Public Law Formal and substantive conceptions of the rule of law: an analytical framework

Paul P. Craig Subject: Jurisprudence. Other related subjects: Constitutional law Keywords: Jurisprudence; Rule of law *P.L. 467 There is a voluminous literature on the rule of law which examines the concept from almost every conceivable perspective. The analysis which follows makes no pretence at being a complete survey of these differing approaches. It does however attempt to address the subject in a way that is both important for public lawyers, and of broader significance outside of any particular legal system. The central theme of the article is the distinction between formal and substantive meanings of the rule of law. This dichotomy is, as will be seen below, of crucial importance in determining the nature of the specific legal precepts which can be derived from the rule of law. The difference between these conceptions of the rule of law will be explored fully below, but the essence of the distinction can be conveyed here. Formal conceptions of the rule of law address the manner in which the law was promulgated (was it by a properly authorised person, in a properly authorised manner, etc.); the clarity of the ensuing norm (was it sufficiently clear to guide an individual's conduct so as to enable a person to plan his or her life, etc.); and the temporal dimension of the enacted norm. (was it prospective or retrospective, etc.). Formal conceptions of the rule of law do not however seek to pass judgment upon the actual content of the law itself. They are not concerned with whether the law was in that sense a good law or a bad law, provided that the formal precepts of the rule of law were themselves met. Those who espouse substantive conceptions of the rule of law seek to go beyond this. They accept that the rule of law has the formal attributes mentioned above, but they wish to take the doctrine further. Certain substantive rights are said to be based on, or derived from, the rule of law. The concept is used as the foundation for these rights, which are then used to distinguish between good laws, which comply with such rights, and bad laws which do not. The structure of the analysis will be as follows. The first part of the article will consider the formal conception of the rule of law. This part of the argument will be divided into three sections. There will be an examination of the work of Raz who articulates the formal conception of the rule of law most clearly and explicitly. This will be followed by an analysis of Dicey's conception *P.L. 468 of the rule of law. It will 1997

be argued that he too was a formalist. In the final section of this part of the article Unger's challenge to the formal conception of the rule of law will be considered. The second part of the article will focus upon a thoroughgoing substantive account of the rule of law provided by Dworkin. The implications of adopting such a conception of the rule of law will be brought out, and will be exemplified through consideration of the work of Sir John Laws and Trevor Allan. The third and final part of the article will consider whether there is some middle way between the adoption of a purely formal conception of the rule of law and the fully substantive version of the doctrine.

1. The formal conception of the rule of law (a) Joseph Raz

It may be helpful at the outset to make clear why those who subscribe to the formal conception of the rule of law insist that the concept should bear this meaning. Raz provides the clearest explanation. If the rule of law is the rule of the good law then to explain its nature is to propound a complete social philosophy. But if so the term lacks any useful function. We have no need to be converted to the rule of law just in order to discover that to believe in it is to believe that good should triumph. The rule of law is a political ideal which a legal system may lack or possess to a greater or lesser degree. That much is common ground. It is also to be insisted that the rule of law is just one of the virtues by which a legal system may be judged and by which it is to be judged. It is not to be confused with democracy, justice, equality (before the law or otherwise), human rights of any kind or respect for persons or for the dignity of man.1 What Raz is getting at here can be explained quite straightforwardly. We may all agree that laws should be just, that their content should be morally sound and that substantive rights should be protected within society. The problem is that if the rule of law is taken to encompass the necessity for good laws in this sense then the concept ceases to have any useful independent function for the following reason. There is a wealth of literature devoted to the discussion of the meaning of a just society, the nature of the rights which should subsist therein, and the appropriate boundaries of governmental action. Political theory has tackled questions such as these from time immemorial. To bring these issues within the rubric of the rule of law would therefore have the effect of robbing this concept of any function independent of such political theories. Laws would be condemned or upheld as being in conformity with, or contrary to, the rule of law in this substantive sense when the condemnation or praise would simply be reflective of attachment to one particular political theory. The message which Raz conveys is an important one. If you wish to argue about the justness of society do so by all means. If you wish to defend a *P.L.

469 particular type of individual right then present your argument. Draw upon the wealth of literature which addresses these matters directly. Nothing however is to be gained by cloaking whatever conclusion you reach in the mantle of the rule of law, since this merely reflects the conclusion which has already been arrived at through the relevant political theory. It is for this reason that Raz insists that the rule of law should be seen in formal terms. The consequence of this reading is, as Raz readily admits, that the rule of law could be met by regimes whose laws are morally objectionable, provided that they comply with the formal precepts which comprise the rule of law. It is equally the case, on this view, that a democratic regime will not necessarily always have laws which do measure up to the rule of law. What then is the proper remit of the rule of law viewed in this formal manner? Raz makes it clear that it cannot just mean that government action is authorised by law since the concept would then be thin indeed. Any law properly passed by Parliament would meet the rule of law defined in this manner. That laws should be passed in the correct legal manner is none the less a necessary facet of a formal conception of the rule of law. It is not howeversufficient. The other important aspect of the rule of law is that the laws thus promulgated should be capable of guiding one's conduct in order that one can plan one's life. It is from this general precept that Raz then deduces a number of more specific attributes that laws should have in order that they could be said to be in compliance with the rule of law. All are related to this idea of enabling individuals to be able to plan their lives. The list includes the following: that laws should be prospective, not retrospective; that they should be relatively stable; that particular laws should be guided by open, general and clear rules; that there should be an independent judiciary; that there should be access to the courts; and that the discretion which law enforcement agencies possess should not be allowed to undermine the purposes of the relevant legal rules. On this view the rule of law is essentially a negative value as Raz himself admits. Given that the law can empower the state to do all manner of things the rule of law minimises the danger created by the law itself. It does so by ensuring that whatever the content of the law, at least it should be open, clear, stable, general and applied by an impartial judiciary. It would however be mistaken not to recognise the more positive side of the rule of law when viewed in this manner. Even if the actual content of the law is morally reprehensible, conformity to the rule of law will often be necessary to ensure that individuals actually comply with the demands which the law imposes.

One final point which is of importance concerning this conception of the rule of law is that, as Raz emphasises, it is only one virtue of a legal system, and may have to be sacrificed to attain other desired ends. We may feel that the rule of law virtues of having clear, general norms must be sacrificed if the best or only way to achieve a desired goal is to have more discretionary, open textured legal provisions. This may be the case in circumstances when it is not possible to lay down in advance in the enabling legalisation clear, prospective rules in sufficient detail to cover all eventualities. Modifications to the rule of law in this manner are not somehow forbidden or proscribed. Given that it is only one *P.L. 470 virtue of a legal system it should not prevent the attainment of other virtues valued by that system.

(b) Dicey

Dicey's conception of the rule of law2 is well known and it has been subjected to analysis from all of the diverse directions set out above. The focus of the discussion which follows will be upon the formal/substantive divide and the way in which this facilitates our understanding of his reasoning. Dicey's first limb of the rule of law was that: no man is punishable or can be lawfully made to suffer in body or goods except for a distinct breach of law established in the ordinary legal manner before the ordinary courts of the land. In this sense the rule of law is contrasted with every system of government based on the exercise by persons in authority of wide, arbitrary, or discretionary powers of constraint.3 There are a number of well known critiques of this principle. One of the most important was that Dicey underestimated both the existence of discretionary power which existed at the time when he was writing; and also the fact that such discretionary power was often a necessary and legitimate consequence of the growth of governmental power in the nineteenth century. To return to the focus of the current discussion, the key question is whether Dicey's first principle is to be perceived in formal or substantive terms? Now there is no doubt that the words used by Dicey could bear a substantive meaning. This is particularly true of the word arbitrary, and some have suggested that Dicey's vision of the rule of law should be viewed in this manner.4 The weight of evidence is, however, clearly against this reading of Dicey's work: his first principle is formalistic and not substantive. This is readily apparent from the first sentence of the above formulation. This requires that laws under which people are condemned should be passed in the correct legal manner and that guilt should only be established through the ordinary trial process. Nothing

here speaks to the content of the laws which an individual will have to face when taken before the courts. But what then of the remainder of the first principle? Does this not have a substantive content? Would not laws which are bad or evil be labelled as arbitrary within the meaning of Dicey's first principle? If not, what then does this word connote? Now, as stated above, it would of course be possible for the word arbitrary to have a substantive content. On this view a law which was properly enacted by Parliament, in compliance with all correct procedures, which was pristinely clear in its application, and which was applied by an impartial judiciary, might*P.L. 471 none the less be tainted as arbitrary if it was thought to infringe certain fundamental rights, or if it entailed excessive punishment. It is equally clear that the word arbitrary can have a formal meaning. When used in this latter sense the word arbitrary would provide the foundation for criticism of two kinds of norm. One category would comprise those allegedly legal rules which, when examined, do not in fact have any legal foundation. They might not have been enacted in the proper manner because, for example, they have not been passed through Parliament and do not come within the ambit of the prerogative. The other category of formal arbitrariness would be used to describe those norms which have been passed in the correct legal manner, but where the resulting law was impossibly vague or unclear, with the result that individuals had no idea how to plan their lives in the light of the relevant legal rule. Formal arbitrariness in either of these senses is independent of whether the content of the legislation was good or bad, just or unjust. So which of these two senses of arbitrary did Dicey have in mind when formulating his rule of law? Two arguments, one positive, the other negative, point strongly to the fact that he was using the term in the latter, formalistic sense. The positive argument is to be found in Dicey's own discussion within later sections of the Law of the Constitution. When discussing freedom of the individual Dicey contrasts continental systems with that in England. He claims that the former were not free from arbitrary power. For Dicey, the Bastille was the visible sign of lawless power, even though it had only a handful of people in it when it fell. This was because it was a symbol of arbitrary power, in the sense that the executive would incarcerate people there without any lawful authority, or for the commission of crimes which were impossibly vague. Dicey spends two pages lamenting the fate of poor Voltaire who was twice placed in the Bastille at royal or aristocratic whim.5 In England, by way of contrast, the singularity of our law was not so much its leniency or goodness, but its legality.6 Although we might have had harsh laws, a

person's fate was not dependent upon the caprice of some other person who might happen to have power. Thus Dicey was under no illusion that all English laws were substantively just; nor does he attempt to claim that they were. His conclusion that England was not subject to arbitrary power, and that in this respect we fared better than those on the continent, was based on the formalistic sense of the term arbitrary considered above. The laws might have been harsh, but they had to be properly passed, and applied by the ordinary courts, before an individual could lose his or her freedom. Moreover, Dicey's discussion of the relationship between sovereignty and the rule of law further reinforces the view that his conception of the latter was formal and not substantive.7 The negative argument which points to the same conclusion is that if Dicey had intended the term arbitrary to bear a substantive connotation then he provided absolutely no criterion as to how this sense of arbitrariness was to be *P.L. 472 determined. We shall see in the discussion which follows the difficulties which have to be faced if one wishes to adopt a substantive conception of the rule of law. Suffice it to say for the present that such a view of the rule of law necessitates the articulation of some criterion which will then provide the foundation for the conclusion that a particular law really was unjust or bad. Now Dicey did of course have strong political views, as is well known. Yet at no stage is there any evidence to suggest that he intended these political and moral precepts to be used to determine that a properly enacted law which was clear, and applied by an impartial judiciary, could none the less be regarded as substantively arbitrary, and hence contrary to the rule of law, on the grounds that it infringed these or any other such precepts. The second principle of the rule of law concerns equality. Dicey's formulation of the principle is as follows. We mean when we speak of the rule of law as a characteristic of our country, not only that with us no man is above the law, but (what is a different thing) that here every man, whatever be his rank or condition, is subject to the ordinary law of the realm and amenable to the jurisdiction of the ordinary tribunals.8 Once again there are well known critiques of Dicey's second principle. His misunderstanding of the French droit administratif was legendary, as was his misapprehension of how much administrative law existed in nineteenthcentury England, with adjudication through specialist tribunals rather than the ordinary courts.9 Our primary concern is, however, as to whether this second principle is formalistic or substantive. As with the first of Dicey's principles, so here too it will be argued that the weight of evidence clearly indicates

that Dicey was thinking in formal rather than substantive terms when formulating his ideas about equality. Dicey's formulation is concerned primarily with formal access to the courts, not with the nature of the rules which individuals will find when they get there. This point is captured well by Marshall: Equality before the law, understood as the equal subjection of all classes to a common rule, might at least be contrasted significantly with chaos or lawlessness, but it does not in itself imply any qualitative view about the sort of law to which all are subject.10 Now to be sure it is true that Dicey was explicitly against officials being accorded any special privileges over and beyond those of ordinary citizens, and in this sense Dicey imported a substantive element into this aspect of his rule of law. But beyond this Dicey's second principle does not touch on substantive equality at all. As Marshall states, speaking of this part of Dicey's analysis: *P.L. 473 It omits, however, to register the truism that the law which all citizens find when they get to the common courts may make unequal provision for some as against others. The same law that bound all could say that the Crown could not be sued, and that policeman and state officials should have powers, privileges, or legal defences not open to private citizens.11 A substantive conception of equality would require the articulation of principles through which the courts would then determine whether the application of one rule to Group A was compatible with the application of a different rule to Group B. Legal systems use varying criteria to resolve questions of this kind. Issues of considerable complexity are involved as courts attempt to decide whether, for example, the division between two groups as to the content of the rules which they face is based on some rational, intelligible difference between them. Any thoroughgoing theory of substantive equality will moreover be based, implicitly, if not explicitly, on some broader back-ground political theory of which it forms but one important part. There is no indication that Dicey in his second rule intended to grapple with such matters, nor that he intended the second limb of the rule of law to have this type of substantive content.12 Dicey's third limb of the rule of law does not sit easily with the previous two. The essence of this precept can be stated as follows. We may say that the constitution is pervaded by the rule of law on the ground that the general principles of the constitution (as for example the right to personal liberty, or the right of public meeting) are with us the result of judicial decisions determining the rights of private persons in particular cases brought before the courts; whereas under many foreign constitutions the security (such as it is) given to the rights of

individuals appears to result from the general principles of the constitution.13 This limb of the rule of law has caused considerable confusion. One common error is to read this aspect of the rule of law as demanding that a society must indeed possess certain individual rights if it is to conform to the rule of law. If this view were correct then Dicey would indeed be imbuing the rule of law with a substantive content. The reason why this is erroneous is quite simply that it is not what Dicey actually said. He might have said it. He might have said a number of things; but he did not say this, What Dicey actually said was crucially different. His argument was not that the rule of law demanded adherence to certain specific substantive rights. It was that if you wished to protect such rights then the common law technique was better than that employed on the continent. This is manifestly clear from Dicey's own formulation of the third principle of the *P.L. 474 rule of law, and from the ensuing discussion. In this discussion Dicey argued that the protection of rights on the continent through Bills of Rights was ineffective, since such constitutional documents could so easily be abrogated at the stroke of a pen. Under the common law, where individual rights were the result of numerous judicial decisions indicating when the individual was at liberty to speak freely etc., it would be considerably more difficult for some authoritarian regime to sweep these rights aside.14 Now this argument may or may not be true on its merits. Even if it did have some empirical validity when Dicey wrote, one might argue that matters are in any event different now and that constitutionally enshrined protections for rights would serve the individual better than the traditional common law methodology. But this species of argument is irrelevant to the point made above, which is that Dicey's third limb of the rule of law is no more substantive than the previous two. It no more demands the existence of certain specific substantive rights than do the earlier limbs of his formulation.

(c) Unger

There have been frequent challenges to the rule of law. One of the most interesting is that advanced by Unger who argues, in effect, both that the formal conception of the rule of law was always a mask for substantive inequalities in power, and that in the modern day this formal conception is in any event increasingly unattainable.15 The contention is that the rule of law provided a convenient legitimating mask for substantive inequality within liberal society. For Unger a liberal society was one in which there were many different groups, with no particular group being able to dominate the whole. There was no preordained caste system, nor was there any fixed hierarchy of the kind which prevailed during the medieval period. In a

liberal society a justification was required for the way in which society was ordered. In Unger's view the rule of law emerged to provide this justification. How did it perform this function? For Unger it achieved this goal by making it appear that power was impersonal. His vision of the rule of law is not markedly different from that espoused by Raz and Dicey. In Unger's terms it enshrines commitment to the generality, autonomy, neutrality and predictability of legal norms. These ideas capture the same themes as those articulated in the earlier discussion. For Unger it was these very formal attributes which played such a large part in legitimating the existing power structures within society, by making it appear that power was impersonal. Unger's contention is that this legitimating function performed by the rule of law was always really a sham. There were two reasons why this was so. On the one hand, one of the premises underlying this legitimating function performed by the rule of law was that most power was concentrated in government. Yet he argues that in practical terms considerable power lay in other places, including the workplace, the family etc. Inequality in these areas *P.L. 475 was not touched by a commitment to formal equality within the legal arena. On the other hand, and more importantly, Unger claims that the assumption behind the rule of law rhetoric was that power could be effectively constrained by rules. If rules were indeed general and impartially applied then it would be difficult for the ruler, or a particular class, to turn them to personal advantage. He contends that this assumption was not sustainable. This was in part because even if rules were general their content would reflect the power of the dominant class. It was in part also because even if a rule was general it would still have to be applied by the judiciary, and this could not be done in a manner that was truly value neutral. Notwithstanding these problems the rule of law could still appear to be a legitimating device within Unger's liberal society. Things are, he claims, different in a post-liberal society, in that the very aims of the rule of law itself, in terms of the generality, neutrality, had autonomy of law, are undermined. A post-liberal society as described by Unger has two characteristics. One is increased governmental involvement in the economy and in the regulation of society in general. The government is forced to intervene in more and more areas, from social welfare to planning, and from utilities regulation to health and safety, in order to ameliorate the problems which flow from unjustified hierarchy. The other related characteristic is that the boundaries between the public and private spheres become eroded. The idea that the government is merely a neutral guardian of the social order comes under increased strain.

These changes in the nature of the societal order have ramifications for the type of legal norms which emerge. Legislation becomes more open-textured and is framed in broader, less precise terms. This is because the complex aims which government now seems to achieve can not be attained through clear and precise rules. More open-ended discretion has to be left to administrative agencies and to the courts. The style of legal reasoning alters. It ceases to be formal and becomes more purposive. Formal legal reasoning could be used when there were clear, general rules which were capable of mechanical application. The nature of the statutes which have to be interpreted in a postliberal society pushes the judiciary towards a more purposive style of reasoning, in which increased attention is placed on the ends which the legislation is intended to serve in order to determine its actual remit. Purposive legal reasoning tends to place a higher premium on the substantive justice of the outcome, and not to focus exclusively on concerns with formal justice. These characteristics of a post-liberal society are said by Unger to have profound implications for the traditional attributes of the rule of law. The generality of law is undermined. This is in part because the complex problems which have to be dealt with often render the formulation of general rules impossible. It is in part because the increased attention which is now placed on substantive as opposed to formal justice means that we might wish to have more particular rules which differentiate between groups to a greater degree. Autonomy is also said to be undermined. Courts will now be forced to apply opentextured, often vague, statutes, which leave many issues unclear. The judiciary will then be placed in a position where they have to weigh a*P.L. 476 wide variety of factors. Their judgments will come to resemble more closely those which are made in the political forum, or by administrators. People will undoubtedly have different views on Unger's provocative analysis. Some might agree wholeheartedly with it, and see it as a timely unmasking or deconstructing of a revered legal myth. Others might regard it as an exercise in historical sociology which is insufficiently grounded in empirical evidence. Yet others might adopt some intermediate position, including the present writer. Exigencies of space preclude any detailed analysis of Unger's thesis, but three related comments may be of help in locating this thesis within the more general structure of this article. The first is to reiterate a point made earlier. Unger's view of what the rule of law actually means does not differ markedly from that of Raz or Dicey. He too adopts a formal version of the rule of law, as expressed in the ideas of generality, neutrality, clarity and autonomy.

The second comment concerns Unger's thesis about the rule of law within what he terms liberal society. This aspect of Unger's thesis has two related components: that the concept was employed as a device to legitimate power inequalities and that it could never properly fulfil this goal. Even if we accept the first component of this argument, the second is more open to question. We have already seen that one of the principal reasons why Unger claims that the objective for which the rule of law was being employed could never be properly achieved was because of substantive inequalities in power. This meant that the content of the resulting norms would be weighted in favour of the dominant grouping within society. Let us assume that this was indeed the case, although such an assumption is contestable. There is none the less an element of circularity in the argument. The rule of law as used by Unger is a formal concept. Adherence to the concept has never been claimed to guarantee a just society, if that phrase is used to connote a society in which thesubstantive distribution of wealth and power is morally acceptable. Nor has the formal concept of the rule of law ever pretended to be a guarantee that the substantive content of particular laws will be just, in the sense of preventing any form of bias within the law for a dominant power grouping. To claim therefore that any legitimating function performed by the rule of law within liberal society was undermined because of substantive power inequalities is to condemn the rule of law for not combating issues which it, as a legal concept, never claimed to be redressing. The third comment relates to the decline of the rule of law which Unger perceives within what he terms post-liberal society. This third comment has both an empirical and a conceptual dimension. In empirical terms one might simply question how serious the problem actually is. How many statutes do actually take the open-textured, vague form which he identifies? How often are the judiciary in fact forced into making legal judgments which are said to replicate closely the type of balancing process which legislatures themselves have to undertake? Are all laws now like this? There are also interesting conceptual issues raised by Unger's analysis. One of the most important may be put in the following terms. We have already seen*P.L. 477 that writers such as Raz emphasise that the rule of law is but one virtue which legal systems should possess. It may well have to be sacrificed if we wish to reach certain ends, the attainment of which is not possible while still adhering to the formal rule of law precepts. Viewed in this way the developments which Unger identifies within post-liberal society could be understood simply as instances in which society has decided that the pursuit of other virtues, such as help for particular disadvantaged groups, necessitates the sacrifice of formal rule of law values.

2. The substantive conception of the rule of law (a) Dworkin

It is not fortuitous or surprising that one of the principal advocates of the formal conception of the rule of law, Raz, is also a leading exponent of legal positivism. The formal conception of the rule of law, and the desire to keep legal questions separate from broader issues of political theory in deciding what the content of the law actually is, fit naturally together. The view of law and adjudication espoused by Dworkin is very different. It is central to this thesis that, subject to questions of fit, the courts should be deciding legal questions according to the best theory of justice.16 On this view broader questions of political theory are central to the resolution of what rights people currently possess. Given this theory, it would be odd, to say the least, to conceive of the rule of law in purely formal terms. That this is indeed so can be seen by considering Dworkin's own discussion.17 He distinguishes between two different conceptions of the rule of law. The first, which he terms the rule book conception, is in effect a version of the formal rule of law discussed above. It says nothing about the content of the laws which exist within a legal system, but merely insists that the government should never exercise power against individuals except in accordance with rules which have been set out in advance and made available to all. As Dworkin recognises, those who have this conception of the rule of law care about the content of the law, but they say that this is a matter of substantive justice, and that substantive justice is an independent ideal, in no sense part of the ideal of the rule of law.18 The second conception of the rule of law is termed by Dworkin the rights conception. He defines it in the following manner: It assumes that citizens have moral rights and duties with respect to one another, and political rights against the state as a whole. It insists that these moral and political rights be recognized in positive law, so that they may be enforced upon the demand of individual citizens through courts or other judicial institutions of the familiar type, so far as this is practicable. The rule of law on this conception is the ideal of rule by an accurate public conception of individual rights. It does not distinguish, as the rule book *P.L. 478 conception does, between the rule of law and substantive justice; on the contrary it requires, as part of the ideal of law, that the rules in the book capture and enforce moral rights.19 Proponents of a rights based or substantive conception of the rule of law will have regard to the values enshrined in the formal conception of the rule of law for two reasons. On the one hand, such values would feature in any serious theory of justice.20 On the other hand, these

values will be of relevance when answering the key question posed by advocates of a rights based conception of the rule of law: whether the plaintiff has a moral right to receive, in court, what he or she or it demands.21 The rule book is relevant to that ultimate question. In a democracy, people have at least a strong prima facie moral right that courts enforce the rights that a representative legislature has enacted. That is why some cases are easy cases on the rights model as well as on the rule book model. If it is clear what the legislature has granted them, then it is also clear what they have a moral right to receive in court. (That statement must be qualified in a democracy whose constitution limits legislative power. It must also be qualified, though it is a complex question how it must be qualified, in a democracy whose laws are fundamentally unjust.)22 Three conclusions can be drawn from the analysis thus far. First, on Dworkin's theory there is no place for a separate concept of the rule of law as such at all. On this view, the rule of law simply captures the theory of law and adjudication which he espouses. That theory directs us to consider what is the best theory of justice as part of the decision as to what rights people presently have. The very need to preserve a firm distinction between legal rules and a more complete political philosophy is rejected by the thesis itself. It is for this reason that one finds virtually no mention of the phrase the rule of law as such within his major work on legal theory, Law's Empire. 23 Secondly, while, as we have seen, Dworkin would continue to have regard to the formal rule of law values within this theory, these values would not be separately demarcated as comprising or constituting the rule of law. They would simply take their place within the theory of law and adjudication which he espouses in the manner described above.24 Thirdly, it should be made clear that the substantive view of the rule of law requires the articulation not simply of general concepts of liberty, equality and the like. It demands that the particular conception of these broader concepts be revealed. It is precisely because many differing theories of justice will*P.L. 479 incorporate some general concept of liberty, equality and the like, that the choice of what is perceived to be the best theory of justice to inform the adjudicative process will necessitate the articulation of a more particular conception of these freedoms.25 Dworkin's own argument serves to emphasise that the very meaning of the rule of law will be inextricably linked with one's definition of law itself and with the proper adjudicative role of the judge. A positivist is likely to subscribe to the formal sense of the rule of law, and to keep this distinct from the content of particular laws. A rights based analysis

of the Dworkin kind is diametrically opposed to this dichotomy between form and substance. The content of laws will be evaluated in order to determine whether they are compatible with the moral rights which individuals possess. On this view the rule of law is nothing more or less than a synonym for a rights based theory of law and adjudication. The preceding analysis can be of assistance in understanding the use made of the rule of law by those writing more specifically about public law. This can be exemplified by focusing upon the work of Sir John Laws and Trevor Allan.

(b) Sir John Laws

In a challenging and important series of articles Laws has articulated the role of the courts in the protection of fundamental rights.26 The detailed nature of the argument is not of immediate concern to us here. Suffice it to say for the present that Laws presents an essentially nonpositivist, rights based conception of law and the role of the judge in cases involving fundamental rights. He posits a higher order law which is binding on the elected Parliament, with the courts as the guardian of both fundamental individual rights, and what may be termed structural constitutional rights.27 In his recent work Sir John Laws has gone a step further and indicated what he perceives to be the content of constitutional rights which individuals possess.28 Space precludes any detailed examination of Laws' thesis. It suffices for the purposes of the present analysis to say that his argument is explicitly Kantian. The conception of individual autonomy, the content of individual constitutional rights, and the divide drawn between positive and negative rights, with priority being accorded to negative rights, are all aspects of the argument expressly posited on this philosophical foundation. The conception of the rule of law which Sir John advances is, not surprisingly, substantive. The rule of law is held to encompass an attachment to freedom, certainty and fairness. The first of these elements is the substantive component of the rule of law, while the second and the third bring in the more *P.L. 480 traditional attributes of the formal rule of law.29 Two related comments on this analysis are pertinent to the present inquiry. First, the fact that the protection of fundamental rights requires us to specify what those rights are, and that this necessitates the articulation of some background political theory, is surely correct. The analysis of Sir John Laws is to be welcomed for his very willingness to specify the background assumptions which he believes give substance and meaning to the rights which are to be regarded as fundamental in this respect. These assumptions will always be controversial. The particular consequences of adopting a liberal Kantian theory may

themselves be open to debate.30 More fundamentally, that theory has itself been criticised. One of the principal debating points within modern political theory is between advocates of some version of Kantian liberalism and variants of republicanism and communitarianism.31 Indeed, Lord Irvine, in his response to Sir John Laws, adverted to this when he pointed to the communitarian critique of the classic liberal notion of the autonomous moral agent.32 Secondly, the role to be played by the rule of law within Sir John Laws' argument can, however, be questioned. The analysis within the preceding section has already shown the consequences of adopting a substantive conception of the rule of law: that phrase ceases to have a function independent of the rights based theory of law and adjudication. Does it then matter that an advocate of such a rights based conception of law continues to employ the rule of law within his or her analysis? The continued usage of the rule of law may be perfectly acceptable, provided that everyone understands how the concept is being used. When it is being used by one who subscribes to a Dworkin type theory it has the following connotation: that a theory of justice is integral to deciding what rights individuals presently have, and that this therefore necessitates the articulation of what that theory of justice actually is. As seen from the preceding discussion, this is the meaning of a substantive conception of the rule of law. The danger of continuing to use the phrase the rule of law when it is being used in this manner is not simply that it adds nothing to the rights based theory of law and adjudication. It is that the phrase can, intentionally or unintentionally, lend added weight to the particular theory of justice/freedom which is being espoused. This is because the phrase the rule of law has a particular force of its own as a result of its use across time. The point can be put quite simply by contrasting the following two statements. Statement 1: the rule of law demands the protection of freedom conceived of in a particular manner. Statement 2: a certain theory of law and adjudication sees justice as an essential element in deciding what the law actually is, and therefore requires the *P.L. 481 articulation of a particular theory of justice and moral rights. The content of these two statements is the same. Statement 1 is, however, likely to arouse considerably less opposition than Statement 2. This is a point of some importance. One may well feel that a theory of law and adjudication of the kind suggested by Dworkin is the best available. Given that this requires the articulation of a background theory of justice/freedom, it is all the more important for the particular theory which is adopted to be transparent and open to critical evaluation in a manner which can be impeded by cloaking it with the rule of law. This is not a criticism of Sir John Laws who does

make very clear the political theory which underpins his analysis. The point being made here is a general one.

(c) Allan

Trevor Allan also adopts an explicitly substantive conception of the rule of law.33 The analysis which he presents is of interest, although it is not unproblematic. The nature of Allan's reasoning may be described as follows. He recognises the dangers of subscribing to a substantive conception of the rule of law which have been articulated by Raz. Nonetheless he is firmly of the view that such a conception is both correct in principle and of most use to those who are engaged in constitutional theory in the context of the British constitution. Four connected arguments can be discerned in his analysis. First, formal conceptions of the rule of law are themselves based upon substantive foundations, namely ideas of moral autonomy and the respect for the individual. Given that this is so, it is unrealistic or implausible to preserve the dichotomy between form and substance.34 Secondly, in normative terms, adjudication, and in particular common law adjudication, involves the application of principles as well as rules. The application of such principles will often require the court to have recourse to considerations of substantive justice or fairness which go beyond the formal conception of the rule of law.35 Thirdly, in descriptive terms, the common law courts in the U.K. do in fact reason in this manner.36 Finally, Allan draws explicitly upon the work of Dworkin to support his thesis.37 A number of points can be made about this line of argument. The first relates to stage one of Trevor Allan's argument, concerning the distinction between form and substance. There is undoubtedly some force in this aspect of the thesis. Formal conceptions of the rule of law are indeed based upon substantive considerations of moral autonomy and the like. This is acknowledged by advocates of the formal conception such as Raz. The fact that this is so does not, however, serve to justify a conclusion that the distinction between formal and substantive conceptions of the rule of law is untenable. Legal systems have always distinguished between formal/procedural and substantive norms, notwithstanding the fact that the former may be *P.L. 482 underpinned by substantive values. More important in this context is the level of abstraction which is in issue. It is one thing to affirm, correctly, that the formal conception of the rule of law is based on some general abstract substantive values which relate to human autonomy. It is quite another matter to conclude that therefore the rule of law must be taken to encompass specific substantive freedoms, such as liberty or equality. This is all the more so once it is realised that in order for these broad substantiveconcepts to be rendered operational, it is necessary to articulate the

particular conception of liberty or equality, etc., which one believes should guide legislative and judicial behaviour. The second point, is that stages two and four of the thesis will clearly get you where you want to go in the following sense. If one seeks to support a substantive conception of the rule of law then it would suffice in this regard simply to declare oneself to be an advocate of a Dworkinian view of law and adjudication, which is effectively what Allan does. Stage four of the argument is therefore in fact the crucial one, and stage two is simply part of the broader Dworkin theory. It was Dworkin who undertook the seminal work on the place of principles within adjudication.38 The third point follows immediately from the second. While Trevor Allan does adopt a substantive view of the rule of law, and while the Dworkinian argument is crucial for reaching this conclusion, Allan appears to believe that the rule of law has some independent value as a constitutional concept. We have already seen from the preceding discussion that, on the Dworkin view, the rule of law simply becomes synonymous with the theory of law and adjudication which he espouses. It does not have an independent role as such. The final point to make is that the descriptive element within Allan's argument, stage three, is problematic in two senses. On the one hand, insofar as it seeks to establish that our courts do reason in the manner for which Allan argues, the sample is far too small to make it possible to draw any really meaningful conclusions from it. More detailed study does not lend support to this conclusion.39 On the other hand, the conclusions which Allan draws from the examples which he does consider are themselves open to question in normative terms. It is assumed that these cases do entail the application of principles rather than rules and that such decisions cannot be explained, or made sense of, in terms of a positivist account of law. Space precludes any exegesis on this general issue, which would clearly take us beyond the remit of this article. Suffice it to say for the present that these conclusions are open to question for the following reason. The fact that a particular court has recourse to moral considerations or conceptions of justice or fairness when deciding a case tells one nothing, in and of itself, as to whether that court is reasoning in a manner consonant with a positivist or non-positivist view of adjudication. Positivists do not deny that *P.L. 483 courts can reason in such a manner. Quite the contrary. For positivists, such as Raz, courts should reason in this fashion when they are faced with cases for which the existing source-based law provides no answer: According to [the sources thesis], the law on a question is settled when legally binding sources provide its solution. In such cases judges are typically said to apply the law, and since it is source-based, its

application involves technical, legal skills in reasoning from those sources and does not call for moral acumen. If a legal question is not answered from legal sources then it lacks a legal answer the law on the question is unsettled. In deciding such cases courts inevitably break new (legal) ground and their decision develops the law (at least in precedent-based legal systems). Naturally, their decisions in such cases rely at least partly on moral and other extra-legal considerations.40 Raz does, moreover, go considerably further in articulating a view as to the nature of common law adjudication. He distinguishes between the role of the courts in what he terms regulated cases, those which fall under a common law or statutory rule which does not require judicial discretion for the determination of the dispute, and unregulated cases, where there is some gap in the law applicable to the case.41 The latter include cases where there is some indeterminacy of language or intention, or those where there are two conflicting rules potentially applicable to the case.42 On this view courts are regarded as making law in cases of unregulated disputes;43 when they do so they should adopt those rules which they believe to be best, in the same manner as a legislator; this may well entail taking into account moral considerations; there may nonetheless, be constraints which cause courts to be less adventurous than legislators; the courts' law making function is not dependent upon the courts necessarily realising that this is what they were doing; and law application and law making may both be present within a particular case.44 The purpose of this part of the discussion is not to demonstrate that the positivist view of adjudication is necessarily correct. The object is much more limited. It is simply to show that the examples which Allan does proffer simply do not prove the point for which he argues. The fact that one can point to cases where courts have adverted to considerations of justice, fairness or morality may be perfectly reconcilable with positivist legal theory. They may simply be *P.L. 484 regarded as cases where the courts have engaged in lawmaking in the positivist sense of that term.45

3. A middle way? (a) Raz

Given the preceding analysis it might well be thought that it is not possible for there to be a middle way between the formal and substantive visions of the rule of law. This may well be so. Before reaching any conclusion on this issue we should, however, consider the views of Joseph Raz who has returned to the topic in his more recent work.46 His analysis concentrates upon the role of the rule of law within Britain, rather than focusing upon the more universal aspects of the concept as he had done within his earlier work.

According to Raz the core idea is the principled faithful application of the law.47 The major features are its insistence on an open, public administration of justice, with reasoned decisions by an independent judiciary, based on publicly promulgated, prospective, principled legislation.48 The principle of the rule of law is addressed to the courts, legislature, and also other bodies such as the police and administrative authorities. While this vision of the rule of law requires the courts to be faithful to legislation which emerges from a democratic legislature, it also sets limits to majoritarian democracy, by requiring principled as well as faithful adjudication.49 The core of the thesis is to be found in this very notion of principled adjudication. It requires that the courts make decisions which are reasoned and public. But the real nub of the idea is captured in the following quotation: In insisting that judicial decisions should be not only faithful but also principled, I am suggesting that the function of the rule of law is to facilitate the integration of particular pieces of legislation with the underlying doctrines of the legal system A particular reform of police powers to search for prohibited drugs should be applied in a manner which is both faithful to the legislative purpose and principled in integrating it with traditional doctrines of the liberties of the citizen.50 This aspect of the judicial function is justified in part by the need to ensure that there is a coherence of purpose within the law.51 It is also held to be*P.L. 485 justified in order to mix52 the fruits of longestablished traditions with the urgencies of short-term exigencies. It is precisely because legislatures can be susceptible to short-term influences, whether generated by elections or the need to respond quickly to public pressure, that the courts should have a role as the guardians of longer-term tradition.53 What then of the place of civil rights within this vision of the rule of law? Raz is careful in this respect. He states that the protection of such rights is partly presupposed and partly implied by the preceding analysis.54 The analysis presupposes such rights in that Raz's present discussion of the rule of law is confined to democratic societies, and a society cannot be democratic without the existence of such rights. The analysis implies the existence of such rights because in insisting on the integration of legislation and other current measures with legal tradition enshrined in doctrine, the rule of law respects those civil rights which are part of the backbone of the legal culture, part of its fundamental traditions.55 The discussion thus far has already touched upon a number of aspects of legal theory stricto sensu, and this is inevitable in any meaningful discussion of the rule of law. It would, however, take us beyond the remit of this article to consider the way in which the vision of the rule of law fits into the broader theory of law which Raz enunciates.56

(b) Jowell
We should not conclude our analysis without considering another attempt to delineate a middle ground between a purely formal conception of the rule of law, and the thoroughgoing substantive version of this doctrine. This is to be found in the work of Jeffrey Jowell.57 He accepts that one must be careful about equating the rule of law with the substance of particular rules. He accepts also that a significant part of the rule of law is concerned with procedure or form as opposed to substance. Jowell does however believe that the rule of law has a substantive dimension.58He perceives the rule of law as a principle of institutional morality and as a constraint on the uninhibited exercise of government power and argues that it does possess a substantive aspect. This aspect is manifest in the judiciary's willingness to strike down agency action if it is unreasonable, arbitrary or capricious. Jowell recognises that in some instances judicial intervention is premised upon the fact that the agency has departed from the sphere over*P.L. 486 which it has been given authority by the legislature. In other circumstances the courts do not even really pretend that they are enforcing legislative intent, and are in reality subjecting agency decision-making to substantive control based on the rule of law. It is reasonably clear from a general reading of Jeffrey Jowell's thesis that he seeks to tread a middle ground between the formal and substantive conception of the rule of law.59 There are, however, two difficulties in holding to this middle ground. On the one hand, the principles of judicial review are not selfexecuting. A judicial decision holding that administrative action should be struck because it is unreasonable or capricious will often, of necessity, involve the identification of various interests, and the assignment of normative weight to them. This has become more evident of late as the courts and commentators have begun to talk of applying these administrative law principles with differing intensity depending upon the nature of the interests at stake within a particular case. The possible inclusion of proportionality as an independent head of judicial review reinforces this point,60 since this doctrine, by its very nature, requires the identification of the competing interests in a dispute and the assignment of normative value to them. Given that this is so it is difficult to perform this exercise of normative evaluation without explicitly or implicitly relying on some background conception of justice. On the other hand, the exclusion of constitutional doctrine is problematic. Jeffrey Jowell argues that the substantive aspect of the rule of law stems primarily from the need to constrain the uninhibited

exercise of governmental power. The administrative law tools of judicial review are the mechanism to achieve this end. Yet given the foundation of the argument it is difficult to see why tools more commonly associated with constitutional law, concerning rights such as expression, equality and the like, should not also be of relevance here. If the argument is based on the need to constrain the uninhibited exercise of government power, and the administrative law principles of judicial review are regarded as serving this facet of the rule of law, then why should the limits of the rule of law61 be set so as to exclude constitutional constraints designed to serve the same end? Once constitutional constraints on governmental power are included it becomes even more difficult to avoid the taxing issues of justice and political theory adverted to above. The preceding argument should not in any way be taken as denying the real importance of constraints on governmental power. The object of the analysis is, rather, to question whether the inclusion of such constraints within the rule of law is readily compatible with the preservation of a middle ground between a formal and substantive conception of this doctrine.

*P.L. 487 4. Conclusion

There will be no attempt to summarise the entirety of the arguments presented above. Three connected points may, however, be made by way of conclusion. First, the rule of law is rightly regarded as a central principle of constitutional governance. It is, therefore, all the more important that we should be as clear as possible about its meaning. It has been argued that clarity in this respect cannot be attained unless public lawyers are aware of the issues of legal theory which underlie the concept. This should not come as a surprise. At the most basic level one might well expect that the meaning to be attributed to a phrase such as the rule of law would be dependent upon what one understands by the term law. That this is indeed so is confirmed by the preceding analysis, since what ultimately divides the formalist and substantive conceptions of the rule of law is disagreement about the way in which we identify legal norms. Secondly, it should be noted that both Raz and Dworkin actually agree on one important issue which is central to us here: the adoption of a fully substantive conception of the rule of law has the consequence of robbing the concept of any function which is independent of the theory of justice which imbues such an account of law. Their fundamental disagreement concerns the very nature of law and the role of the courts in adjudication. Thirdly, it is of course open to public lawyers, and indeed any one else, to choose between the contending views of the rule of law presented

above. Debate on this issue is helpful. It is nonetheless important to understand the consequences of adopting a particular position on this matter. The phrase the rule of law has a power or force of its own. To criticise governmental action as contrary to the rule of law immediately casts it in a bad light. Such criticism may well be warranted depending upon the circumstances. Yet if the nub of the critique is posited upon the substantive conception of the rule of law then intellectual honestly requires that this is made clear, and it also demands clarity as to the particular theory of justice which informs the critique. P.L. 1997, Aut, 467-487