Home | Databases | WorldLII | Search | Feedback University of New South Wales Faculty of Law Research Series

Krygier, Martin --- "The Rule of Law: Legality, Teleology, Sociology" [2007] UNSWLRS 65

Last Updated: 28 April 2008

The Rule of Law: Legality, Teleology, Sociology

Martin Krygier

Forthcoming in Gianluigi Palombella and Neil Walker, eds, Re-locating the Rule of Law, Hart Publishers, Oxford, 2008.

Abstract

The concept of the rule of law is no new coin. It has long been the stuff of legal cliché, but also of extensive conceptual analysis and scholarly debate. The concept has a strong presence in legal theory and in traditions and branches of political theory. It has been less noticed or analysed by social theorists, however, which is odd. That neglect is unfortunate, for some of the central questions about the rule of law are sociological ones. So my suggestion is that we would do well to explore a hardly existent sociology, the sociology of the rule of law.

I provide nothing like that here, only some reasons to seek it. The argument is briefly this. The proper way to approach the rule of law is not to offer, as lawyers typically do, a list of characteristics of laws and legal institutions supposedly necessary, if not sufficient, for the rule of law to exist; let me call that the anatomical approach. Rather, one should begin with teleology and end with sociology. That is, I suggest we start by asking what we might want the rule of law for, by which I mean not external ends that it might serve, such as economic growth or democracy, but something like its telos, the point of the enterprise, goals internal to, immanent in the concept. Only then should we move to ask what sorts of things need to happen for us to achieve such a state of affairs, and only then move to ask what we need in order to get it. That third question, the bottom line, as it were, will of course involve legal institutions but it cannot be answered without looking beyond them to the societies in which they function, the ways they function there, and what else happens there which interacts with and affects the sway of law. For the rule of law to exist, still more to flourish and be secure, many things beside the law matter, and since societies differ in many ways, so will those things.

‘Das Vergessen der Absichten is die häufigste Dummheit, die gemacht wird’ - Nietzsche[1]

The concept of the rule of law is no new coin. It has long been the stuff of legal cliché, but also of extensive conceptual analysis and scholarly debate. The concept has a strong presence in legal theory and in traditions and branches of political theory. It has been central to centuries of political thought about how power might be restrained, without being emasculated. It has been less noticed or analysed by social theorists, however, which is odd. For if the rule of law matters legally and politically, it certainly matters socially. It is typically contrasted with arbitrary exercise of power. Since that is a common cause of social disorientation and in the worst cases catastrophe, what might be done to prevent or lessen it is a proper matter of social concern. Moreover, the success of the rule of law as a restraint on power has indispensable social conditions as important as, or more than, any particular legal specifications one might suggest.
But there are many things that conspire against close sociological exploration of the rule of law. Prominent among them are purists’ fears of disciplinary contamination. The rule of law is so associated with law and politics that sociologists have tended to keep their distance from this hallowed legal ideal – too normative, too legal, too political, too formal, too disconnected from life; and how is it to be measured? That neglect is unfortunate, for some of the central questions about the rule of law are sociological ones.
Conversely, if sociological innocence about the rule of law is striking, so too are the immaculate conceptions of legal and political theorists, untainted as they have remained by social theory or empirical social research. That is an odd way to work. Presumably if they were confident that by prayer they could eliminate arbitrariness in the exercise of power, they would think more about prayer and less about law. Instead we are sent to particular sorts of legal arrangements which on their own, I have sought to argue, often don’t amount to much. So my suggestion is that we would do well to explore a ‘social science that does not quite yet exist,’[2] the sociology of the rule of law.
I provide nothing like that here, only some reasons to seek it. I have more confidence in the questions I ask and the goals I postulate, than in my ability to answer and reach them all successfully. My argument has been developed in a number of pieces and in answer to a number of different questions.[3] This chapter attempts to restate that argument in general terms, refine it where I have noticed weaknesses in earlier renditions, and raise some new questions which any such sociology needs to answer.
The argument is briefly this. The proper way to approach the rule of law is not to offer, as lawyers typically do, a list of characteristics of laws and legal institutions supposedly necessary, if not sufficient, for the rule of law to exist; let me call that the anatomical approach. Rather, one should begin with teleology and end with sociology. That is, I suggest we start by asking what we might want the rule of law for, by which I mean not external ends that it might serve, such as economic growth or democracy, but something like its telos, the point of the enterprise, goals internal to, immanent in the concept. Only then should we move to ask what sorts of things need to happen for us to achieve such a state of affairs, and only then move to ask what we need in order to get it. That third question, the bottom line, as it were, will of course involve legal institutions but it cannot be answered without looking beyond them to the societies in which they function, the ways they function there, and what else happens there which interacts with and affects the sway of law. For the rule of law to exist, still more to flourish and be secure, many things beside the law matter, and since societies differ in many ways, so will those things.
My concern is with questions that need to be put about the rule of law, narrow or broad, thick or thin, and, just as important, the order in which they should be put. My belief[4] is that the movement through these questions takes one from universal human needs (more or less. I wouldn’t bet my life on universality; pretty general will do me fine) through some also fairly general conditions, to extremely variable ways in which they might be met in particular societies with particular histories and particular problems at particular times. So a universal, institution-based, answer to what the rule of law is, is implausible. And it will often mislead. Indeed it might well lead us away from the rule of law. Or maybe not those of us from rule-of-law-rich states, but those who seek or are advised to seek, to emulate us in this regard. For my thinking in these matters has been provoked by the problems of those who lack the rule of law but want it or are being told they need it. I have recently started to suspect that the approach I defend is less appropriate for understanding elements of the rule of law where it is established, than it might be for seeking to introduce it where it is not. In both cases, however, the reason is the same: we lack, as a matter of principle rather than epistemological shortcoming, universal legal-institutional prescriptions for the rule of law. The reasons for that, in turn, have to do with social complexity rather than lack of legal ingenuity.
In what follows, I start with an outline of the anatomical approach that I oppose, draw on some historical and one modern ground for suspicion of it, and move to teleology and sociology, which I commend. I conclude by considering whether what I recommend is the opposite of what it is best to do in relation to the rule of law. I argue that at times it is, but I find a way not to be embarrassed by that conclusion.

1. Anatomy

Anatomies of the rule of law typically have two features in common: first, their focus is on legal institutions and the norms and practices directly associated with them; second, a list of elements of such institutions and practices is presented as adding up to the rule of law. That is not a surprise, since law is plausibly assumed to be central to the rule of law, lawyers are typically taken to be the experts on it, and they provide the lists. Not a surprise, but perhaps a mistake.
A famous example is Albert Venn Dicey’s three point definition of the rule of law, considered by Judith Shklar to be ‘the most influential restatement of the Rule of Law since the eighteenth century’:[5]

1. ‘no man is punishable ...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.’[6]
2. ‘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.’[7]
3. ‘the general principles of the constitution ... 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 results, or appears to result, from the general principles of the constitution.’[8]

We will return to Dicey. Here my observation is not that he is wrong about English law, though he might be;[9] nor that he is parochial about the rule of law, though he is.[10] Simply it is to point to what he takes to be the appropriate way to ground and explain ‘a trait of national character which is as noticeable as it is hard to portray.’[11]
Not everyone picks on the same features as Dicey. A more abstract account, overlapping but not identical, is developed by Friedrich von Hayek, at least in one of his renditions. On this view, the ideal type of the rule of law (from which modernity has steadily fallen away) depends on abstract, general and certain laws rather than particular commands, where ‘[l]aw in its ideal form might be described as a “once-and-for-all” command that is directed to unknown people and that is abstracted from all particular circumstances of time and place and refers only to such conditions as may occur anywhere and at any time.’[12] Hayek concedes that in reality laws shade all the way down into concrete commands, but the rule of law, he insists, depends on the extent to which the bulk of laws are not commands. Many writers have agreed with him in this stipulation.
Other influential accounts follow the eight elements that Lon Fuller characterised as constituting ‘the internal morality of law.’ This list has frequently been adopted or adapted to constitute the conditions of the rule of law. Briefly these conditions are that: there must be 1. general rules; 2. made public; 3. non-retroactive; 4. comprehensible; 5. non-contradictory; 6. possible to perform; 7. relatively stable; 8. administered in ways congruent with the rules as announced.[13] There are many other such accounts of the form required of institutional hardware, but I needn’t delay with them for I am not concerned with their details. The issue I take is with their character. What they all have in common is: a) the confident assumption that the central ingredients of the rule of law are legal institutions, and b) the equally confident assumption that we are in a position to stipulate in general terms what aspects and elements of these institutions produce the results we seek. Many other accounts of the rule of law are even more specific than these, and mention the configuration of institutions, presence or absence of bills of rights, institutional measures to guarantee judicial independence, and so on.
Let me stress, this is not a definitional dispute. I am not saying this is a semantically mistaken way of proceeding. People can define as they wish and since the rule of law has such aura today, there is no tying it down. Moreover, as we will see, there might be circumstances where it makes sense, better sense than I once believed, to take this path. Nevertheless, I think it is liable to mislead us about important substantive matters, and to do so precisely where the stakes are highest.
Many countries have emerged relatively recently from dictatorships that, whatever else can be said about them, ignored, denied, and/or defiled the rule of law. Many citizens of these countries seek, or are urged, to establish it. Here some clarity about the rule of law matters, since the assumption is that there has not (some places, has never) been much of it there, and so it has to be developed from the ground up. In the post-communist world, for example, a taste for the rule of law was made all the more tantalizing (for some) by its absence from local experience, and its alleged presence elsewhere - in what were, in the communist and early post-communist period, called ‘normal countries.’[14] What to do and how to do it?
It was common in 1989 to insist that what distinguished these revolutions from any of their forebears was that the former intended ‘no more experiments.’ Successful models existed in normal countries, and the job was to adopt them. Those locals who wanted the rule of law thought of it a bit like working telephones, or roads without potholes, which they also lacked: they have them, we want them, let’s get them. Or, even if local enthusiasm was less strong, there were plenty of foreigners to insist: we have them, you need them, here take them. Timothy Garton Ash faithfully captures this sentiment of the time:

In politics they are all saying: There is no "socialist democracy," there is only democracy. And by democracy they mean multi-party, parliamentary democracy as practiced in contemporary Western, Northern, and Southern Europe. They are all saying: There is no "socialist legality," there is only legality. And by that they mean the rule of law, guaranteed by the constitutionally anchored independence of the judiciary.[15]

This taste for democracy and legality ‘without adjectives,’ as dissidents used to put it, can be readily appreciated. They were rightly allergic to such substance-cancelling qualifiers. But, to the extent that saying ‘there is only legality’ might suggest that there exists one obvious incarnation of legality which merely needs to be copied by eager imitators, then the taste for legality unqualified is misleading. As Stephen Holmes remarked a few years ago,[16] a production technology is easier to transplant than an interaction technology. I imagine he would agree that is all the more the case when you only have a very vague idea what the relevant technology is, and no one seems to have much of a clue how it works.
In the past twenty years, over a billion dollars has been spent internationally to bring the rule of law to benighted countries thought to need it. Some of the promoters and observers of these efforts have recently begun to issue crestfallen reports.[17] One observation might suffice:

In legal circles in developing countries and in international development circles, rule of law has become almost synonymous with legal and judicial reform. Basic questions about what legal systems across diverse countries actually do, why they do it, and to what effect are either inadequately explored or totally ignored. In developed and developing countries, larger questions about the relationship of the rule of law to human rights, democracy, civil society, economic development, and governance often are reduced to arid doctrinalism in the legal fraternity. And in the practice of the international donor community, the rule of law is reduced to sectors of support, the most prominent of which is the judicial sector.
... During the last seven years, we have witnessed an explosion of literature related to legal and judicial reform. Yet very little attention has been paid to the widening gap between theory and practice, or to the disconnection between stated project goals and objectives and the actual activities supported.[18]

Common to many of these reports is the complaint that, notwithstanding the vast amounts of money funnelled into legal and judicial reform, in the name of the rule of law, corruption still rules, hidden structures of power decide, networks are key, those who win outside the law win inside, if they ever need to venture inside. One conclusion you might draw is that even though the rule of law has been installed, it’s just not worth the money spent on it. Even when people get it they won’t have got much. Thus, roughly following the model of that old and bleak hospital joke: ‘the operation was successful; the patient died,’ Frank Upham laments about:

The likelihood that Western mischaracterization of the appropriate roles of law will be accepted by developing countries, thus leading to misallocation of domestic effort and attention, and perhaps most important, eventually to deep disillusionment with the potential of law. When the revision of the criminal code does not prevent warlords from creating havoc in Afghanistan and the training of Chinese judges by American law professors does not prevent the detention of political dissidents – or, perversely, enables judges to provide plausible legal reasons for their detention – political leaders on all sides may turn away from law completely and miss the modest role that law can play in political and economic development.[19]

Typically, Upham identifies the rule of law and exaggerated expectations of it, rather than an inadequate understanding of it, as the source of his fears. However, what if the problem is less that the rule of law was installed but failed to do much good, than that what was installed was not the rule of law? That is my view. When legal institutional tinkering fails to prevent havoc, when people who count ignore the law and those who don’t merely suffer it, the rule of law is in very poor shape if it exists at all, whatever the laws look like, and commonly that should not have been a surprise. On their own, the legal institutional features so often identified with the rule of law are not up to the task. Indeed, they never are, but always need supporting circumstances, social and political structures and cultural supports, which are not always available and are difficult to engineer. Some peoples are lucky to be born into societies where those supports are old and embedded. They should recognize their luck. Others face challenges, which are never merely legal-institutional. And yet, not just disappointed political leaders but also misdirected critics are liable to indict the rule of law, even though it has yet to visit the scene of the crime.[20]

One possibility, then, is that one could have elements of what many take to be the rule of law, but not have the rule of law. Another is that you might have the rule of law without the elements. Let me offer some quickly sketched examples of this second option, from two domains on both of which my expertise is slight and derivative: one is English history, the other American modernity.

2. History

Within Britain, and inherited by its dependants, the concept of the rule of law is deeply embedded and very old indeed. This has recently been well demonstrated by the American legal historian, John Philip Reid. Thus he quotes Bracton in the thirteenth century declaring of the king:

Let him, therefore, temper his power by law, which is the bridle of power, that he may live according to the laws, for the law of mankind has decreed that his own laws bind the lawgiver, and elsewhere in the same source, it is a saying worthy of the majesty of a ruler that the prince acknowledges himself bound by the laws. Nothing is more fitting for a sovereign than to live by the laws, nor is there any greater sovereignty than to govern according to law, and he ought properly to yield to the law what the law has bestowed upon him, for the law makes him king.[21]

Perhaps it is this heritage that impressed those ‘foreign observers of English manners, such for example as Voltaire, De Lolme, de Tocqueville, or Gneist,’ of whom Dicey wrote that they have been ‘far more struck than have Englishmen themselves with the fact that England is a country governed, as is scarcely any other part of Europe, under the rule of law.’[22] Whatever slippage we allow between self-preening ideology and actual historical practice, there is something in the claim. But if there is, it is important to know what that claim amounts to. It is not that the sort of institutional arrangements now identified with the rule of law have a long history, that law was, or was even thought to be, certain, prospective, promulgated, general, and so on. Rather, it is that the power even of the King was required to be exercised within bounds set, however, vaguely, by existing law. And that was predominantly common law.[23] That in turn, as Blackstone put it in the eighteenth century, was an ‘antient collection of unwritten maxims and ccustoms,’[24] and ‘[t]he only method of proving that this or that maxim is a rule of the common law, is by shewing that it hath always been the custom to observe it.’[25]
Reid shows that in English tradition until the eighteenth century, law was identified with misty, murky but ages-old custom, traced to a time when ‘the memory of man runneth not to the contrary.’ He writes, ‘the medieval constitutional law out of which today’s rule of law developed would not have met the requirements of clarity or precision. There was always an air of indefiniteness, a smoky vagueness surrounding this all-embracing restraining “law” of English constitutionalism. Even its authority as law was shrouded in immeasurability.’[26] On the old view, as Reid puts it, ‘what mattered was not its intrinsic qualities but that it was customary practice, not deliberative decision.’[27] The rule of law tamed unruly exercise of power, because even the sovereign was not above it, indeed not sovereign in Cromwell’s or Blackstone’s or Austin’s sense, but subject to higher law.
The last great defence of that old English conception, Reid argues conscious of the irony, was the American Revolution against the British Crown. In the eighteenth century, the Americans insisted that no government was above the law, but the English had moved beyond them to regard the lawmaker as legally sovereign, outstripping though (and perhaps thus) losing its about-to-be-former colony. The Americans still defended an older understanding of law and the rule of law: ‘In truth, the American Revolution, if understood from the perspective of the development of the concept of rule of law in England and Great Britain should be seen as one of the last – if not the very last – constitutional stands for the old ideal of rule by customary, prescriptive, immutable, fundamental law ... the American Revolution was the greatest triumph for the rule of law.’[28]
It can, then, make sense to speak of the rule of law, and thereby mark a significant distinction between different sorts of polities, without putting some particular institutional recipe at the centre of things. Indeed, it makes sense to talk about the rule of law without saying much about the specifics of legal institutions at all. Take another English example that has intrigued me for a long time: the subject of E.P. Thompson’s famous/notorious conclusion to Whigs and Hunters. Readers will recall that though the bulk of this book is a denunciation of many particulars of that law, in his conclusion Thompson reflects that:

there is a difference between arbitrary power and the rule of law. We ought to expose the shams and inequities which may be concealed beneath the law. But the rule of law itself, the imposing of effective inhibitions upon power and the defence of the citizen from power’s all-intrusive claims, seems to me to me an unqualified human good.[29]

Perhaps fortunately, Thompson was not a lawyer, and unlike Dicey and most other lawyers who write about the rule of law, he did not seek to spell out just what legal elements allegedly produced it. In an ‘I know it when I see it’ way, he insisted upon the ‘obvious point’ that ‘there is a difference between arbitrary power and the rule of law,’ and the latter was identified by what it was claimed to achieve rather than by any recipe or précis of ingredients. Thompson identified the rule of law by the good it did – ‘the imposing of effective inhibitions upon power and the defence of the citizen from power’s all-intrusive claims.’ It was only if and to the extent that law and the rule of law made that sort of difference that it mattered.
And where did he look for evidence of that difference? Well not to particular legal forms, which he thought were constantly being ‘created ... and bent’ by ‘a Whig oligarchy ... in order to legitimise its own property and status.’[30] But that oligarchy could not do as it wished; its hands were often tied by the law it sought to exploit. How did Thompson show this? By describing the character of legal institutions and norms? No. Rather, he called in aid facts such as that ‘[w]hat was often at issue was not property, supported by law, against no-property; it was alternative definitions of property-rights ... law was a definition of actual agrarian practice, as it has been pursued “time out of mind” ... “law” was deeply imbricated within the very basis of productive relations, which would have been inoperable without this law. And ... this law, as definition or as rules (imperfectly enforceable through institutional forms) was endorsed by norms, tenaciously transmitted through the community.’[31] It is facts like these that lead Thompson to declare that ‘the notion of the regulation and reconciliation of conflicts through the rule of law – and the elaboration of rules and procedures which, on occasion, made some approximate approach towards the ideal – seems to me a cultural achievement of universal significance.’[32] ‘Cultural achievement’ is a well chosen phrase.
Of course, analysis of the rule of law must go further than this, if only to check that the claimed good actually existed, and that its purported causes have been well identified. But I think Thompson was right at least to seek his evidence where he did, rather than in contingent descriptions of institutional particulars. Still more, to avoid taking these contingent particular elements, as Dicey appeared to take them, to be the universal essence of the Ding an sich.
Particularly if the Ding is so elusive. Whatever Voltaire admired about the English rule of law, as Dicey boasted, it is unlikely to have been that it was ‘general, equal, and certain.’ Listen (these were lectures, which it must have been a delight to listen to) to Dicey’s contemporary Maitland, on eighteenth century English law. I quote this passage at length since it is so charmingly counter-intuitive, at least counter to Hayekian intuition:

I take up a list of the statutes of 1786. There are 160 so-called public acts, and 60 so-called private acts. But listen to the titles of a few of the public acts: an act for establishing a workhouse at Havering, an act to enable the king to license a playhouse at Margate, an act for erecting a house of correction in Middlesex, an act for incorporating the Clyde Marine Society, an act for paving the town of Cheltenham, an act for widening the roads in the borough of Bodmin. Fully half of the public acts are of this petty local character. Then as to the private acts, these deal with particular persons: an act for naturalizing Andreas Emmerich, an act for enabling Cornelius Salvidge to take the surname of Tutton, an act for rectifying mistakes in the marriage settlement of Lord and Lady Camelford, an act to enable the guardians of William Frye to grant leases, an act to dissolve the marriage between Jonathan Twiss and Francis Dorrill. Then there are almost countless acts for enclosing this, that and the other common. One is inclined to call the last century the century of privilegia. It seems afraid to rise to the dignity of a general proposition; it will not say, ‘All commons may be enclosed according to these general rules,’ ‘All aliens may become naturalised if they fulfil these or those conditions,’ ‘All boroughs shall have these powers for widening their roads,’ ‘All marriages may be dissolved if the wife’s adultery be proved.’ No, it deals with this common and that marriage.[33]

Either eighteenth century England had a strong measure of the rule of law or it didn’t. I think Dicey, Thompson, and Reid are right to think it did. I also think the sources of this blessing need to be sought somewhere else than where lawyers are accustomed to seek them.

3. Modernity

There was a time, 1989 to be precise, when I assumed the rule of law without adjectives was what the world needed and that it was well captured in Lon Fuller’s ‘internal morality’ or something like it. And so it seemed to me that in all the excitement of the collapse of communism, this was a product ripe for export. So I was shocked in that same year to read an article by Edward Rubin which took this alleged morality apart, as it applied, or rather was argued not to apply, to ‘Law and Legislation in the Administrative State’[34] Rubin argues that the bulk of modern legislation is not, as Lon Fuller thought law to be, ‘the enterprise of subjecting human conduct to the governance of rules,’[35] but rather ‘a series of directives issued by the legislature to government-implementation mechanisms, primarily administrative agencies, rather than as a set of rules for the governance of human conduct.’[36] A great deal of modern legislation is ‘internal,’ that is, concerned at least initially with administrative agencies rather than individual citizens. Within ‘external’ legislation, moreover, much is ‘intransitive,’ that is, though concerned ultimately with citizens, it does not specify precisely what rules an agency is expected to apply to them. There is a vast amount of such legislation in the modern state, and it ‘did not arise out of some lapse of moral vigilance. It is central to our beliefs about the role of the government in solving problems and delivering services.’[37] Of this legislation Rubin argues that Fuller’s principles are unhelpful, and ‘[e]ven for transitive statutes, most of Fuller’s principles are persuasive only when the statute relies on courts as its primary implementation mechanism. When a transitive statute is enforced by an agency, our normative system simply does not make the demands that Fuller perceives.’[38] It still makes sense to oppose arbitrary uses of power against citizens, but a great deal of law needs to be thought about in other terms, and where the concern is appropriate, antidotes to it are often likely to be very different from those that Fuller suggests.
Now it might be that a closer reading of Fuller would reveal that he was not setting up universal measures simply to be applied, whatever the form of law, whatever the circumstance. I think that is likely,[39] but my point is different. We don’t have legal institutional recipes that explain the rule of law, even in the places where it is strong. And what we do have is of unclear application to a great deal that modern legislatures do. Why think we have products ready for export?

4. Teleology

So, and particularly for those in the export business, I advocate starting with the ends of the rule of law, rather than what purports to be its institutional anatomy. The overarching end I have focused on, as does so much discussion of the rule of law, is opposition to arbitrary exercise of power. I am uncomfortably conscious that I, and not only I, have yet to provide a satisfactory and sufficiently complex and textured analysis of what arbitrariness includes (caprice? whim? unreasonableness? unreasonedness? discretion? If not all discretion, how much? And so on) and excludes. The concept is key and would repay close attention. I find it easier to give examples than an analysis. That might make the conceptual haziness less dramatic, though, since many of these examples are far from subtle. Stalin provides lots, Saddam Hussein many others. Fuller, away from his formula, and in his explorations of law as a form of social architecture, has sensitively analysed others, closer to home. So too Philip Selznick.[40] Still, it’s a weakness.
It might be, as Gianluigi Palombella has argued,[41] that to focus merely on ‘arbitrariness’ miscasts (or misses) some of the dangers that the rule of law is thought to combat. In particular, the Rechtsstaat tradition which simply takes law to be ‘the structure of the State, not an external limitation to it,’ also seeks to avoid arbitrariness but is less than the rule of law, which extends to the taming of even un-arbitrary ways of acting, of invading jurisdictio. I find Palombella’s argumentation, his insistence that the rule of law adds the protection of elements of right to legal pursuit of the good, extremely persuasive. But if arbitrariness is not the only danger the rule of law is meant to promote, it is a central one. Moreover, if further values are added to my rather lean conception of the rule of law, that would actually strengthen my claim that it is to those values that we should look first, rather than merely to institutional structures that too often threaten to be treated as ends in themselves. And for the meantime, if the edges are blurred, the importance of arbitrariness as an (even if perhaps not the) anti-value among those who have written about the rule of law for centuries is not open to doubt.
I have suggested two reasons to applaud reduction of arbitrariness in the exercise of power, and doubtless there are others. Mine are quite unoriginal, though, and I take that to be a strength. Political theory has reiterated the first for millennia, and economic theory depends on the second, even if not all modern economists are aware of it. Two reasons to welcome reduction of arbitrariness are that it is frightening, and that it is confusing.[42] So, in the contrast between tyranny and government under law, reduction of reasonable fear of power has been a central motivating concern. And given the need that Adam Smith attributed to all members of ‘civilised society,’ of ‘the co-operation and assistance of great multitudes, while his whole life is scarce sufficient to gain the friendship of a few persons,’[43] reduction of arbitrary power and the reasonable apprehension of its exercise, is equally a valuable thing to have.
Where the rule of law is strong, confident interaction and co-ordination among non-intimates are reasonable expectations. These are crucial conditions for a large modern society in good shape. The rule of law can provide fellow citizens with crucial information and security, ‘a basis for legitimate expectations,’[44] by enabling them to know a good deal about each other, though many of them are strangers; to co-ordinate their actions with them; and to feel some security and predictability in their dealings with them. For though not everything can ever be made predictable, much that would otherwise be up for grabs can be tied down. Fixed and knowable points can be established in the landscape, on the basis of which the strangers who routinely interact in modern societies can do so with some security, autonomy, and ability to choose. This can provide a foundation and scaffolding for the building of ‘civil’ relations between state and citizens and among citizens themselves.[45] They can rely upon the state, the law, and each other, not merely live at suspicious or fearful distance from them.
However much we add to, refine, or render precise, the goals that have motivated attention to the rule of law, my suggestion has been and still is (with one significant qualification to be taken up later),[46] that that is where we should start. Goals in view, investigations might begin into how they might be attained. Of course in this no one today is Christopher Columbus. Many people have exercised a lot of thought on these matters. Lessons have been learnt. It would be foolish to ignore them. One very general lesson, very old indeed, is that if you want to avoid arbitrary exercise of power, don’t just trust to luck or virtue. When there is room for those with power to act repressively, they are sooner or later likely to. If you want to avoid it, something must be done and someone must be in a position to do it. And once is not enough, so the ability to restrain the ways in which power is exercised needs to be institutionalised. That is certainly what Montesquieu believed and he was right, very right. Montesquieu also believed that it was not a great idea for all the jobs to be done by the same institutions or for the same people to run all the institutions. These are good ideas too. There have been others.
Thinking about such ideas can yield, in the first instance, some conditions that institutions must satisfy to be able, routinely, reliably, to help avoid arbitrary exercise of power. I have suggested four such general conditions, relying on nothing much more than a combination of intuition and reflection. This level could be explored much more than I have done. Anyway my conditions for institutional contributions to the rule of law have to do with four general criteria. The first concerns the scope of the reach of institutions of restraint – if they are to matter they have to be able to reach those who matter. This must include both state and non-state actors. The significant question is not where they sit but what they can do.[47] The second has to do with the character of the norms that guide, channel and restrain – they have to be such that people can know what they require. This is the basis of those lists of characteristics of norms that lawyers commonly take to be the rule of law. Fuller and Raz have produced influential lists of this sort; Geoffrey de Q. Walker has magnified them.[48] I think the lists are systematically inadequate, as I will argue below, for they imagine that one can read off how laws will be received in societies from lawyers’ intuitions about them, but they have a rationale: unless people can know what the law requires, they can’t abide by it or hold power-wielders to it. Nor can the reciprocity among lawmakers, law-interpreters, and law-receivers work on a common base of understanding and knowledge. Thirdly, there must be a real and knowable link between the norms and the ways they are administered. This will often require complex practices of interpretation, and there is room for real and extensive controversy, but unless the controversy is about what the norms require, you have left the rule of law. Finally, the condition most important of all and least explored in the legal literature, the institutionalised norms need to count as a source of restraint and a normative resource, usable and with some routine confidence used in social life.
The last condition is socially, if not doctrinally, the most important, since unless the norms do count nothing else much matters. We need to know in what ways they need to count, for potential arbitrariness to be diminished. An account of what it means and how it happens that law counts must be developed further than I have done.[49] One way in which laws count is that people, by and large, obey them; and particularly if our concern is to restrain the possibility of arbitrary power, that the powerful do. Why people obey laws, who does and when, are large questions, the answers to which vary greatly between societies, and depend only in part on the character of the laws themselves. Apart from obedience, patterns of use and manner of use are other major sources of distinction between societies where law counts and those where it doesn’t. I am taken with the Bulgarian saying that law is like a door in the middle of an open field. Of course, you could go through the door, but only a fool would bother. Where the saying has resonance, the rule of law is not likely to. We need then to explore what generates circumstances in which the norms do count in these useful, nay precious, ways. You won’t find the answer in Dicey.

5. Sociology

Specification of all these conditions could be refined, and should be. Particularly the last and most important. Whatever we decide them to be, my point is only that these legal conditions themselves depend on conditions that are not legal. For they all have to do with the social reach and weight of law, which are matters of sociology and politics, as much as of law. Indeed social and political questions are central ones to ask about the place of law in a society, and they will be answered differently in different societies, whatever the written laws say or have in common. This is not because the law has no significance, but because the nature and extent of that significance depend on so many factors outside, or underlying, the law itself.
Particularly when these questions are asked by someone concerned with how law might to be encouraged to count where it has not, or in ways it has not, we should keep in mind and generalise Holmes’s observation about Russia:

Lawyers are trained to solve routine problems within routine procedures. They are not trained to reflect creatively on the emergence and stabilization of the complex institutions that lawyering silently presupposes. Ordinary legal training, therefore, is not adequate to the extraordinary problems faced by the manager of a legal-development project in Russia. The problem is not Russian uniqueness and exceptionalism, but the opposite. In Russia, as everywhere else, legal reform cannot succeed without attention to social context, local infrastructure, professional skills, logistic capacities, and political support. .... So legal knowledge alone is never enough.[50]

It’s not enough in practice, but it’s not enough in theory either. Recall Thompson. What was key for him, as it has been for dissidents under countless despotisms, was ‘the imposing of effective inhibitions upon power and the defence of the citizen from power’s all-intrusive clams.’ This is a social and political result, to which law is supposed to be able to contribute and, needless to say, it depends on many things beside the qualities of the formal law. Yet far too often lawyers and philosophers discussing the rule of law move from some legalistic conception of the first three of the conditions distinguished above to the assumption that where they exist so does the rule of law. Which it might, if the law were the single unmoved mover of the social world. Since no one believes that, this move is as odd as it is common.
Take, for example, the second condition, which has to do with knowledge of the law. One a priori hypothesis, for example, extremely common among lawyers and legal theorists, is that whatever contributes to making legal rules less vague, ambiguous, open-ended and renders them more precise, tightly-specified and univocal contributes to making law more certain, and therefore reliable. It seems to stand to reason, after all, that if a rule is sharper, more precise, less open to interpretation, it is easier to understand and follow. Indeed Max Weber built a theory of law’s contribution to capitalism on this premise. He argued that modern capitalism depended on predictability, and that since formal rational civil law promoted the greatest degree of legal generality, clarity and formal certainty, it must be the most predictable. Therefore, maximum formal rationality of law, as found in continental Europe, was indispensable for the rise of modern capitalism; except, as in the great capitalist nations of the nineteenth and early twentieth centuries, England and America, where it was not! Given that these were exemplars of modernity and capitalism this, the so-called ‘England problem,’ was not a small embarrassment for the theory. Driven by the logic of ‘the more the better,’ but chastened by facts he was too observant and too honest to deny, he was led to a series of ad hoc explanations of how Great Britain managed to do capitalism better than anywhere else with a highly ‘irrational’ legal system, compared, say, to the rationality of German law, which accompanied a less developed economy. These concessions were a tribute to his character and powers of observation, but not to this aspect of his social theory.[51]
For, as his contemporary Eugen Ehrlich[52] emhasized, so much that promotes security of expectations is not the doing of formal legal institutions, but of what Ehrlich identified as the ‘living law’ that regulates the lives of communities for so much of the time. The interrelationships between official ‘rules for decision,’ as Ehrlich called them, and ‘living law’ are complex and variable, but there is no reason to believe that ratcheting up the formal rationality of the former will produce its direct and faithful reflection in the latter.
Again, Joseph Raz gives, as one ‘fairly obvious’ reason for preferring rules to principles in the direct regulation of behaviour, that ‘[p]rinciples, because they prescribe highly unspecific acts, tend to be more vague and less certain than rules’ and ‘[s]ince the law should strive to balance certainty and reliability against flexibility, it is on the whole wise legal policy to use rules as much as possible for regulating human behaviour because they are more certain than principles and lend themselves more easily to uniform and predictable application.’[53] On that assumption, numerous advocates of the rule of law insist that it should be a ‘law of rules,’[54] where rules are understood to act as ‘exclusionary reasons,’[55] rather than more open-ended principles, since the former are assumed to be more certain and predictable than the latter.[56] Even those, like Ronald Dworkin, who are fond of principles are so not on the grounds that they are as predictable as rules, indeed they concede that they are not. Dworkin commends them for offering other virtues of justice which a strict regime of rules might thwart.
Yet whether or not precision of legal rules yields certainty of law is a major and unresolved issue of socio-legal investigation. Not only is it unresolved, it is very difficult to resolve, since it is an empirical question for which it is hard to gather evidence. Such evidence as we have suggests, at least to John Braithwaite, that while rules might be more certain than principles in relation to ‘simple, stable patterns of action that do not involve high economic stakes’ - like driving a car - ‘with complex actions in changing environments where large economic interests are at stake’ principles are more likely to enable legal certainty than rules. Indeed, Braithwaite argues, ‘[w]hen flux is great it can be obvious that radically abandoning the precision of rules can increase certainty.’[57] The argument is complex and the evidence, as Braithwaite readily concedes, incomplete and hard to obtain, but his arguments are powerful and the evidence on which he draws, though limited, is strong. A complex order of fixed and rigid rules, for example, is typically more open to ‘creative compliance,’ ‘legal entrepreneurship’ and ‘contrived complexity’ particularly at ‘the big end of town.’ This is both because certain sorts of precise rules, and regimes where such rules predominate, lend themselves to such exploitation more readily than certain sorts of principles and also because ‘there is uncertainty that is structurally predictable by features of power in society rather than by features of the law.’[58] One might speculate that some of the tendencies Braithwaite identifies might even be stronger in less ruly countries than the western, comparatively law-abiding polities (Australia, UK, US) on which this and allied research primarily draws.
Whether or not Braithwaite’s particular hypotheses are confirmed by further work, the point remains that we won’t be able to confirm or deny them without such work. Yet the literature of the rule of law is largely innocent of these sorts of inquiry. Lawyers often stop at the place where social investigation should start, the legal vehicle of transmission, or at a somewhat skewed sample of law-affected behaviour later, where legally relevant bruises and projects are brought to them. They do not regularly investigate those places where legal transmissions are most typically and crucially received and acted upon – in the myriad law-affected everyday interactions of individuals and groups, which go nowhere near lawyers or officials but where law in a rule-of-law society does its most important work.
Moreover, sources of and impediments to legal knowledge differ between societies. So even were lawyers interested and equipped to look more widely, they would still typically only have local knowledge. And since philosophers of law rarely go beyond the writings of lawyers for their data, they have even less to work with: vicarious local knowledge. This would need to be supplemented by comparison and reflection, and of sorts which need to go beyond where lawyers usually feel comfortable looking or philosophers thinking. One does not expect lawyers or philosophers to do something alien to their natures, viz. empirical social research, but it would be gratifying if, once in a while, they acknowledged the signficance of such investigations for so much that they say in ignorance of them.
In any event, whether the law is known or knowable cannot just be read off merely from legal forms. For success in communication of law surely depends on how the law is received, not on how it is expressed or even delivered. And that depends on many – and various - factors that intervene between law and life. But what in a particular society are the sources of and impediments to orienting one’s actions by law are essentially empirical, socio-legal questions to which we have few certain answers. And since we don’t it is odd that lawyers and philosophers are so confident we do.
This is just one example of a more general point, that the successful attainment of the rule of law is a social (broadly understood: it is obviously political and other things as well) outcome, not a merely legal one. What matters, here as everywhere with the rule of law, is how the law affects subjects. But since the distance between law in books and action is often long, the space full of many other things, and in different places full of different things, it is a matter of comparative social investigation and theorization what might best, in particular circumstances, in particular societies, further that goal. A docket of the rechtsstaatlich features of legal instruments, even buttressed by citations to Fuller, Hayek, Raz, or even Weber, will not do the trick.
And even if there were a linear relationship between the formal purity of our legal instruments and the predictability of the law, it is not obvious that social predictability would increase in proportion. Law can only offer us tolerable threshold conditions, not total security or foreseeabililty. That is to say, what people need from the rule of law (together with other things), and what successful institutionalisation of it can help provide, is, first, an adequate shield against the worst sorts of fears, uncertainties and surprises that arbitrarily exercised power can produce and, second, adequate and commonly interpretable cues by which strangers can orient their behaviour and interact with some confidence and mutual understanding. Without such a shield and such cues, life can be intolerable. But nothing can protect us against all surprises, since there are so many that the law cannot control. Nor should we hope for such security, for that would be the life of a prisoner not a free citizen. So we must recognise that more rule of law, above threshold levels, is not necessarily better.
Extremes of achievement are easier to identify than thresholds, but that there are thresholds and that they are valuable should not be controversial. Unceasing cranking up of the clarity, certainty, consistency, etc., of legal provisions is not obviously the only way, nor the best, to deliver what we need the rule of law to deliver. As a corollary, some diminution in these features is not necessarily the beginning of a slide into the abyss.
Again, to move from my second condition to the fourth and most important, the only time the rule of law can occur, when then law might be said to rule, is when the law counts significantly, distinct and even in competition with other sources of influence, in the thoughts and behaviour, the normative economy, of significant sectors of a society. But we don’t know what makes law count.[59] Knowability of legal provisions is obviously only a part of the story. Jurists say little about this large issue, beyond bromides about ‘legal effectiveness’ or, more occasionally, the importance of legal culture or a culture of lawfulness. However, as seekers of the rule of law in societies without it are discovering in many parts of the world, what these generalities depend upon, and even more how to produce them, are mysteries. And, since what works somewhere does not necessarily work in the same way or at all elsewhere, many mysteries.
The notion of legal effectiveness merely hints at the complexity of the conditions of the rule of law, far greater complexity than is needed merely (!) to ensure the effectiveness of a legal order. That is no simple matter either, of course, but one can imagine that, for a while at least, effectiveness might come ‘out of the barrel of a gun.’ But not the rule of law.
If the laws are there but governments by-pass them, it is not the law that rules. So exercises of governmental power must be predominantly channelled through laws that people can know. But governments, as we have seen, are not the only addressees of the rule of law. And for the rule of law to count in the life of its subjects, as important as mere submission to law, or even adequate access to and supply of laws and legal institutions, though far less remarked upon than either, is constraint by demand for, and (often unreflective) use of legal services and resources. [60] Such demand and use extend beyond, and frequently will not involve, direct enlistment of legal officials or institutions. They are manifest in the extent to which legal institutions, concepts, options, resources, frame, inform and support the choices of citizens.
More socially significant than citizens’ (generally rare) direct invocations of official channels, is the extent to which they are able and willing to use and to rely upon legal resources as cues, standards, models, ‘bargaining chips,’ ‘regulatory endowments,’ authorizations, immunities, in relations with each other and with the state, as realistic (even if necessarily imperfect) indicators of what they and others can and are likely to do. For it is a socio-legal truism, which still escapes many lawyers, that the importance of legal institutions is poorly indicated by the numbers who make direct use of them. The primary impact of such institutions, as Marc Galanter has emphasised,[61] is not as magnets for social disputes, a very small proportion of which ever come to them, but as beacons, sending signals about law, rights, costs, delays, advantages, disadvantages, and other possibilities, into the community. Of course it helps if the beacons are bright rather than dim, but that is not all that is needed. It is the job of legal officials to try to make the signals they send clear and encouraging (or, in the case of criminal law discouraging), and of enforcement agencies to try to make them salient. But even when these signals are bright and visible, they are not the only ones that are sent out or received in a society.[62] They can be blotted out by more immediate, urgent, extra-legal, often anti-legal messages, sent from many quarters. Or by discouraging messages, such as that whatever the courts say, it won’t be implemented (often alleged in Russia), or that the courts are less powerful than local patrons (ditto and elsewhere), or that whatever one gets from the courts won’t compensate for the costs, difficulties, delays and even dangers of getting it. And other systems, not always co-operative with the law, come into play. Finally, even after the legal messages have been sent, and not diverted, occluded or misdirected, there are still the receivers, who are nowhere a single entity or homogeneous group but plural, different, self-and-other-directed, within numerous, often distinct, sometimes and in some respects overlapping, ‘semi-autonomous’ groups which affect them, often deeply. Law ‘means’ different things to different ‘communities of interpreters,’ especially since for most of them interpretation of law is not their major interest.
The extent to which citizens are able and willing to use and to rely upon legal institutions to protect and advance their interests varies, again within and between societies and over time. In many times and places, citizens are willing to use the law but excluded from access to it. In others it appears that they are unwilling to make much use even of laws they could use. In yet others, such as the United States, many citizens, perhaps too many, are both willing and able. We know a bit about how to affect the supply of law, but we know a good deal less than we might about how to affect demand for it.
Law never means everything in people’s lives, and it rarely means nothing either. But to speak sensibly of the rule of law as a significant element in the life of a society, the law’s norms must be socially normative. If people know nothing of the law, or knowing something think nothing of it, or think of it but don’t take it seriously, or even, taking it seriously don’t know what to do about it, then their lives will not be enriched by the rule of law (though if it applies to governments they might still be partly protected by it). As to how such normativity might be generated, we have few universal prescriptions worth offering.
Where the law really does count, we can foreshorten the question why, as lawyers commonly do, and answer it in terms of the provisions and institutions of the law. For when the law is socially and politically significant, the legal position will bear closely on the factual position and the hour of the lawyer is at hand. But that is only because what lawyers do not know, the conditions of legal effectiveness, gives significance to what they do, the law. When those conditions are lacking, lawyers’ talk is beside the point. For if no one is listening it does not matter too much what the law is saying.

6. Locating and Relocating

I have used Dicey as something of a whipping boy in this paper and in the articles it draws upon. That was probably because his self-satisfied parochialism irritates me and because he has had such a large influence at least in the English-speaking world. But I have recently come to think there is a reading of him (and of Fuller, for that matter) that is more interesting than the one I have given them, whether or not it is the right one. I have been writing in a context where the rule of law has been proposed to help many societies where it was not strong or long embedded, and where it often faces fierce competition from forces that have no concern with it, and whose major interests allow no accommodation for it. My argument is that responses to such proposals, that begin with the legal-institutional features of success-stories, are for many reasons a bad way to start. Roughly, I have been saying: start with function not with form. However, it is arguable that some of the greatest success stories of the rule of law started the other way around. No one designed them from the ground up; typically they were inherited, occasionally tinkered with and at least once, in the United States, tinkered with greatly and to great effect. Their rule of law was not a grand rationalist program of institutional design, but what Michael Oakeshott has called the ‘pursuit of intimations’[63] of existing, sometimes very old, traditions. I don’t think that is the only way that institutions can develop, but a society is very lucky indeed when it has good institutional intimations to pursue.
In those circumstances, exploring how we do things here might well start better by exploring the genius loci, to begin with existing forms, try to understand what they do, and how they have come to do what one values in what they do. Perhaps that is all that Dicey, and Fuller when he examined the forms and limits of different legal practices with which he was familiar, sought to do. If so, I think that is a locally important and valuable enterprise, even if limited in scope because it rests with only part of the whole. For in a sense this is an exception that proves the rule. The reason one is better off to start with forms that are established and of which one approves, with however many qualifications, is that one already has a live and healthy organism, the further growth of which one supports. If we don’t have the full story of its conditions of life, even if we can’t fully understand how it works, that doesn’t really matter, since it’s done fine before us and without us. Taken as a basis for transplantation to a landscape full of organisms one wishes to transform, cadavers, clones, embryos, and other species, some quite voracious, the anatomical/legalistic approach to transplanting the rule of law seems to me in principle misguided.
Relocating the rule of law has its own rigours. Not every locus has a genius for the rule of law. The intimations of local traditions must be explored even if they are odious, both because unless they are taken into account there will usually be a price to pay, and they may not be hospitable to every import. Moreover, much that goes to make up the intimations of rich and complex institutional traditions will not accompany their institutional emanations when the latter are asked to travel. For institutions rest on and are interwoven with traditions, interpretations and understandings that are themselves not easily identified or transported. They might depend, too, upon ingredients which are not found in the donee country or be nullified by ones which are. Function, in other words, will not necessarily follow form; so we cannot avoid the difficult task of working the other way around, with all the attention to particularity, local history and traditions, and allowance for variety, that that implies.

[1] Lon Fuller’s chapter, ‘the concept of law.’, in his The Morality of Law, (New Haven, Yale University Press, 1969) begins with this epigram.. ‘Forgetting purposes is the commonest form of stupidity,’ is H.L.A. Hart’s translation in his review of The Morality of Law, (1965) 78 Harvard Law Review, 1291.

[2] I borrow the phrase from Karol Sołtan. He has used it of Philip Selznick, Lon Fuller, and Charles Anderson, alleged pioneers of such a science, that of ‘civics’ (see his ‘Selznick and Civics,’ in Robert A. Kagan, Martin Krygier and Kenneth Winston, eds, Legality and Community. On the Intellectual Legacy of Philip Selznick, (New York, Rowman & Littlefield, 2002) 357; or ‘eunomics’ (see his ‘A Social Science That Does Not Exist,’ in Willem J. Witteveen and Wibren van der Burg, eds, Rediscovering Fuller. Essays on Implicit Law and Institutional Design, (Amsterdam, Amsterdam University Press, 1999) 387). Since Selznick and Fuller are arguably also pioneers in the sociology of the rule of law, it is not plagiarism but merely respectful homage to have borrowed the phrase from Sołtan.

[3] Among them: ‘Institutional Optimism, Cultural Pessimism and the Rule of Law,’ in Martin Krygier and Adam Czarnota, eds, The Rule of Law after Communism, (Aldershot, Ashgate, 1999) 77-105, ‘The Rule of Law’ International Encyclopedia of the Social and Behavioral Sciences, editors-in-chief Neil J. Smelser and Paul B. Bates, (Oxford, Elsevier Science, 2001) vol.20, 13403-408; ‘Transitional Questions about the Rule of Law: Why, What, and How?’ (2001) 28, part 1, East Central Europe/L’Europe du Centre-Est, 1-34; ‘The Grammar of Colonial Legality: Subjects, Objects and the Rule of Law,’ in Geoffrey Brennan and Francis G. Castles, eds, Australia Reshaped. Essays on 200 Years of Institutional Transformation, (Cambridge, Cambridge University Press, 2002) 220-60; ‘False Dichotomies, Real Perplexities, and the Rule of Law’ in András Sajó, ed., Human Rights with Modesty. The Problem of Universalism, (Leiden/Boston, Martinus Nijhoff) 2004, 251-277; .’Rethinking the Rule of Law after Communism,’ in Adam Czarnota, Martin Krygier, and Wojciech Sadurski, eds, Rethinking the Rule of Law after Communism, (Budapest, Central European University Press, 2005) 265-77; ‘The Rule of Law. An Abuser’s Guide,’ in András Sajó, ed., The Dark Side of Fundamental Rights, (Utrecht, Eleven International Publishing, 2006) 129-161.

[4] Argued in ‘False Dichotomies ...’ n.3 supra.

[5] ‘Political Theory and the Rule of Law,’ in Political Theory and Political Thinkers, (Chicago, University of Chicago Press, 1998) 26.

[6] Introduction to the Study of the Law of the Constitution, 10th edition 1959 (first edition 1885), (London, Macmillan) 188.

[7] Ibid., 193.

[8] Ibid, 195-96.

[9] See Ivor Jennings, The Law and the Constitution, (5th edition, University of London Press, London, 1959).

[10] Cf. Judith Shklar on ‘Dicey’s unfortunate outburst of Anglo-Saxon parochialism ... The Rule of Law was thus both trivialised as the peculiar patrimony of one and only one national order, and formalised, by the insistence that only one set of inherited procedures and court practices could sustain it.’, op. cit., 26.

[11] Ibid., 187.
[12] The Constitution of Liberty, (London, Routledge and Kegan Paul, 1960) 149-50.

[13] For a good discussion in this spirit, see Joseph Raz, ‘The Rule of Law and Its Virtue,’ in The Authority of Law, (Oxford, Clarendon Press, 1979,) 210-29.

[14] Cf. Martin Krygier, ‘Marxism and the Rule of Law. Reflections on the Collapse of Communism,’ (1990) 15 Law and Social Inquiry, 633-63 at 637.

[15] T. G. Ash, "Eastern Europe: The Year of Truth," New York Review of Books, February 15 (1990) p. 21.

[16] In a workshop on ‘Rethinking the Rule of Law after Communism.,’ European University Institute, Florence, 2003.

[17] See two good collections, Erik Jensen and Thomas C. Heller, eds., Beyond Common Knowledge. Empirical Approaches to the Rule of Law, (Stanford, Stanford Law and Politics, 2003) and Thomas Carothers, ed., Promoting the Rule of Law Abroad. In Search of Knowledge, (Washington, Carnegie Endowment for International Peace, 2006).

[18] Erik G. Jensen and Thomas C. Heller, eds, op. cit., , 1-2.

[19] ‘The Illusory Promise of the Rule of Law,’ in András Sajó, ed. Human Rights with Modesty. The Problem of Universalism,, (Leiden/Boston, Martinus Nijhoff, 2004) 281.

[20] For another example of this style of argument, see R.P. Peerenboom, ‘Human Rights and Rule of Law: What’s the Relationship?’ (2005) 36 Georgia Jounal of International Law, 809.

21. [21] Bracton, On the Laws and Customs of England, vol.2, 305-6, quoted in John Philip Reid, Rule of Law, (Illinois, University of Northern Illinois Press, 2004) 11.

[22] Op. cit., 184.

[23] See Martin Krygier, ‘Common Law,’ Routledge Encyclopedia of Philosophy, General Editor Edward Craig, vol. 1, 440-46 (London, ROutledge, 1998).
[24] Commentaries on the Laws of England, vol. 1, (Chicago, Chicago University Press, 1979) 17.
[25] Ibid., 68.

[26] Reid, op. cit., 16.

[27] Reid, op. cit., 13.

[28] Op. cit., 75.

[29] Whigs and Hunters. The Origin of the Black Act, (Harmondsworth, Penguin, 1977) 266.

[30] Ibid., 260-61.

[31] Ibid., 261.

[32] Ibid., 266.

[33] The Constitutional History of England, (Cambridge, Cambridge University Press, 1965) (first edition 1908), 383. I am grateful to Mark Aronson for bringing this passage to my attention.

[34] (1989) 89 Columbia Law Review, 369-426.
[35] Lon L. Fuller, The Morality of Law, Yale University Press, New Haven, 1969, 106.
[36] Op. cit., 371-72.

[37] Ibid., 406-07.

[38] Ibid., 399.

[39] See his The Principles of Social Order, revised edition edited by Kenneth I. Winston, (Oxford, Hart, 2001) and the insightful introductory essays by Winston.

[40] In his discussion of the principles of due process, near the end of Law, Society, and Industrial Justice, Selznick suggests some examples: ‘Rule-making that is based on evident caprice or prejudice, or that presumes the contrary of clearly established knowledge violates due process. Procedure cannot be ‘due’ if it does not conform to the canons of rational discourse or if it is otherwise outside the pale of reasoned and dispassionate assessment. Thus legislative classification of persons or groups may be struck down as arbitrary and against reason if they have no defensible connection with, or inherently frustrate, the professed aims of the legislation. Similarly a host of administrative actions, though they may enjoy large grants of discretion, are subject to this ultimate appeal.’ (New Brunswick, Transaction Books, 1969), 253)
[41] Personal communication. And see his article in this volume.

[42] I elaborate on these two reasons in ‘The Rule of Law,’ and ‘Transitional Questions ...,’ n.3 supra.

[43] An Inquiry into the Nature and Causes of the Wealth of Nations, (Indianapolis, IN, Liberty Fund, 1981) vol. 1, 26.

[44] J. Rawls, A Theory of Justice, (Cambridge Massachusetts: Harvard University Press, 1971), p. 238[45]

[46] See Locating and Relocating, below.

47. [47] Cf., G. O’Donnell, ‘Polyarchies and the (Un) Rule of Law,’ in The (Un) Rule of Law and the Underprivileged in Latin America, eds. J.E. Méndez, G. O’Donnell and P.S. Pinheiro, (Notre Dame, Indiana University of Notre Dame Press, 1999), 318: ‘if the legal system is supposed to texture, stabilise, and order manifold social relations, then not only when state agents but also when private actors violate the law with impunity, the rule of law is at best truncated. Whether state agents perpetrate unlawful acts on their own or de facto license private actors to do so, does not make much difference, either for the victims of such actions or for the (in) effectiveness of the rule of law.’

[48] Geoffrey de Q. Walker, The Rule of Law. Foundation of constitutional democracy, (Carleton, Victoria, Melbourne University Press) 1988.

[49] I have a few guesses in ‘Transitional Questions ...,’ supra n.3 at 12-18, but the work is left to do.

[50] ‘Can Foreign Aid Promote the Rule of Law?,’ (1999) 8, 4 East European Constitutional Review, 71.

[51] This argument is developed at greater length in my ‘Ethical Positivism and the Liberalism of Fear,’ in Tom Campbell and Jeffrey Goldsworthy, eds, Judicial Power, Democracy and Legal Positivism, (Aldershot, Ashgate, 2000) 73-77. It gains further, even poignant, support from analyses then unknown to me that suggest the common law is superior to civil law in supporting economic growth. See Frank B. Cross, ‘Identifying the Virtues of the Common Law’ http://papers.ssrn.com/sol3/papers.cfm?abstract_id=812464, and references cited there.

[52] See his Fundamental Principles of the Sociology of Law, with introduction by Roscoe Pound and a new introduction by Klaus A. Ziegert, (New Jersey, Transaction Publishers, 2002), first published in English 1936, and in German 1913.

[53] ‘Legal Principles and the Limits of Law,’ (1972) 81 Yale Law Journal 823 at 841.

[54] Cf., Antonin Scalia, “The Rule of Law as a Law of Rules” (1989) 56 University of Chicago Law Review 1175. This is the central theme of Tom Campbell’s The Legal Theory of Ethical Positivism, (Aldershot, Dartmouth, 1996). Campbell’s ‘ethical positivism’ is ‘an aspirational model of law according to which it is a presumptive condition of the legitimacy of governments that they function through the medium of specific rules capable of being identified and applied by citizens and officials without recourse to contentious personal or group political presuppositions, beliefs and commitments.’ (2)

[55] Joseph Raz, Practical Reasons and Norms, (London, Hutchinson, 1975), 15-84. See Campbell, op. cit., 5: ‘a system of law ought to be a system of rules. Further, the rules in question must be ‘real’ rules, that is rules which have, in Raz’s term, “exclusionary force”.’

[56] See Campbell, Scalia, Geoffrey de Q. Walker, op. cit., P.S. Atiyah, From Principles to Pragmatism: Changes in the Function of the Judicial Process and the Law, Inaugural Lecture, delivered at Oxford University, Febrary 17, 1978, and published by the Clarendon Press.

[57] ‘Rules and Principles: A Theory of Legal Certainty,’ (2002) 27 Australian Journal of Legal Philosophy, 54.

[58] Ibid., 58-59.

[59] For some intelligent, still controversial and unsettled, speculations in a particular context, see Kathryn Hendley, Stephen Holmes, Anders Åslund, András Sajó, ‘Debate: Demand for Law,’ (1999) 8, 4 East European Constitutional Review 88-108. Cf. also ‘Citizen and Law after Communism, (Winter 1998) 7, 1 East European Constitutional Review, 70-88 and Ilian G. Cashu and Mitchell A. Orenstein, ‘The Pensioners' Court Campaign: Making Law Matter in Russia’ with reply by Kathryn Hendley, ‘”Demand” for Law – A Mixed Picture,’ (Fall 2001) 10, 4 East European Constitutional Review.

[60] See Hendley et al, ‘Debate: Demand for Law.’

[61] See his ‘Justice in Many Rooms: Courts, Private Ordering, and Indigenous Law,’ (1981) 19 Journal of Legal Pluralism, 1-47. As Galanter observes, “[t]he mainstream of legal scholarship has tended to look out from within the official legal order, abetting the pretensions of the official law to stand in a relationship of hierarchic control to other normative orderings in society. Social research on law has been characterised by a repeated rediscovery of the other hemisphere of the legal world. This has entailed recurrent rediscovery that law in modern society is plural rather than monolithic, that it is private as well as public in character and that the national (public, official) legal system is often a secondary rather than a primary locus of regulation” (at 20).

[62] For a classic statement of these points, see Sally Falk Moore, ‘Law and social change: the semi-autonomous social field as an appropriate subject of study,’ in Law as Process, (London, Routledge and Kegan Paul, 1978) 54-81.

[63] ‘In politics, then, every enterprise is a consequential enterprise, the pursuit, not of a dream, or of a general principle, by of an intimation.,’ Michael Oakeshott, ‘Political Education,’ in Rationalism in Politics and other essays, new and expanded edition, (Indianapolis, IN., Liberty Press, 1991) 57. And see 66-69, ‘The Pursuit of Intimations.’