).
— Glanvill 1189/1993. The Treatise on the Laws and Customs of the Realm of England Commonly Called Glanvill ed and tr GDG Hall(Oxford Clarendon P).
— Goldsworthy 1999. J Goldsworthy The Sovereignty of Parliament: History and Philosophy (Oxford Clarendon P).
— Grotius 1625/1913-25. H Grotius De Iure Belli ac Pacis tr FW Kelsey (Washington and Oxford Carnegie Institution and Clarendon P).
— Grotius 1631/1953. H Grotius The Jurisprudence of Holland tr RW Lee (Oxford Clarendon P).
— Hall 1959/1963. R Hall ‘Excluders’ in Philosophy and Ordinary Language CE Caton (ed) (Urbana U Illinois P) pp 67-73.
— Hart 1961/1994. HLA Hart The Concept of Law (2nd edn Oxford Clarendon P).
— Hartland 1924. ES Hartland Primitive Law (London Methuen).
— Hughes 1987/1988. R Hughes The Fatal Shore: a History of the Transportation of Convicts to Australia 1787-1868 (London Pan).
— Hunt and Wickham 1994. A Hunt and G Wickham Foucault and Law: Towards a Sociology of Law as Governance (London Pluto P).
— HWB 1971-. Historisches Wörterbuch der Philosophie (Basel Schwabe V).
— Izorche 2001. M-L Izorche ‘Propositions méthodologiques de la comparaison’ (2001) 53 Revue internationale de droit comparé 289-324.
— Jamin 2000. C Jamin ‘Le vieux rêve de Saleilles et Lambert revisité’ (2000) 52 Revue internationale de droit comparé 733-751.
— Kant 1763/1960. I Kant ‘Versuch den Begriff der negativen Grössen in die Weltweisheit einzuführen’ Werke (Wiesbaden Insel-V) 1: 777-819.
— Kant 1781/1929. I Kant Critique of Pure Reason tr N Kemp Smith (London Macmillan).
— Kelly 1995. P Kelly November 1975: the Inside Story of Australia’s Greatest Political Crisis (St Leonards Allen & Unwin).
— Kelsen 1934/1960. H Kelsen Reine Rechtslehre (2nd edn Wien Deuticke).
— Kelsen 1945/1961. H. Kelsen General Theory of Law and State tr A Wedberg (New York Russell & Russell).
— Kelsen 1965. H Kelsen ‘Was ist juristischer Positivismus?’ (1965) 20 Juristenzeitung 465-469.
— Kelsen 1979/1991. H Kelsen General Theory of Norms tr M Hartney (Oxford Clarendon P).
— Kennedy 1983. AG Kennedy ‘Cnut’s Law Code of 1018’ (1983) 11 Anglo-Saxon England 57-81.
— Kincaid 1998. H Kincaid ‘Positivism in the Social Sciences’ in Routledge Encyclopedia of Philosophy ed E Craig (London Routledge).
— Kirby 2000. M Kirby ‘Constitutional Interpretation and Original Intent: a Form of Ancestor Worship?’ [2000] Melbourne University Law Review 1 (http://www.austlii.edu.au/au/journals/MULR/2000/1.html)
— Kuttner 1936. S Kuttner ‘Sur les origines du terme “droit positif”’ (1936) 15 Revue historique de droit français et étranger (4^e s) 728-740.
— Levi 1949. EH Levi An Introduction to Legal Reasoning (Chicago U Chicago P).
— Liebermann 1903-16/1960. F Liebermann Die Gesetze der Angelsachsen (Aalen Scientia).
— Llewellyn 1930/1951. KN Llewellyn The Bramble Bush: on Our Law and its Study (New York Oceana).
— Maitland 1895. FW Maitland (ed and tr) Selected Passages from the Works of Bracton and Azo Publications of the Selden Society vol 8 (London Quaritch).
— Malinowski 1926. B Malinowski Crime and Custom in Savage Society (London Routledge & Kegan Paul).
— Markesinis 1997. BS Markesinis Foreign Law and Comparative Methodology: a Subject and a Thesis (Oxford Hart).
— Markesinis 2001. BS Markesinis Always on the Same Path (Oxford Hart).
— Maurer 1999. H Maurer Staatsrecht (München Beck).
— Moens and Trone 1974/2001. G Moens and J Trone Lumb and Moens’ The Constitution of the Commonwealth of Australia Annotated (Chatswood Butterworths).
— Muir Watt 2000. H Muir Watt ‘La fonction subversive du droit comparé’ (2000) 52 Revue internationale de droit comparé 503-527.
— NSW Legislative Council 2001. New South Wales Legislative Council Standing Committee on Law and Justice A NSW Bill of Rights (Sydney Parliament of New South Wales).
— Parashar 2000. A Parashar ‘Exclusions and the Voices of the Excluded’ (2000) 25 Australian Journal of Legal Philosophy 323-332.
— Pascal 1657/1967. B Pascal The Provincial Letters tr AJ Krailsheimer (Harmondsworth Penguin).
— Pfersmann 2001. O Pfersmann ‘Le droit comparé comme interprétation et comme théorie du droit’ (2001) Oxford U Comparative Law Forum 4; also (2001) 53 Revue international de droit comparé 275-288.
— Polanyi 1962/1967. M Polanyi The Tacit Dimension (London Routledge & Kegan Paul).
— Pound 1952. R Pound ‘Natural Natural Law and Positive Natural Law’ (1952) 68 Law Quarterly Review 330-336.
— Raz 1975. J Raz Practical Reason and Norms (London Oxford UP).
— Reynolds 1996. H Reynolds Aboriginal Sovereignty (St Leonards Allen & Unwin).
— Samuel 2001. G Samuel ‘Comparative Law as a Core Subject’ (2001) 21 Legal Studies 444-459.
— Sartre 1946/1965. J-P Sartre Anti-Semite and Jew tr GJ Becker (New York Schocken).
— Schmiedel 1966. B Schmiedel Consuetudo im klassischen und nachklassischen römischen Recht (Graz Bühlau).
— Searle 1969. JR Searle Speech Acts: an Essay in the Philosophy of Language (Cambridge Cambridge UP).
— Searle 1995. JR Searle The Construction of Social Reality (Harmondsworth Allen Lane The Penguin P).
— Shiner 1992a. RA Shiner ‘Exclusionary Reasons and the Explanation of Behaviour’ (1992) 5 Ratio Juris 1-22.
— Shiner 1992b. RA Shiner Norm and Nature: the Movements of Legal Thought (Oxford Clarendon P).
— Stewart 1981. I Stewart ‘Sociology in Jurisprudence: the Problem of “Law” as Object of Knowledge’ in Law, State, and Society ed B Fryer and others (London Croom Helm) pp 107-135.
— Stewart 1982. I Stewart Law and Custom: Critique of Right PhD thesis University College London.
— Stewart 1986. I Stewart ‘Droit et clôture: essai sur la conceptualisation critique du droit’ tr A Strowel (1986) 16 Revue interdisciplinaire d’études juridiques 33-89.
— Stewart 1987. I Stewart ‘Closure and the Legal Norm: an Essay in Critique of Law’ (1987) 50 Modern Law Review 908-933.
— Stewart 1990. I Stewart ‘The Critical Legal Science of Hans Kelsen’ (1990) 17 Journal of Law and Society 273-308; also http://www.law.mq.edu.au/HTML/staff/istewart/istewart.htm.
— Stewart 1997. I Stewart ‘Positivist Natural Law in Spencer’s Social Darwinism’ Archiv für Rechts-und Sozialphilosophie Beiheft 70(4): 78-86.
— Stewart 1998. I Stewart ‘Kelsen Tomorrow’ (1998) 51/1 Current Legal Problems 181-204; also http://www.law.mq.edu.au/HTML/staff/istewart/istewart.htm.
— Stewart 2002. I Stewart ‘Structure of the Australian Legal System’ in G Moens and R Biffot (eds) The Convergence of Legal Systems in the 21st Century: an Australian Approach (Brisbane CopyRight Publishing for U Queensland) pp 181-208.
— Summers 1968. RS Summers ‘Legal Philosophy Today—an Introduction’ in Essays in Legal Philosophy ed RS Summers (Oxford Blackwell) pp 1-21.
— Thompson 1991. EP Thompson Customs in Common (London Allen Lane).
— Thornton 2000. M Thornton ‘Towards Strategic Universalism’ (2000) 25 Australian Journal of Legal Philosophy 315-321.
— van de Kerchove and Ost 1988/1994. M van de Kerchove and F Ost Legal System Between Order and Disorder tr I Stewart (Oxford Clarendon P).
— Varga 1992. C Varga (ed) Comparative Legal Cultures (Aldershot Dartmouth).
— Waline 1935. M Waline ‘Positivisme philosophique, juridique et sociologique’ in Mélanges Carré de Malberg (Paris Sirey) pp 517-534.
— Weber 1922/1978. M Weber Economy and Society: an Outline of Interpretive Sociology various tr (Berkeley U California P).
— Whitelock 1979. D Whitelock (ed) English Historical Documents c.500-1042 (London Eyre & Spottiswoode).
— Williams 2000. G Williams A Bill of Rights for Australia (Sydney U of New South Wales P).
— Wilson 1997. R Wilson Bringing Them Home: Report of the National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families (Canberra Human Rights and Equal Opportunity Commission); also http://www.austlii.edu.au/au/special/rsjproject/rsjlibrary/hreoc/stolen.
— Zweigert and Kötz 1969-71/1998. K Zweigert and H Kötz Introduction to Comparative Law tr T Weir (3rd edn Oxford Clarendon P).

Notes

1 Senior Lecturer, Division of Law, Macquarie University, Sydney 2109, Australia; www.law.mq.edu.au/HTML/staff/istewart/istewart.htm; iain.stewart@mq.edu.au. An earlier version of this article (Stewart 2002) appeared under the title ‘Structure of the Australian Legal System’ in the volume of Australian national reports to the XVIth Congress of the International Academy of Comparative Law, Brisbane, 14-20 July 2002.

2 Pfersmann 2001: before and after note 3.

3 Pfersmann 2001: part I introduction, part IB.

4 cp Jamin 2000.

5 Pfersmann 2001: parts IA and IIA; Izorche 2001; Samuel 2001; cp Markesinis 1997: 25.

6 This has always been an especially strong element in public international law—but that is, equally, a special case. It is not confined to romanist legal orders: even in England the rule has now gone, that learned authors cannot be cited until they are safely dead (and unable to protest that they have been judicially misunderstood).

7 If one legal order incorporates a norm of another legal order by reference, as the Australian legal order often does with norms of English law, that is reference and not comparison.

8 I have endeavoured to explain how his system as a whole, through what would later be denounced as the ‘naturalistic fallacy’, contains a philosophical-positivist doctrine of natural law: Stewart 1997.

9 cp Pound 1952.

10 Pfersmann 2001: part IA.

11 Pfersmann 2001: part IA.

12 Pfersmann 2001: part IA.

13 Dostoyevsky 1880/1958: 691-6, 764.

14 Stewart 1998: 181-2.

15 Stewart 1990; 1998.

16 Hunt and Wickham, developing a Foucauldian theory of law, propose a definition of law that would include among ‘operations of law’ the ‘procedures of a company concerned with proper bookkeeping’ (Hunt and Wickham 199: 102). While Davies goes for broke: ‘Law is a limit’ and ‘any limit is a law’ (Davies 1996: 15).

17 Stewart 1981; 1998.

18 A ‘theory of law’, Dworkin holds, constitutes ‘the elaborate scheme of justification required to justify the body of laws’ (Dworkin 1977/1978: 66-8). Hart misses the point when he thinks that Dworkin pursues justification rather than description (Hart 1961/1994: 241ff). Dworkin’s aim (Dworkin 1986) is to make description serve justification by showing law in (as he states repeatedly) its ‘best’ light.

19 Pfersmann 2001: part IIA.

20 This point is distinct from that of the relation to general theory. Thus one Europhile comparatist envisages a ‘ius commune Europaeum’ (Markesinis 1997: 13). Within a federation, comparison between the laws of the States is readily accomplished within the federal legal order as a whole.

21 Pfersmann 2001: penultimate par.

22 Pfersmann 2001: final par.

23 As in the relation between the legal order of an EU member state and that of the EU as such.

24 Kelsen’s ‘static’ and ‘dynamic’ ordering: Kelsen 1945/1961: 112-13.

25 Levi 1949. I shall leave aside the old myth that in a common-law system judges do not make law: not even the judiciary will now say that; on the contrary, they enjoy the recognition of their lawmaking capacity. Australian courts have at least avoided one mistake of US federalism: they have not permitted the development of a separate ‘common law’ federally and in each State; in Australia there is but one common law, and it is now Australian common law and not part of the common law of England

26 cp Berman 1977; Arnaud 1975: 45-74.

27 A decision of the Federal Court, which will set out the facts exhaustively, is likely to be longer. One recent Federal Court decision, by a single judge, runs to nearly 500 pages: Cubillo v Commonwealth (2000) 174 ALR 97.

28 Kartinyeri v Commonwealth (1998) 195 CLR 337.

29 Zweigert and Kötz 1969-71/1998: 69.

30 Samuel 2001: 451-5.

31 The law student is required to read case after case until the instruments of analogy are acquired through a kind of osmosis. Llewellyn advised, ‘To do the work is not: to do the classes. Rather must you immerse yourself for all your hours in the law. Eat law, talk law, think law, drink law, babble of law and judgments in your sleep. Pickle yourself in law—it is your only hope.’ (Llewellyn 1930/1951: 96). Even this is relatively recent: English law was not taught in universities until the eighteenth century or, widely in universities, until well into the twentieth. Even then, some law schools in provincial universities had previously been training schools for intending solicitors and many legal ‘academics’ were moonlighting practitioners. Australia imitated this model and, unlike England, has yet to establish a legal academic tradition that is firmly distinguished from the legal profession.

32 Muir Watt 2000.

33 An example, jointly produced by the High Court of Australia and the House of Lords, is a radical muddle over one of the central concepts of torts (civil delict), ‘proximity’. This concept seems barely to have progressed since it was satirised by Pascal (Pascal 1657/1967: Letter I).

34 Recently, an interpretation of the Constitution of New South Wales was urged with reliance on English cases from the seventeenth century; however, the High Court was not altogether patient with that strategy: Durham Holdings v NSW (2000-01) 177 ALR 436.

35 It just should not happen, that in interpreting a constitutional provision that has stood unamended for a century a split decision of a supreme court should strike down a long-standing fiscal practice and, in the particular case, as much as A$5bn worth of current taxation: Ha v NSW (1997) 189 CLR 465. The sole beneficiaries of this decision, apart from the plaintiffs, appear to have been tobacco companies, who are estimated to have profited considerably from selling stockpiled tobacco products tax-free in the brief gap between the invalidation of State taxes and their replacement by new Commonwealth taxes.

36 ‘Scarce any man has the means of knowing a twentieth part of the laws he is bound by. Both sorts of law are kept most happily and carefully from the knowledge of the people: statute law by its shape and bulk; common law by its very essence. It is the judges … that make the common law. Do you know how they make it? Just as a man makes laws for his dog. When your dog does anything you want to break him of, you wait till he does it, and then beat him for it. This is the way you make laws for your dog: and this is the way the judges make law for you and me. They won’t tell a man beforehand what it is he should not do—they won’t so much as allow of his being told: they lie by till he has done something which they say he should not have done, and then they hang him for it. What way, then, has any man of coming at this dog-law? Only by watching their proceedings: by observing in what cases they have hanged a man, in what cases they have sent him to jail, in what cases they have seized his goods, and so forth. These proceedings they won’t publish themselves, and if anybody else publishes them, it is what they call a contempt of court, and a man may be sent to jail for it.’ (Bentham 1792/1843: 235).

This position has been much improved by the production of law reports and by placing law on the Internet: Australia is excellently served, in this respect, by the website of the Australasian Legal Information Institute: http://www.austlii.edu.au; the Commonwealth Government legal website (http://scaleplus.law.gov.au) has in some respects more information, but is difficult to use and is being wound back. However, physical access is one thing, intellectual access another.

37 In Australia, legal aid for civil cases has nearly disappeared. For criminal cases it is still widely available, yet has become so restricted in amount that a tactic now used by defence counsel is to prolong the proceedings until the defendant’s legal aid has run out and, as justice must be seen to be done, the case cannot continue.

38 An outstanding example of this is Re McBain (2002) 188 ALR 1. The High Court unanimously found that the Commonwealth Attorney-General had been wrong to issue a fiat that granted standing to the Australian Catholic bishops to appeal against the outcome of a case on in-vitro fertilisation in which the bishops had not been a party. He had done so even though both of the actual parties had been satisfied with the decision and had not wished to appeal. The bishops had required his fiat because the time for appeal had long expired.

39 Liebermann 1903-16/1960; Whitelock 1979; Kennedy 1983.

40 The work counted as the first English legal textbook, attributed to Glanvill (Glanvill 1189/1993), alludes in its preface to the prefatory constitution of Justinian’s Institutes, and quotes and relies on that text, although Glanvill’s book is just a list of procedures. While this outlook has something in common with ‘classical’ Roman law, ‘Roman law’ as we now know it is the codified Byzantine product.

41 Bracton c1256/1968-77. Bracton actually plagiarised extensively from the Italian glossator Azo (Maitland 1895).

42 Goodrich 1992.

43 The seventeenth century philosopher of science Bacon was Lord Chancellor (head of the judiciary); among later lawyer-scientists, in the nineteenth century Lyell was one of the founders of geology.

44 I will not provide detailed references for this discussion, since references are readily obtainable in recent textbooks: notably Blackshield and Williams 1996/2002; Moens and Trone 1974/2001.

45 See, further, Hughes 1987/1988.

46 Mabo v Queensland (No. 2) (1992) 175 CLR 1; since incorporated in statute (Native Title Act 1993) but later reduced (Native Title Amendment Act 1998).

47 Which has been demanded for the past 30 years by the Aboriginal Tent Embassy outside the old parliament house in Canberra.

48 The sole original mention of Indigenous peoples, as ‘the aboriginal race in each State’, was deleted in 1967; it had been a licence for the States to continue the genocide without federal interference: Constitution s 51(xxvi). The High Court, despite several opportunities, has never decided whether this power for the federal parliament to legislate with respect to any ‘race’ can be used only for the benefit of a ‘race’. The Constitution also still allows a State to disqualify ‘all persons of any race’ from voting in State elections, which would also deprive them of a federal vote: Constitution s 25. The aim of that section seems actually to be to discourage a State from taking such a step, by making the State forfeit part of its federal representation; nonetheless, the section contemplates and permits such disqualification.

49 Kelsen 1934/1960: 228-9. Or, ‘constitutional law’.

50 Ibid.

51 The expression ‘the Commonwealth’ has a double meaning: it refers to the whole Australian polity, ‘the Commonwealth of Australia’; but more often it specifies the federal power, over against the States.

52 Constitution ss 51-52.

53 Constitution s 109. The High Court has interpreted this section to the benefit of the Commonwealth.

54 Constitution s 3, absurdly, still states: ‘There shall be payable to the Queen out of the Consolidated Revenue fund of the Commonwealth, for the salary of the Governor-General, an annual sum which, until the Parliament otherwise provides, shall be ten thousand pounds.’

55 The records of the framing (http://setis.library.usyd.edu.au/fed) do not reveal what understanding of ‘separation of powers’ the framers may have had in mind: the version that appears to have been adopted does not follow Locke, Montesquieu or the Constitution of the USA. (On Montesquieu, see further my ‘Montesquieu in England’ forthcoming on this website.) Considerable difficulty has been caused by placing the Territories power (Constitution s 122, the Commonwealth’s power to govern a part of Australia directly) both outside the chapters dealing with separation of powers and within a chapter, on the States, where it seems not to belong. There are two main Territories: the Australian Capital Territory (corresponding to Washington DC) and the Northern Territory, which has been pressing for Statehood.

56 Constitution s 128. By statute, voting in all federal elections and referenda is compulsory. While it is possible for a proposal to go to referendum with the support of only one house, it is hard to imagine that it would then be approved by the people. Even with the support of both houses, a proposal is very unlikely to succeed in the referendum unless it also has bipartisan support. A further quirk is that voters in the Territories count in the overall total, although it is not necessary for the proposal to obtain a majority in any Territory.

57 Constitution s 44(i); Sue v Hill (1999) 199 CLR 462.

58 Constitution s 2.

59 References in the Constitution to ‘the Queen’ are to Queen Victoria and to her ‘heirs and successors in the sovereignty of the United Kingdom’: covering cl 2.

60 Constitution ss 58-60.

61 The Royal Style and Titles Bill 1973 was reserved for HM’s pleasure and received her personal assent (note to s 1 of the Act), since it concerned her titles. A formal matter, but a demonstration that this provision was not yet a dead letter.

62 See e.g. Kelly 1995.

63 Australian Citizenship Act 1948, as now titled.

64 Constitution s 51(xix).

65 cp Re Patterson (2001) 182 ALR 657.

66 Blackshield and Williams 1996/2002: 944-955.

67 Constitution s 51(xxix).

68 Sue v Hill (1999) 199 CLR 462.

69 Australia (Request and Consent) Act 1985 (Cth).

70 E.g. Durham Holdings v NSW (2000) 177 ALR 436, Kirby J at par 75.

71 Williams, in Blackshield and Williams 1996/2002: 170-2.

72 In 1999, proposals for constitutional amendment that would have exchanged the monarchy for a non-executive presidency selected by the parliament passed the parliament, against government resistance, but in the ensuing referendum were not carried nationally or in any State.

73 The ‘deep rights’ theory, involving an independence from the legislature of the judiciary as authentic makers of the common law and especially as controlling the conceptualisation of sovereignty (Allan 1993), has not taken root in Australia. It is widely accepted that the argument is historically mistaken: Goldsworthy 1999; Durham Holdings v NSW (2000-01) 177 ALR 436.

74 Constitution s 51(xxxi); this provision is fictionally applied in the delightful Australian film The Castle.

75 Kruger v Commonwealth (1997) 190 CLR 1. The meaning of the provision that guarantees a trial by jury where there is a prosecution on indictment for a Commonwealth offence (Constitution s 80) is judicially disputed; the herrschende Meinung is that it leaves the federal parliament free to determine what offences shall be triable on indictment: Cheng v The Queen (2000) 203 CLR 248. The minority view (maintained in that case) is that this cannot be correct, since it would render the section empty; however, the minority have not managed to specify what meaning would be appropriate.

76 Nulyarimma v Thompson (1999) 165 ALR 621. The Anti-Genocide Bill 1999 was a response to this decision.

77 Constitution ss 92, 117.

78 Constitution s 117.

79 Constitution s 116.

80 That is, although there must be voting, no individual or group has a constitutional right to participate in the process.

81 The first non-fiscal legislation of the Australian federal parliament was a statute inhibiting what the proposer, Prime Minister Barton, described as ‘Asiatic influxes’ to the detriment of ‘the Aryan races’: Immigration Restriction Act 1901 (Cth); HR Debs 7 August 1901, 3497-3503. This statute institutionalised for the new Commonwealth the White Australia policy, which continued until the 1970s. Even today, a naturalised Australian Citizen is not on the same footing as a native-born Australian—by a decision of the High Court, interpreting the federal constitution, even though no such distinction appears in the Australian Citizenship Act: Re Patterson; ex parte Taylor (2001) 182 ALR 657.

82 Genocide Convention Act 1949 (Cth).

83 A recent private member’s bill to achieve that, the Anti-Genocide Bill 1999, failed to obtain all-party support and foundered in committee on the definition of ‘genocide’. The conventional wisdom is that international law has no effect in Australian domestic law unless and as specifically incorporated by statute, although a convention that has been ratified may give rise to a legitimate expectation: Minister of State for Immigration and Ethnic Affairs v Teoh (1995) 183 CLR 273; legislation to reverse this decision has been introduced but not proceeded with. Recently, however, there has been judicial sympathy for the view that established norms of international law may be automatically incorporated in the common law: Merkel J in Nulyarimma v Thompson (1999) 165 ALR 621.

84 For an introduction to the issues, see Williams 2000. Much of a bill of rights could be enacted by bringing the provisions of international conventions into domestic law under the ‘external affairs’ power: Constitution s 51(xxix). But it would be difficult to use that power to enact a recognition of Indigenous peoples and without such recognition a bill of rights would have a gaping hole. That hole might be filled by relying on the ‘race’ power, Constitution s 51(xxvi), but there are serious problems with the current interpretation of that power, which the framers undoubtedly intended to license racial oppression. A statutory bill of rights could be enacted by the federal parliament with the agreement of the States: Constitution s 51(xxxvii, xxxviii). But there would be little point if not all States agreed both to go down this road and on a particular text, and the chances of that seem remote.

85 NSW Legislative Council 2001.

86 On Magna Carta, see Clark 2000; the version acknowledged is that of 1297. The Bill of Rights 1689 (1688 old style) applies so far as suitable to Australian conditions and not superseded by Australian constitutional or statutory provisions. Its provision as to the freedom of parliamentary proceedings, s 6, has been adopted by statute: Parliamentary Privileges Act 1987 (Cth) s 16. However, it has been judicially doubted that that section of the 1987 Act is entirely consistent with the freedom of political communication that the High Court has implied into the Commonwealth Constitution, so that it is now uncertain how far that section of the English Bill of 1689 now applies at the federal level: Blackshield and Williams 1996/2002: 402-10. Since the implied freedom applies at all levels of government—Commonwealth, State and local—the problem is further complicated.

87 The leading case is Lange v Australian Broadcasting Corporation (1997) 189 CLR 520.

88 This is both wider and narrower than the corresponding, explicit provision in the First Amendment to the Constitution of the USA. It is wider in that it is not restricted to ‘speech’ and has been held to include symbolic communication (waving a dead duck in front of a TV camera in protest against duck shooting). And it is narrower in that it is restricted to ‘political’ communication, i.e. to communication for a purpose related to the political system; although there is no restriction on the subject matter of the communication, since anything could be the subject of parliamentary discussion.

89 Blackshield and Williams 1996/2002: ch 8. Just one judge has rejected originalism wholesale: Kirby J in Abebe v Commonwealth (1999) 197 CLR 510 at par 203; Kirby 2000.

90 Mabo v Queensland (No. 2) (1992) 175 CLR 1. In that case, by recognising native title the Court arguably landed itself in the contradictory position of recognising Indigenous autonomy in private law while denying it in public law, with regard to cultures that do not acknowledge such a distinction between private and public (Reynolds 1996).

91 Kruger v Commonwealth (1997) 190 CLR 1.

92 Constitution s 51(xxvi); Kartinyeri v Commonwealth (1998) 195 CLR 337. In that case, counsel for the Commonwealth argued that the ‘race’ power would support legislation similar to the ‘race’ laws of apartheid South Africa and of Nazi Germany. The response of Kirby J was to assert that, where the Constitution is ambiguous—which, it being a constitution and hence couched in general terms, would be nearly everywhere—it could be interpreted by reference to international law: at par 166.

93 Wilson 1997.

94 Cubillo v Commonwealth (2000) 174 ALR 97.

95 Border Protection Bill 2001 (Cth).

96 Constitution s 51(xxvii).

97 The difference between a statement of purpose and a statement of subject matter can be hard to discern: which of them, for example, is ‘acquisition of property’—s 51(xxxi)? If the distinction can be made, it is nonetheless hard to see how any head of legislative power could be without purpose. The distinction to be made would then be between those heads that also have an evident subject matter and those that do not. An example of the latter would be ‘defence’—s 51(vi). The High Court has never settled on an evident subject matter of the immigration power, s 51(xxvii): Blackshield and Williams 1996/2002: 857-874.

98 cp Maurer 1999: 234.

99 Grundgesetz Art 1-19.

100 van de Kerchove and Ost 1988/1994.

101 Although for other purposes the tradition may be divided and parts of it distributed into other sets, for instance in terms of ‘legal families’: Zweigert and Kötz 1969-71/1998: 63-73; David and Brierley 1964/1985: 17-21; Varga 1992.

102 For example: ‘administrative law’ and droit administratif, although their signifieds are radically different; but ‘common law’, (la) common law (and variously in German), and ‘general law’, droit commun, allgemeines Recht.

103 For one example, in Roman law ‘natural law’ appears mainly under the label ‘justice (iustitia)’ while ius naturale is a kind of instinct, shared with animals, of entitlement (Justinian: Inst 1.2; Dig 1.1.1.3, 1.1.11). For another, ‘customary law’ and its equivalents are modern: in Roman law and through the Middle Ages, including Norman law, ‘law’ (or ‘act’ or ‘statute’, lex, loi, Gesetz) was contrasted with ‘custom’ (consuetudo, coutume, Gewohnheit) or ‘usage’ (consuetudo, usage, Brauch). There are even contexts (cp Schmiedel 1966) in which consuetudo could be translated as ‘legitimate expectation’.

104 The following discussion of the expressions ‘positive’ and ‘positive law’ draws mainly on HWB 1971-: ‘Positiv, Positivität’, ‘Positivismus’; Kuttner 1936.

105 A commentary by the Neoplatonist Calcidius on Plato’s Timaeus distinguishes ‘positive justice (iustitia positiva)’ from ‘natural justice (iustitia naturalis)’ as respectively artificial and natural. This commentary became widely known. Its influence, with the distinction between ‘positive’ and ‘natural’ justice, can be traced through to scholastics of the twelfth century. It accompanied Calcidius’ partial Latin translation of the Timaeus, which until the twelfth century was the only translation of a Platonic dialogue.

106 Dialogus inter Philosophum, Judaeum et Christianum: Abelard 1136-9/1995: 117-18; 1136-9/1979: col 1656; partially quoted, Kuttner 1936: 730.

107 It seems to translate (and in the seventeenth century was used to translate) Aristotle’s expression ‘thetic law (νόμους θετικούς)’ (Politics 1274 b 4). Aristotle is discussing laws to regulate procreation, for the purpose of property division.

108 When it appeared in the commentary by Bernard of Botone on the Decretals (1234) of Pope Gregory IX (Kuttner 1936: 736). This commentary would later be printed with those Decretals, which were to form the principal part of the Corpus Iuris Canonici up to 1918.

109 Grotius 1625/1913-25: II.38-9, 44-5; Grotius 1631/1953: I.7.

110 Summers 1968: 15-16. The variety has expanded considerably since he wrote.

111 Kelsen 1965.

112 Kincaid 1998; cp Benton 1998.

113 Where the emphasis is strongly on the methodology of physical science, it is also known as ‘scientism’.

114 cp Waline 1935.

115 I shall summarise what I have written elsewhere: Stewart 1990; 1998.

116 cp Stewart 1981.

117 Stewart 1982; 1986: 81-4.

118 By this time, the sense of ‘custom’ that, with the rise of sociology, would be replaced with the vocabulary of patterns of social behaviour had already parted company with the legal meaning of ‘custom’.

119 Inst 1.2.3-4 and 9.

120 The Laws of William, compiled in the late eleventh or early twelfth century, begin with it: ‘These are the laws and customs (les leis e les custumes) which King William granted to the people of England after the conquest of the land, which are the same that King Edward his cousin held before him.’ (Liebermann 1903-16/1960: I.492). A probably contemporary Latin translation gives leges et consuetudines (Liebermann 1903-16/1960: I.493).

121 Inst 1.2.3-4 and 9; Dig 1.1.6-9, 1.3-4.

122 Though the popular must not be confused with the democratic. As Thompson, studying eighteenth century England, finds: ‘Because law belongs to people, and is not alienated, or delegated, it is not thereby made necessarily more “nice” and tolerant, more cosy and folksy. It is only as nice and as tolerant as the prejudices and norms of the folk allow. […] In Bavaria the last manifestations of [H]aberfeldtreiben were linked to mafia-like blackmail, anti-semitism and, in the final stage, to ascendant Nazism. For some of its victims, the coming of a distanced (if alienated) Law and a bureaucratised police must have been felt as a liberation from the tyranny of one’s “own”.’ (Thompson 1991: 530-1).

123 Con Imperatoriam (prefacing the Institutes) title.

124 Hartland 1924; attacked in Malinowski 1926, although Malinowski’s image is vulnerable to Thompson’s stricture.

125 Customary international law would have to be a special case, since in its law-creating subjects are juristic persons.

126 This is to take existing legal theory not as something to contribute to but as a social phenomenon. Although there is little space to analyse it further here, I shall generally assume it as a referent.

127 Weber 1922/1978: 4.

128 Even Marx commits this error, seduced by his intellectual access to the texts of political economy into examining ‘ideology’ as an internal relation rather than as something encountered.

129 I will not take on board the postmodernist tenet that all is ‘text’. There is, at least, an important difference between reading a text and being hit on the head with it. During a police investigation in the former Byelorussia, it is reported:

They falsified evidence. They hung suspects upside down until the blood clogged their heads and they were ready to confess. They beat them with fists and boots, slammed their heads against an office safe, and in one chilling case, repeatedly hammered the head of a suspect with the thick and weighty tome, the Byelorussian Criminal Code.

‘This may have been to ensure that the suspect had no illusions regarding the legality of the investigation process,’ Literaturnya Gazeta reported drily in its account of the investigation that finally led to the arrest of the real killer. (The Guardian Weekly 13 March 1988)

Likewise, death is dangerous but it is not a supplement.

130 Kelsen does, however, valuably stress that a legal person is but the personification of a bundle of legal norms: Kelsen 1945/1961: 93-109, 191; Stewart 1998: 194-5. Thus one can say that in legal discourse an encounter between real human beings is re-presented as an abstract relation between legal persons. In particular, an irrelation between a legally authorised real person and another real person is re-presented as a relation between their legal personalities. That re-presentation forms part of the irrelation. Since a relation between legal persons is abstract, one need not be surprised to find a legal system in serious difficulty with concepts, such as causation or proximity, under which a relation between legal persons is to be determined by reference to spatio-temporal reality.

131 A common feature, for example, of the French Code civil.

132 Raz’s ‘exclusionary rules’ (Raz 1975), so far as I can see, simply do not exist. Albeit that his diagnosis of an exclusionary effect is intuitively compelling (Shiner 1992a; 1992b: ch 3).

133 I will give French equivalents, to assist discussion. Neither the English nor the French expressions are neologisms.

134 cp Aristotle /1941: 7; Kant 1763/1960: 783-97; Colletti 1974/1975.

135 cp Kelsen 1979/1991: 224.

136 cp the Monty Python ‘dead parrot’ sketch.

137 There is an extensive feminist literature on ‘exclusion’ and ‘silencing’, and these concepts have also played major rôles in critical race theory ever since (at least) Sartre’s Anti-Semite and Jew (Sartre 1946/1965), but I want to give them a more general turn here, with a hope that this turn could contribute to those debates.

138 cp Foucault 1961/1967; 1975/1979; Althusser and Balibar 1968/1970: 26-30; Berger and Luckmann 1966/1971: 132-4.

139 See e.g. Polanyi 1962/1967.

140 Searle terms this ‘the Background’ (Searle 1995: ch 6).

141 Austin 1955/1975: 4-7.

142 Briefly but succinctly in Speech Acts (Searle 1969: 50-71); then at length throughout The Construction of Social Reality (Searle 1995). Austin quickly dropped it, unsatisfied that he could distinguish ‘performatives’ as a class of utterances; Searle seems content that at least some utterances can be clearly so classed

143 Searle 1969: 33-4. Searle assumes that rules are either regulative or constitutive. For many legal rules this does not hold, since they both establish the possibility of (constitute) a ‘game’ and require (regulate) participation in it. An example (of which Searle himself is fond) is citizenship.

144 Searle 1969: 50-71. I shall not develop here a commentary on the ‘new institutionalism’ of MacCormick and Weinberger.

145 cp, on verbal ‘excluders’, Hall 1959/1963

146 See, further, Stewart 1987. In particular, the concept of the ‘direct imaginary’ introduced there (1987: 916-22) can I think be developed as a tool for answering two major questions that I shall not address here. First: can closure ever be complete? I would say that closure can effect a complete hegemony of determinate reality, but can never affect the direct imaginary because the latter is not determinate (and, for the same reason, is not a supplement). Second: if legal norms are characterised by closure, what price rights? Much of my possible answer to that is implicit in the Conclusion to this article. Beyond that, I would speculate on the possibility of ‘rights’ in a sense of norms that do not involve closure, on the criterion that they are compatible with assuming the priority of the direct imaginary over determinacy. Such as a right not to be killed, not to be maimed or not to starve. To construct that argument, however, would involve formulating what I say about ‘dark performatives’ compatibly with a conception of a signifier that does not correspond to a signified but points directly to (directly imaginary) reality.

147 In general, nothing hangs on whether a word used has an initial capital. Although in today’s English it is relevant that ‘God’ has an initial capital, that is not so for language in which it is normal to give an initial capital to a noun—such as modern German or seventeenth or eighteenth century English.

148 cp Searle 1969: 73-6. The employment of scare marks, e.g. ‘God’, is one way to effect this distinction.

149 Hart 1961/1994: 2-3.

150 I am not opposing ‘ideology’ and ‘science’: the opposite of ‘ideology’ is ‘critique’; science can be ideological or critical.

151 Hyde Park, London, circa 1972.

152 Kelsen 1945/1961: 112-13.

153 The Australian national anthem begins: ‘Australians all let us rejoice / For we are young and free’. Here is already a closure toward the country’s Indigenous population, who are neither young nor free.

154 I distinguish ‘form’ and ‘content’ here mainly to facilitate the exposition.

155 Parashar 2000; Thornton 2000.

156 I have not been able to find a source, but I recall hearing in the 1970s that a constitution of Sri Lanka was replaced in the following way: a national referendum approved the text of a new constitution; then the members of the parliament declared themselves to be a constituent assembly, adjourned to a different building, formally adopted the new constitution, and returned to the parliament building as the new parliament. In Canberra there is an ideal venue to adjourn to: the old parliament building, just downhill from the new.

157 Kant 1781/1929: 22.

158 I have drawn particularly upon Searle’s The Construction of Social Reality (Searle 1995), which from its title looks like a retort to the highly constructivist sociological classic The Social Construction of Reality (Berger and Luckmann 1966/1971), although Searle does not discuss it. Yet Searle’s theory of ‘institutional’ knowledge is highly constructivist.

159 Stewart 1987; 1998.

160 e.g. Markesinis 1997; 2001. It has not generally been supposed that the jurist should be at the service of just any legal system or necessarily at the service of the home system, yet the moral conditions under which service would be refused have not generally been set out.

161 Markesinis 2001: 56-7.

162 Stewart 1987.

© 2002 I. Stewart. This HTML edition © 2002 University of Oxford.
The contents of this page may be downloaded and printed out in single copies for individual use only. Making multiple copies without permission is prohibited.