DAVID DYZENHAUS - Rule Law as Rule Liberal Principle

7/28/2019 DAVID DYZENHAUS - Rule Law as Rule Liberal Principle

1/28

Ronald Dworkin

Edited byARTHUR RIPSTEINUniversity ofToronto

CAMBRIDGEUNIVERSITY PRESS


2/28

CAMBRlDGE UNIVERSITY PRESSCambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao PauloCambridgeUniversityPress32 Avenue of the Americas,NewYork, NY 10013-2473, USAwww.cambridge.orgInformation on this title: www.cambridge.org/9780521664127 Cambridge UniversityPress 2007This publication is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,no reproduction of any part may take place withoutthe written pennission of Cambridge University Press.First published 2007Printed in the United States ofAmericaA cawlog record for this publication is available franz the British Library.Lilrrary ufCongress Cataloging in Publication DatoRonald Dworkin I edited byArthur Ripstein.

p. em. - (Contemporary philosophy in focus)Includes bibliographical references and index.ISBN-l3: 978-0-521-66289-5 (hardback)ISBN-10: 0-521-66289-3 (hardback)ISBN-l3: 978-0-521-66412-7 (pbk.)ISBN-lO: 0-521-66412-8 (pbk.)1. Dworkin, Ronald. 2. Law-Philosophy. 3. Law- United States-Philosophy.I. Ripstein, Arthur. II. Title. ill. Series.K230.D92.R66 2007340'.1 -del2 2006038179ISBN 978-0-521-66289-5 hardbackISBN 978-0-521-66412-7 paperbackCambridge UniversityPress has no responsibility forthe persistence or accuracy ofURLs for external orthird-party InternetWeb sites referred to in this publicationand does not guarantee that any content on suchWeb sites is, orwill remain, accurate or appropriate.


3/28

2 The Rule of Law as the Ruleof Liberal PrincipleDAVID DYZENHAUS

,I

[TJhe "rights" conception [of the rule of law] assnmes that citizens havemoral rights and duties with respect to one another, and political rightsagainst the state as a whole. It insists that these moral and political rights berecognized in positive law, so that they may be enforced upon the demllnd ofindividual citizens through courts or other judicial institutions of the familiartype, so far as this is practicable. The rule of law on this conception is theideal of rule by an accurate public conception of individual rights. It doesnot distinguish, as the rule-book conception does, between the rule of lawand substantive justice; on the contrary it requires, as partof the ideal oflaw,that the rules in the rule book capture and enforce moral rights. l

Ronald Dworkin

INTRODUCTION

In popular opinion, the rule of law is essential to legitimate government.Philosophers of law do not dispute popular opinion but divide over anambiguity in it. They agree that a legitimate government is one that, amongother things, operatesunder the constraints ofthe rule oflaw.They disagreeabout whether the constraints suffice to make government to some extentlegitimate. As we will soon see, this disagreement turns in part on politicalconsiderations about the role of judges in democracy. But it also turns onthe debate between positivists and antipositivists about the nature of law.IT, as antipositivists argue, there is some necessary connection between lawand morality, the rule of law is to some extent the rule of good law. IT, aspositivists argue, the rule of law is simply the rule of the rules certified asvalid in a particular legal order, it is the rule of those rules with the contentthey happen to have.Positivists concede that the rule of law makes a difference to government. Rules are applied in accordance with the requirements of strict

56


4/28

The Rule of Law as the Rule of Liberal Principle 57

legality so that official coercion is authorized only if there is a clear legalwarrantfor it, likecases are treated alike, and so on.Under a regimeoflegitimate government, it is important that strictlegalityis byand large observed,and thus legitimate government requires the rule of law. But, positivistsmaintain, an oppressive regime can also operate under the constraints ofthe rule oflaw. Since its rules will be rigorously enforced, its rulemight bemorally worse for its subjects than that of an equally evil but disorganizedregime.

Ronald Dworkin's account of the rule of law as the rule of liberal prin-ciples is the most recent in a long line of distinguished attempts to loosenpositivism's grip on legal theory. Its power is evidenced by the fact thatpositivists have found themselves forced to concede that the rule of law isnot just the rule of rules, since it is also the rule ofprinciples. As Dworkinargues, such principles playa central role in judicial decision of"hard cases,"cases where there is reasonable disagreement aboutwhat the law requires.

Positivists do not, however, suppose that that concession damages thefundamental tenet oflegal positivism- the "separationthesis," whichassertsthat there is no necessary connectionbetween law and morality. They thinkthat Dworkin fails to show that the rule ofprinciples is necessarily the ruleof liberal principles, thus illustrating their contention that any attempt toshow that the rule oflaw is the rule of good law must fail.

In the next section, I identifY the main problems that an antipositivistaccount of the rule oflaw faces. "With one exception, they come in a packageof positivist objections to the tradition of judicial antipositivism to whichDworkin belongs.Then, in a long expositorysection, I show thatDworkin'saccount of the rule oflaw cannot surmount these problems, and thus I closeby arguing that his account should be modified.

But before I embark onmy account ofa complex debate, I want to notea curious feature of it. Most of the principal figures in philosophy of lawin the twentieth century paid little attention to the issue of the rule of lawbecause for them that issue was approached by :first determining what thelaw is that rules, after which there is little left to say about the 17tle of law.Wewill see thatin Dworkin's case there seems little left to say, which iswhythe essay from whichmy epigraph is taken is a rare example of focus on therule oflaw.2 It is more true ofhis position than ofany other that his accountof the rule oflaw is an account of the particular theory of liberal justice thathe argues is embeddedwithin the law. I will narrow the scope ofdiscussionby focusing on three central objections to Dworkin's theory of the ruleoflaw.


5/28

The first positivist objection to an antipositivist account of the role oflaw isthat it is politically dangerous. To suggest that the role of law is the role ofgoodlawis to support oppressiveregimes in their claim to legitimacymerelybecause they observe the role oflaw. But this is not the onlypolitical dangerpositivists detect in antipositivism, at least in the judicial antipositivismof the common law tradition to which Dworkin belongs, a tradition tJ.1atmaintains that the common law is the repository of the morality of the lawand that judges are the guardians of its principles.

JeremyBentham regarded this view as the main weapon in the armoryof the self-aggrandizing, usurpatory legitimacy claimsmade by judges.Hispositivism was clearly put in the service of debunking the claims of judgesminded to uphold a private order that privileged property-owning elitesin the face of democratic legislative reforms. This political association ofpositivismwith democracy, and apreference for stamtes as theprimaryformoflaw against liberal antipositivists whomay seem to give an exalted role tojudges and the common law, is a constant theme in debates about the role oflaw. It informs the critique of judges in whatwe might think of as populistconceptions of democracy, which require law to be nothingmore than theexpression ofwhat the people in fact want, unconstrained insofar as thisis possible by formal legal requirements and completely unconstrained byjudicially interpreted principles.

The main stream of contemporary positivism, however, followsH. L. A. Hart in detaching legal positivism from democratic theory, indeedfrom any particular political commitments. Although contemporary positivism stresses the political dangers ofantipositivism, it ultimatelyrelies ona conceptual reason for adopting positivism- the need for a general theoryof law.3 A theory of lawmust be general not only in that it can account forthe fact that law can be used as an instrument ofoppression. Itmust also beable to accountfor the fact that legal orders have verydifferent institutionalfeatures, for example, the institution of constitutional review of theUnitedStates ofAmerica and theWestminstermodel ofparliamentarysupremacy,which requires judicial deference to clearly expressed legislative intention.

So a second objection to antipositivist accounts of the role of law isnot so much political as theoretical. I t is that such accounts are parochial,unable to offer a general account of the law. They rely on features oflegalsystems that are entirely contingent - for example, a written constitutiongiving judges the authority to test legislation against a bill of liberal rightsand freedoms.

58

THREE OBJECTIONS TO DWORI


6/28

The Rule of law as the Rule of Liberal Principle 59

The structure of the positivist claim for superiority over antipositivism makes life difficult for judicial antipositivism. Positivists distinguishbetween a general theoryoflawand theories ofadjudication that tell judgeshow to decide hard cases. They also claim that in hard cases the positivelaw does not supply an answer. Since the positive law supplies no answer,and since law is positive law, the question is settled by the discretion ofthe official with final authority. In otherwords, the official, who may be ajudge, bas to legislate an answer. An accouut of the proper exercise ofsuchdiscretion is a theory of adjudication whose details will depend both onthe institutional makeup of the particular legal order and on political arguments about the considerations onwhich judgesmay legitimately rely.Theparticularity of such a theory ofadjudication tells us that it cannot providea theory of law given that atheory oflawmust aspire to be general. In sum,judicial antipositivism is more than theoretically inadequate and politicallysuspect; it does not even live in the same space as a general theory oflaw.4

The last problem for judicial antipositivism isnot so much an objectionput by positivists as one to which their theory of law is not vulnerable.However, it is damaging to judicial antipositivism.

The central issue of the rule of law in the twentieth century stemmedfrom changes in the formoflaw- the growth oftheregulatorystate,broughtaboutby legislative creation ofadministrative agencies that are oftenhybridin nature since they combine legislative and adjudicative functions. Theseagencies can be given the task of developing the legislation - the policymandate that they are charged with implementing as well as the task ofadjudicating conflicts that arise about the interpretation of that legislation.Usually, the officials who staff these agencies will get their authority froma statute, and then the question is what are the limits set by the statute orby other sources of law, for example, the constitution or the common law.

In hard cases about the scope ofofficial authority, the statutorygrant ofauthority to the official is usually couched in discretionary terms. The official, or panel ofadjudicators when an administrative tribuual has to decide,will get its authority either explicitly or by clear implication from theislature. So for positivists it follows that in this situation - the ordinarysituation that faces common law judges in judicial reviewwithin administrative law - judicial discretion as to what the law requires is piled on topof administrative discretion on the same issue.

All that positivist legal theory can say about this situation is that it issbot through with discretion. However, this is not a concern for positivismsince it does not assume a division ofpower between the legislature, whichmakes the law; the judiciary, which interprets it; and the administration,


7/28

60 David Oyzenhaus

which implements it.5 Since positivists maintain that judges make as wellas apply the law, the change in the fonn oflaw seems to positivists to be no.change at all butmerely the addition ofanother hybrid body to many legalorders.Dworkin, in contrast, does generally assume a strict division of powersbetween legislature, judiciary, and administration, whereby the legislaturehas a monopoly on making law, the judiciary on interpreting the law, andthe administration merely implements the law. Hence, the change in t;hefonn oflaw provides the basis for a third objection to his theory of the ruleoflaw.

THE RULE OF LIBERAL PRINCIPLES

A. An Attack on Two FrontsIn his "Introduction" to Taking Rights Seriously, Dworkin undertook to supply a "general theoryoflaw" thatwould unseat "the ruling theoryoflaw" oflegal positivism.6 Bygeneral theoryhemeant something different from thepositivist thought that a general theory of law does not depend on contingent institutional features or political ideologies. For him, a general theoryunites a nonnative - or politicallyprescriptive- partwith a conceptual part.Bentham, Dworkin said, was the "last philosopher in the Anglo-Americanstream to offer" such a general theory oflaw, one that united positivismwiththenonnative theoryofutilitarianism. Dworkin sawhis task as anargumentagainst the two streams that flow on seemingly separate7 courses fromBentham's general theory: the conceptual legal positivist stream, whichholds that one's legal rights are simplythose the positive lawexplicitlystates;and the economic/utilitarian stream, which holds that the only ground forprotesting a legislative decision is that it does not in fact serve the generalwelfare.

In the "Introduction," Dworkin outlined the distinction for whichTaking Rights SerioZlsly is bestknown, namely, the distinction betweenpolicyand principle, according to which the role of judges is to be the guardiansof the moral "principles" immanent in the law, whereas legislatures makedecisions about policy or collective welfare - decisions that are legitimateunless they violate principles. But he also said that another distinctionwas more fundamental one between "two fonns ofpolitical rights":

background rights, which are rights that hold in an abstract way againstdecisions takenby the communityor the society as awhole, and morespecific


8/28

The RuLe of Law as the RuLe of liberaL PrincipLe 61

institutional rights that hold against a decisionmade bya specificinstitution.Legal rights may then be identified as a distinct species of a political right,that is, an institutional right to the decision of a court in its adjudicativefunction.8To appreciate the importance of this distinction, recall the positivist

claim that the rule oflaw does not guarantee legitimate rule. That claim isstronglysupported ifany attemptat refutation cannot stop shortofequatingthe rule of law with the rule of good law, meaning rule by a full theory ofliberal justice.

To succeed, then, an antipositivist theory has to showmore than that therule oflaw is the rule ofprinciples. It has to show- withoutmaking the fatalequation - that the principles of the rule of law are sound moral principles.Hence it is the distinction between background and institutional rights,rather than the one between principle and policy, that "provides a bridgebetween the conceptual and normative parts,,9 ofDworkin's own theory.

I t is clear from the way Dworkin drew the distinction that judges are atcenter stage in determining institutional rights. Since he wanted to highlight a principled basis for the decisions of courts "in their adjudicativefunction," the rule of law is the rule of institutional rights based on theprinciples exposed in adjudication. I t is rule by judges based on their interpretation of the principled basis of their law. This rights model of the rule oflaw- the rule of liberal principles over which judges have an interpretativemonopoly- must seem like judicial antipositivism written as large as it canbe. And so Dworkin hastened to assure the reader that the bridge his "normative theory ofadjudication" provides is one that also shows that "judicialdecisions based on arguments of principle are compatible with democraticprinciples."lO

B. The ConceptuaL ChallengeDworkin's conceptual challenge to legal positivism demonstrates the pervasive role ofprinciples in adjudication. Since this challenge is the topic ofother essays in this volume, and since positivists responded byacknowledgingthat principles playa pervasive role, I want simplyto note themain formof the positivist response - that the operation of principles is the subjectmatter for a theory of adjudication, not a theory of law. The question ofwhat principles are available will depend on contingent features of a particular legal order, and their role in adjudication will depend on a politicaltheory about how judges should exercise their discretion.


9/28

62 David Oyzenhau5

Tills response is problematic since it accounts for principles by openingup an ever-widening gap of discretion for judges and thus may seem evenmore suspect from the point of view of a populist democratic politicaltheory than Dworkin's attempt to build a theory of law out of a theory ofadjudication. ll ButDworkincannotprovidesufficientgrounds for rejectingthe response, unless he can show that the right to a particular decision in ahard case, and thus the judge's duty to give thatdecision, is both amoralandalegal right.And this is so even if the contemporarynormative developme.ntsofBentham's utilitarianism arewrong,for Dworkin's theory shows that in ademocracy individuals have rights that "trump" collective judgments aboutwelfare.12 Such a theoryshows only that rights should be recognized in law,not that they are already recognized. Yet it is the latter result thatDworkinneeds ifhe is to show that the role of principles goes beyond a concepmalchallenge to positivism, because only that resultestablishes the link betweenlaw and morality that positivism denies.

C. A Bridge Too Far?Dworkin's foundation for ills bridge is the distinction between backgroundand institutional rights. It is supposed to show the difference between thequestions "What are moral rights, whatever the law says?" and "What arethe moral rights under the law of this particular legal order?"

Dworkin sets out the bridge in "Hard Cases." Again ills foil is legalpositivism, described as making two main claims. First, positivism putsjudges "in the shadow of legislation" - they should "apply the law thatother instimtionshavemade; they shouldnotmakenew law." Second, whenjudges have to make law because of the inherent vagueness of legal rules,they should act "not only as deputy to the legislature, but as a deputylegislature" and make "law in response to evidence and arguments of thesame character as would move the superior instimtion i f i twere acting onits own." Tills, says Dworkin, "is a deeper level of subordination because itmakes anyunderstandingofwhatjudges do in hard cases parasitic on apriorunderstandingofwhatlegislatorsdo all the time. Tills deepersubordinationis thus concepmal as well as political."13

H, however, judges should decide on the basis of principle rather thanpolicy, they move Dworkin supposes -legitimately out of the shadow ofthe legislature. The legal order is one inwillch there is no subordination ofjudges to the legislature, but there is a division aflabar between the legislature,whose province is policy, and the judiciary,whose province is principle.


10/28

The Rule of Law as the RuLe of liberaL PrincipLe 63

Legislatures may rely on arguments ofprinciple, but theymay also rely onpolicy considerations alone.

According to Dworkin, a democratic objection to an enhanced judicialrole has most force ifwe suppose that judges' decisions are based on policy.Judges are not in as good an institutional position as legislatures to adjudicate competing claims of interest, and they should not determine people'srights retrospectively on the basis, say, of an assessment ofwise economicpolicy. But if judges decide on the basis ofprinciple, it makes sense to insulate them from "the demands of the political majority." Judicial decisionsenforce "existing political rights" because the principles onwhich they arebased are to be found in "institutional history." But institutional historycannot by itself supply the answer - the judge has to make a "fresh judgment" about what those rights are, one which engages his or her sense ofjustice at the same time as his or her understanding ofhistory.14

This claim about engagement is important because it opposes the ideathat judges have discretion. "Judges," says Dworkin, "do not decide hardcases in two stages, first checking to see where the institutional constraintsend, and then setting the books aside to stride off on their own." Rather,there is a "pervasive interaction" of "personal and institutional morality.,,15

Judges, like "all political officials," are subject to a doctrine of politicalresponsibility that requires them to make "only such political decisionsas they can justifY within a political theory that also justifies the otherdecisions they propose to make." This requirement is one of "articulateconsistency," demanding of judges arguments of principle, which showthat the principle cited is "consistent with earlier decisions not recanted,and with decisions that the institution is prepared to make in hypotheticalcircumstances."16

Dworkin summarizes his claims in the "rights thesis," the thesis that aparty has a right in a hard case to the decision supported qy principle. Thisis not a right to the decision that would be given by "background rights,"the "rights that provide a justification for political decisions by society inthe abstract," but a right to the decision in terms of "institutional rights,that provide a justification for a decision by some particular and specifiedpolitical institution." And within both categories of right, he distinguishesbetween "abstract" and "concrete" rights, where the former are generalpoliticalaims, whereas the latteraremorepreciselydefined to "expressmoredefinitely the weight they have against other political aims on particularoccasions."17 The rights thesis provides, then, that judges decidehard casesby "confirming or denying concrete rights," which are institutional rights


11/28

64 David Dyzenhau5

rather than background rights, and legal rights "rather than some otherform of institutional rights."18

But though theyare institutional rights, theyare, Dworkin asserts, genuine rights despite the fact that institutional autonomy will insulate an"official's institutional duty from the greater part of background politicalmorality.,,19 The judge who accepts the "settled practices of his legal system .. .must, according to the doctrine of political responsibility, acceptsome general theory that justifies these practices."zo .

In "Hard Cases," Dworkin examines three kinds of legal materialwritten constitutions, statutes, and the common law. Here I confine myselfto his discussion of statutes and the common law since an antipositivistconception of the rule of law cannot rely on the contingent presence of awritten constitution.

Even if Dworkin need not rely on the existence of a written constitution, he does rely on the claim that a judge in any legal order requiresa "constitutional theory." The judge's approach to statutes and to precedent - his or her doctrine of political responsibility - will be informed bya constitutional theory, in the sense of an overarching theory ofwhat justifies his or her approach to both statute and precedent as authoritativesources oflaw.21 This bare understanding of constitutionalismmoves oneout of the terrain ofconceptual challenge and into the terrain ofnormativepolitical theory. Dworkin's foil here is notHart's conceptual positivism butBentham's general theory. And that theory is a foil to Dworkin's judicialantipositivism since it is a theory oflegislative positivism, one that seeks tosubordinate judges as far as possible to the will of the majority, expressed asclearly as possible in the only form oflaw that Bentham thought politicallyresponsible - statutes.

What then is the constitutional theory that informs judicial antipositivism? According to Dworkin, a judge interpreting a statute has to askwhich interpretation of the statute best "ties the language the legislatureused to its constitutional responsibilities."z2 The policy of the statute isdetermined by a combinationofinterpretation of the language used by thestatute with a sense of the legal rights of the individuals affected by plausible interpretations. The judge must select the interpretation that best livesup to his or her sense of the legislature's constitutional responsibilities.Dworkin emphasizes that this is not a theory ofwhat the legislature wouldhave done or shouldhave done had it contemplated the particular situation,but ofwhat it did do, with the limits ofwhat it could plausibly be said tohave done set by the words of the s.tatute.23


12/28

The Rule of law as the Rule of Liberal Prindple 65

The common law in contrast does not operate by setting limits throughlanguage to judicial interpretation. Here Dworkin introduces a distinction between "enactmentforce" and "gravitational force," where enactmentforce is about the limits on the political decisions that may be ascribed toan authority and are set by language, whereas gravitational force is a kindof authority that "escapes . . . language." It is explained by the principle offairness, which requires that like cases be treated alike. But likeness here islikeness with respect to principle. The gravitational force of earlier decisions is limited "to the extension ofthe arguments ofprinciple necessary tojustify those decisions."24

Dworkin does not, however, suppose that statutes have only enactmentforce, whereas judicial decisions always have gravitational force. On hisaccount, a judge deciding a hard case must seek to show that his or herinterpretation of the law is best on two dimensions fit and soundness.25The judge must show that the interpretation is among those consistentwith- thatfit- relevantinstitutionalhistoryand thatoftheseinterpretationsit is soundest in that it shows that history in its best moral light. The set ofprinciples that shows all the law of a particular jurisdiction in its best lightis what Dworkin terms a theory of law.

Total consistency or fit is not achievable, and so the judge will have todevelop a theory of "institutional mistakes." Dworkin emphasizes that atheory of mistakes cannot be "impudent" a judge cannot declare as mistaken any piece of institutional history merely because it is incompatiblewith his or her own theoryof law, otherwise the requirementofconsistencywould be "no genuine requirement at all." A judge will often lack authori ty to overrule either statutes or past judicial decisions, and so, even if hisor lier theory deems these to be mistakes, they are "embedded" mistakes.Here the judge should deal with the mistake by limiting its legal force toits enactment force - to what its explicit language requires.26

But even the decision so to limit the force of a past decision must beconstrained by the principle of fairness, since that principlewill say that adecision declaring a past decision a mistake is "prima facie weaker than onethat does not." This gives rise to two maxims a judge must observe:

I f he can show, by arguments of history or by appeal to some sense of thelegal community, that a particular principle, though it once had sufficientappeal to persuade a legislature or court to a legal decision, has now solittle force that it is unlikely to generate any further such decisions, then theargument from fairness that supports that principle is undercut. I f he can


13/28

66 David Dyzenhaus

show by arguments of political morality that such a principle, apart fromits popularity, is unjust, then the argument from fairness that supports thatprinciple is overriddenPNow the secondmaximclearlyallows the judge's own convictions about

justice to decide the issue.28 Thus it immediately attracts the charge thatit permits the judge to decide on the basis of his or her own convictions,which, Dworkin acknowledges, many will regard as "unfair, contrary todemocracy, and offensive to the rule oflaw."29 .

Dworkin points out that even if a judge decides in. hard cases to deferto the community's sense of justice, such deference relies on a convictionabout justice - that this is what the judge's institutional duty requires. So itmight seem that one can characterize the issue as one between the populistdemocrat judge who regards his or her duty in such cases as requiringdeference to community morality, understood as conventional morality,and a judgewho relies on his or her own convictions aboutmorality.

Dworkin argues that this is a mischaracterization. His judge, whom hecalls Hercules because of his "superhuman skill, learning, patience, andacumen,"30 does not "enforce his convictions against the community's."Rather, the judge takes the "record" of the community's "moral traditions"to be that "captured in the whole institutional record that it is his officeto interpret." This record establishes a "constitutional morality," one thatmay be at odds not only with conventional morality - the opinion of theday - but also with other discrete bits of the record. "There is," he says,"no question as to how such a conflictmust be resolved":

Individuals have a right to the consistent enforcement of the principles onwhich their institutions rely. It is this institutional right, as defined by thecommunity's constitutionalmorality, that Hercules must defend against anyinconsistent opinion, however unpopular.31We can now see why the distinction between background and institu

tional rights is so important. In hard cases, judges are decidingwhat legalrights individuals have, and those rights depend on moral principles. Butthe principles are not the moral principles that the judge would enact intothe law, had he or she the power to do so. Rather, they are the principlesalready established by the legal record. Since that record is a legal one established by the political decisions of a particular community over time the community cannot complain that the judge relies illegitimately on hisorher own convictions about justice in interpreting that record, even i f thatinterpretation is both unpopular and declares some part of the record to be


14/28


mistaken. For the judge is interpreting that community's record, the moralprinciples i t regarded as so important that it gave them public expression inits law. In short, there is no objectionfrom democracy, since themoralityonwhich the judge relies is a community's own constitutionalmorality.Nor isthere an objection from the rule oflaw, since a judge in decidingsuchcaseshas to rely on his or her substantive convictions about justice at some point.

Dworkin acknowledges that occasionallya judgewill find thathis orhersense of justice will be in conflict with concepts in institutional morality.Generally, this will not be the case because "he is likely to value mostof the concepts that figure in the justification of the institutions of hisown community," and so he "will be able to put to himself, rather than tosome hypothetical self, questions about the deep morality that gives theconcept value."But ifsuch conflict arises, the judgewill find that his or herinquirywill be more "sociological" in nature; the judge will ask what arethe convictions about morality held by those who enacted that part of therecord rather thanwhat his or her own convictions are.32

Positivists think this acknowledgment is fatal. A distinction is usefulhere, one that falls within D w o ~ k i n ' s category of principles, between substantive and procedural principles. The distinction is notoriously slippery,since procedural principles are often valued for substantive reasons. But itusefully captures the dIfference between, on the one hand, principles aboutthe rights and liberties of the individual and, on the other, the principlesthat likecases shouldbe treated alike and that, in a legal regime oflegislativesupremacy, judges should defer to thewill of the legislature.

Now imagine a legal order where these two procedural principles holdand the legislature is the captive of a minority racial group, which uses itspower both to enact racist principles into the law and to ensure thatmostjudges are either supporters of its policies or at least are likely to adoptthe view that in hard cases their duty lies in deferring to the group's views.Moreover, legislative and judicial reaction to the decisions of the few liberaljudgeswho make it to the bench shows that the legal communityregards aslegitimate this conventionalist mode of interpretation. In this legal order,the substantiveprinciples ofthe law are racist, whereas the procedural principles require their faithful enforcement. Even in cases where some roomfor interpretation is available, it seems clearhow judges are supposed to usethat room.

This profoundly undemocratic legal order is not fanciful - it is prettymuch the legal order ofapartheid South Africa, which was frequently citedas a counterexample to Dworkin's theory of law. For the apartheid legalorder seemed to show that a liberal judge who emulates Hercules will be


15/28

68 David Dyzenhaus

driven by his adherence to the distinction between institutional and background morality to an evermore sociological inquiry. And such an inquiryWill detect racist substantive principles as the animating principles behindboth statute and common law.

33At first, Dworkin appeared to concede this point.He said that the liberal

judge would either have to resign or to make "the difficnlt moral decision"to lie about what the law is.34 As Hart pointed out, that concession raisesinsuperable difficnlties for Dworkin's claim that a judge always has a m9ralreason to apply the soundest theory of the law.35 It concedes that the substantive principles of the law can be morallyodious. Hence, it concedes thepositivist case against equating the rule of lawwith the rule of good law.Dworkin thus offered two further responses. First, he said that it is a

mistake to take wicked legal systems as a crucial test for the connectionhe claims exists between moral and legal rights, since that assumes thatweknow what people's legal rights are in such systems, and we can know thatonly if we have a theory of legal rights. ''"Wicked legal systems should betreated, that is, like hard cases that turn on which conception of law is bestrather than like easy cases whose proper resolution we already know andcan therefore use to test any particular conception for adequacy."36 In thisregard, he suggested that his conception of law would prove adequate in a"wicked legal system" as long as one distinguishes between explanation andjustificationso as to see that an explanation does not "provide a justificationof a series of political decisions if it presents, as justificatory principles,propositions that offend our ideas ofwhat even a badmoral principlemustbe like."And he asserted that he had more faith than he had expressed inearlierwork in the "screening power of the concept ofa moral principle" the power of the justificatory dimension of his test for hard cases to ruleout blatantly discriminatory principles.37

Second, he said it is a mistake to suppose that the source of the moralweight of the judge's duty to apply the soundest theory of the law is foundin the principles that figure in the soundest theory itself. Rather, the sourceof themoralweight is a "generalpolitical situation . . . that the centralpowerof the community has been administered through an articulate constitutional structure the citizens have been encouraged to obey and treat as asource of rights and duties, and that the citizens as a whole have in fact doneso." In short, the moral reason - or "moral kind of reason," as he qualifiesit - is whether the public accepts the legitimacy of the law.38The first response is problematic since it seems to collapse explanationor fit into jnstification.Themore the liberal judge's convictions about background morality are put under stress, the less the test of fit counts in hisor her approach to hard cases until the point where all that counts is his


16/28

The Rule of law as the Rule of lii:leral Principle 69

or her convictions. The distinction between institutional rights and background rights disintegrates when the judge finds the moral content of thelawobnoxious. AndDworkin's suggestion that hemight say ofsuch a situationthatthereareno legal rights because the lawseems so devoidofmoralitydoes nothelp since it raises the sameproblemofequating lawwithgood law.

The second response is problematic because it collapses justificationinto explanation. It appears to concede that the principled basis of the lawwill not ofitselflegitimate the legalorder because the substantive principlesmight be morally obnoxious. It therefore resorts to an external source oflegitimacy - large scale public endorsement of the law. But it is unclearwhy such endorsement will provide judges with a reason to enforce thelaw, unless it is the case that one might expect that where there is suchendorsement, it occurs in part because principles attractive to the majorityare already incorporated into the law. It will follow, of course, that mostjudges will also find the principles of the law morally attractive, and sotheywill think there is a moral reason to apply them. But this provides asociological explanation ofwhy judges are likely to think there is a moralreason to apply the law of the land, not a normative argument aboutwhythere is such a reason.

In sum, the wicked legal system seems an effective counterexample toa liberal model of the rule of law. Dworkin's :first response to the wickedlegal system pushes the bridge the model builds from law to morality toofar because it equates law and morality. His second response has the bridgestop short of reaching the moral shore, at the same time as it seeks to t intthat situation with morality. It is perhaps awareness of this dilemma thatled Dworkin to remark that wicked legal systems "are not very importanthard cases, from the practical pointofview, because the judgmentswemakeabout foreign wicked legal systems are rarelyhinged to decisions wehave totake."39 But this remark, made in themidstofarticulatinghis first response,tied his argument to the fatal concession of his second response - that histheory of the rule of law is contingent on judges' finding themselves in alegal order in which the substantive principles arenot badlyout oflinewithliberal morality. And, as we will now see, an analogous dilemma bedevilshis rights model of the rule oflaw, even in democratic legal orderswhere itshould bemost at home.

D. Is Administrative Law a Mistake?We saw that Dworkin has a theory of "institutional mistakes," whichexplains when a judge is entitled either to expunge some bit of the public record oflaw or to limit its force. There is another sense of institutional


17/28

70 David Dyzenhaus

mistake in the historyof legal theory, one that pertains not to the substanceofparticularbits of the record but towhole institutions. Bentham, for example, thought that to give judges the kind ofauthorityover the interpretationof the law that licenses judicial antipositivism is to make a grave politicalmistake. In his view, the institution of the common law is a mistake thatmust be confined or, evenbetter, eradicated. Therefore, a constitutionalismthat explicitlygives judges the authority to strike down statutes is that samemistakewrit large.

Conversely, A v: Dicey's model of the rule of law is designed to findthat the law made by administrative agencies is a mistake because i t is intensionwith the common law. This is in part for the reason that his modelof the rule oflaw cannot accommodate administrative agencieswith hybridfunctions ofmaking and interpreting the law, since themodel requires thatlegislatures haveamonopolyonmaking the lawandjudgeshaveamonopolyon interpreting the law.40 But it is also - andmore fundamentally - becauseDicey is opposed to the political program of regulation and redistributionof economic and social resources, which prompted the development of theadministrative state.

ForDicey,whatwas fundamentallywrongwith the administrative statewas that it flouted the principles of a morality of liberty and protection ofprivate property, a kindofconstitutionalmorality inDworkin's sense,whichhad been embedded in the common law. Since these principles formedthe backdrop for judicial interpretation of statutes, statutes that delegatedboth legislative and adjudicativepower to agencies removed them from thecontrol of a constitutional morality whose guardian is the jndiciary. Andthis same theme is to be found in LordHewart's The New Despotism 41 andin F. A Hayek's The Road to Serfdo11L.42

Dworkin's account of the substance of the liberal principles that judgesshould protect is different from Dicey's, Hewart's, or Hayek's, all ofwhomare libertarians concerned to maximize the space of negative liberty forindividuals by limitingstate intervention to the greatest possible extent. Asan egalitarianliberal, Dworkinhas argued that equality, not negative liberty,is the foundational value of liberalism. And he gives the state a large role inredistribution, as long as redistributive decisions do not violate a principleof state neutrality expressed inwhat he takes to be the fundamental liberalprinciple of equal concern and respect.

But though there is this political difference between Dworkin and libertarians about the content ofwhat is protected by the rule of law, there islittle or no difference from the perspective oflegal theory. The libertariansandDworkin agree that the rule of law existswhen judgesprotectsome core


18/28

rThe RUle of Law as the Rule of Liberal Principle 71

of liberal principles from majoritarian decision making. And this stance of"judicial supremacism" is made up of two components that together createenormous difficulties for judicial review of administrative action, whateverthe details of the liberal ideology behind it.The first component is institutional - the place of judges in "Law'sEmpire." No less than Dicey, Dworkin understands legislatures as havinga monopoly on making law and judges a monopoly on its interpretation.There is no room in his account for administrative agencies that have anauthority to make or interpret the law in the sense that such administrativedecisions are ones to which courts have reason to defer. Atmost, Dworkincan concede to administrative agencies the authority to make decisionsabout the policy implications of their constitutive statutes utilitarian calculations about what decision will best advance the policy or decisionsabout what the effects of different arrangements of natural justice mightbe. But theyhave no authority over legal principles- the exclusive provinceof the judiciary.43

The second component is substantive - the assumption that the essenceof law is some particular substance, a coherent conception of the commongood. It does notmatter so muchwhat the precise content of thatvision ismore or less libertarian, more or less egalitarian as that there is such anassumption. For it is such an assumption that requires that judges have theinstitutional place just sketched.44

Together these two components create difficulties familiar to administrative lawyers in common law legal orders. On the one hand, Dworkin'sunderstanding of the imperialism of principles over which judges presideinvites the extensionofprinciples derived from one area oflawinto another,and thus, for example, invites judges to attempt to govern the public lawregimes ofadministrative law in the lightof their understandings ofprivatelaw.45

On the other hand, the distinction between principle and policy inDworkin's account of adjudication suggests thatwhen one is on the policyside of the divide, no crucial issues of legal principle arise.46 The administration either gets squeezed into a procrustean common law bed or left toits own devices because it is somehownot for themost part involved in law,properly so called.

In the last sectionDworkin's dilemmawas betweenan implausible equation of the principled content of the law with liberal morality and a concession that the principled content of the law is only contingently sound.Here the dilemma is between a judicial supremacism hostile to the administrative state and a concession that judges do not have an interpretative


19/28

72 David Dyzenhausmonopoly. But the root cause of both dilemmas is the same - an equationof the principled basis of the lawwith a particular set ofprinciples ofliberalmorality. And that equation makes Dworkin's position vulnerable to thevarious objections listed earlier: of ascribing legitimacy to unjust regimesmerely because they govern through law, of supporting usurpatory legitimacy claims made by judges, and ofa parochial dependence on a particularkind of legal order.

CULTURES OF LAW

I remarked that few philosophers of law focus on the issue of the ntle oflaw, because most regard that issue as resolved by :first determining whatthe law is that rules. The main exception is Lon L. Fuller, for whom therule oflaw involves principles oflegality that, taken together, amount to an"inner morality" of law.

Fuller argued that eight principles of legality inhere in the idea of therule of law: generality, promulgation, nonretroactivity, clarity, noncontradiction, possibility of compliance, constancy through time, and, the onehe took to be the most complex, congruence between official action anddeclared rule.47 These principles are more procedural in nature than substantive in that they are supposed to pinpoint what is involved in the production ofvalid law.

Asystem that fails completely tomeet oneprinciple, or fails substantiallyto meet several, would not, in Fuller's view, be a legal system. It would notqualify as governmentunder law- as government subject to the rule oHaw.Further, a tyrant who wanted to govern through the medium oflawwouldhave to comply with the eight principles, and this would preclude rule byarbitrary decree and secret terror, which, Fuller said, is the most effectivemedium for tyranny.48

Positivists argued that Fuller's principles oflegalitymerely enhance theability of law to guide behavior,49 and Dworkin joined vigorously in thispositivistcritique.50 He suggested thatFullerhadmissedhis positivist targetaltogether, since Fuller's arguments were about a purported innermoralityoflaw, whereas the separation thesis is about a connection between lawandsubstantive moral standards external to law.51

Dworkin also said that if tyrants preferred not to rule through law,this would be because they feared publicity and not because there was anyconstraining inner morality of law.52 Thus, he challenged Fuller's claimthat laws enacted to achieve evil ends are necessarily vague and lacking in


20/28

The RUle of Law as the Rule of Liberal P r i n c i ~ l e 73

the quality of legality, saying that a "perfectly evil statute can be draftedwith exquisite precision." The only substance in Fuller's position, Dworkinthought. was that i fwe have to interpret a lawwhose fundamental purposeis obnoxious to our sense ofmorality, we might have difficulty in understanding the law becausewe would "gain no help from our sense offairnessinmaking discriminations under it."n

Finally,Dworkinclaimed that the projectofestablishingan innermorality of law is fundamentally mistaken because it fails to see an important difference betweenlawandmorals- thata "moral principle . . . cannotbe established by deliberate act, as some sorts oflaw can.,,54 His point here is thatwhereas there might be criterial standards ofmorality analogous to Fuller'scriteria for legality for example, the rule against self-contradictionsuch moral criterial standards do not tell us whether an act is moral orimmoral. Only the standards ofsubstantive morality can tell us that; hence,the criterial standards ofmorality do not identify the standards ofmoralityin theway that the criterial standards of lawmight.

In making this point, Dworkin relied explicitly and with wholeheartedapprovalonHart's account in The ConceptofLaw of the relationship betweenmoral and legal obligation, specifically on the section where Hart maintained that certain important differences between law and morality arisefrom the fact thatmorality, unlike law, is immune from deliberate change.Hart said that"the idea ofa moral legislature with competence to make andchange morals is repugnant to the whole notion ofmorality."S5

But this idea is not obviously repugnant. Something like the idea of a"morallegislature"with "competence to make and changemorals" animatesthe democratic tradition, includingBentham's utilitarianism. By supposingotherwise,Hart andDworkin affinn a crude dichotomybetween liberalismand democracy.

On the one hand, we have liberals determined to preserve the substantive standards of morality from statutory encroachments by democraticlegislatures. They thus hold up the judiciary as the legal guardian of thepolitical constitution. The ideal system for many liberals, especially in theUnited States ofAmerica, entrenches liberal political morality in a writtenconstitution - one that gives to judges the authority to strike down statutesas invalid when these cannot be interpreted in a fashion consistentwith thatmorality. On the other hand, we the populist democrats referred toearlier, who require law to be nothing more than the expression ofwhatthe people in fact want, unconstrained insofar as this is possible by formallegal requirements and completely unconstrained by judicially interpretedprinciples.


21/28

74 David Dyzenhaus

Curiously, what these liberals and democrats share is an instrumental,basically positivist, account oflaw. Law is just the instrument ofsubstantive .values established by political morality, and disagreement is about onlywhether it is liberal political philosophyor "the people" who should decideon the content ofmorality. And i f it seems odd to suggest that Dworkinturns out to be a positivist in this very limited sense, one should recall thata constant criticism of his theory has been that it boils down to a versionofAmerican constitutionalism - to a political theory with a claim to h a ~ eestablished itself in a particular legal order.

It is also worth recalling that Fuller too argued against the positivistaccount of adjudication and, especially in the 1960s, advanced a theoryof adjudication that was in many respects quite similar to Dworkin's.56However, he never tied his theory of adjudication to a particular versionof liberal political morality, nor did he rest his challenge to the separationthesis on such a theory. He stuck to his idea that a sincere commitment togovernmentwithin the constraints oflegality involvesmoral constraints ongovernment and that it is in attending to the values that such constraintsserve that judges, like other legal officials, will express their fidelity to law.In other words, Fuller's theory of adjudication is a subtheory within hismain theory, which is a theory of law as legal order or legality. And it isthe moral order put into place by the legal order to which judges must befaithful rather than to any particular political ideology.

Fuller's antipositivism thus agrees with one of positivism's principalobjections to Dworkin's position - that Dworkin goes wrong in trying tobuild a theory of law out of a theory of adjudication. The error seemsdemonstrable in Dworkin's case because he ties his account of adjudicationto a particular version of liberalism and so commits himself either to theodd claim that every legal order must instantiate that particular politicalideology or to the counterintuitive claim that judges are morally as well aslegally bound to apply morallyunsound law.

Further, Dworkin - with the positivists - failed to appreciate the complexity ofFuller's principleofpublicity. Publicityfor Fuller is not onlyaboutstaring one's aims in public. Rather, publicity involves a commitment to aprocess of reasoned justification by legal authority to those subject to it.A public law is one that is the result of such a process.57 And in generalall of Fuller's principles are best understood as instantiations of the twomain democratic principles - participation and accountability - principlesthat can be realized in very different ways within legal institutions with nothought that judges have a monopoly on their guardianship.


22/28

The Rule of Law as the Rule of Liberal Prindple 75

Inmy view, the embryonic forms ofdemocratic accountabilityare foundmerely in an insistence on positive law as the form that political policymust adopt. Not only is positive law highly visible, but it also establishesa standard to which those with power can be called to account. In otherwords, the existence of positive law is linked to its justification - visibilityplus accountability.58 As the idea ofwhat ensures accountability changes,so will ideas of what legal institutions are required and thus ofwhat thefundamental forms of law are. But such ideas will always be linked to themoral principles that legal forms are meant to serve.

I have argued elsewhere thathow theyare linked can best beunderstoodby distinguishing between three different conceptions of legal culture: aDworlcinian liberal culture of neutrality, which requires judges to upholdliberal standards on judicial review; a positivistic culture of reflection, inwhich lawmerely mirrors or reflects judgmentsmade in the political arenaand judges are required to defer submissively to the clearly expressed willof the legislature; and the Fullerian culture of justification, in which law isjustified when it is adequately supported by reasons, and thus the judicialduty is to promote a culture of reason giving.59 It is that idea that providesthe principled basis of an accommodation of the change in the form of lawbrought about by the regulatory state, one that neither consigns administrative decision making to the positivist black hole of discretion, nor setsup judges as supreme arbiters of the law.

CONCLUSION

It is a mark of Dworkin's great influence that most positivists becamealmost entirely focused on adjudication on putting together a conceptualresponse to Dworkin's demonstration of the role ofprinciples in adjudication. Recently, however, some positivists, against the trend of conceptualism, have returned to the task of trying to build what we saw Dworkin calla general theory oflaw.60 They sense that the place of the democratic legislature in legal order has almost completely disappeared from legal theory.And they seek to marry a positivistic account of legal institutions with apolitical account oflegitimacy, thus beginning a response to the challengeDworkin articulatedmore than twentyyears ago in Taking RightsSeriously.61

The return ofpositivism to its substantive political rootswill, I suspect,highlight even more starkly the problems faced by Dworkin's account ofthe rule of law. His most significant contribution to our understanding of


23/28

76 David Dyzenhaus

the rule of law could then come to be seen as his prompting of a revivalof legal positivism. But that result would not be ironical i f one understoodDworkin's project in line with his "Introduction" to 7akingRightsSeriously.That is, Dworkin's primary aimwas not so much to provoke acceptance ofhis theory of adjudication as to show that the issues raised by that theorycompelled legal positivists and others to resituate the great debates oflegaltheorywithin political philosophy.

It would, or so I have suggested, be a mistake to leave Fuller out ofthe story. A revival of the Fuller-Dworkin debate mightwell hold the cluesto a stronger, antipositivist theory of the rule of law. In particular,onewould want to dwell on those moments in Dworkin's theory of law that Ihave neglected in this chapter because,whenpush comes to shove, Dworkinhimselfhas chosenanother path.Here I refer to the idea that judicialfidelityto law is not primarily about the judicial role in promoting articulate consistency or integrity with principles of some substantive vision of politicalmorality. Rather, judges, together with the other institutions oflegal order,namely, the legislature and government, have a role in promotingintegritywith the principles oflegality - the more proceduralmorality of the rule oflaw.62

ACKNOWLEDGMENTS

I thankAlan Brudner, Sujit Choudhry, Evan Fox-Decent, Arthur Ripstein,and Hamish Stewart for comments on drafts of this chapter. My greatestdebt in regard to its themes goes back to discussions between 1984 and1988with a highly imaginative and conscientious thesis supervisor, RonaldDworkin. With time I have come to understand better the luck involvedin working with a supervisor whose only concern was to help his studentsmake their arguments into the best they could be.

Notes1. RonaldDworkin, "Political}udges and the Rule ofLaw," inAMatter ofPrinciple(Cambridge,Mk Harvard University Press, 1985), 11-12 (his emphasis).2. And, as one commentator has pointed out, there is hardly any mention of thephrase "the rule of law" within Dworkin's major work on legal theory, Law's

Empire (London: Fontana Press, 1986). See Paul Craig, "Formal and SubstantiveConceptions of the Rule ofLaw: An Analytical Analysis," Public Law (Autumn1997): 467-87. Inpart, thisisa productofthe highlyparochialnatureofAmericanpublic law discourse, which assumes that debate about the rule ofIaw is a debateaboutU.S. style constitutionalism. (For an exception, see RichardH. Fallon,}r.,


24/28

The RuLe of Law as the RuLe of liberaL PrincipLe 77

"'The Rule ofLaw' as a Concept in Constitutional Discourse," Columbia LawReview 97, no. 1 Uanuary 1997]: 1-56.) It is thus unsurprising that Dworkin'sonlysustained treatmentof the rule oflawoccurs on the occasion ofa lecture tothe British Academy and that the bestwork on his legal theory and the rule oflaw is by a British administrative lawyer, Trevor Allan; see, for example, Allan's"The Rule ofLaw as the Rule ofReason: Consent and Constitutionalism," LawQuarterly Review 115 (1999): 221-44, and, more recently, ConstitutionalJustice:A Liberal Them) of the Rule ofLaw (Oxford: Oxford University Press, 2001).More recently, Dworkin has taken up the topic of the rule oflaw in response toJeremyWaldron, "The Rule ofLaw as a Theatre ofDebate," in Dworkin andHis Critics, ed.Justine Burley (Oxford: Blackwell, 2004), 387-8. In otherworkDworkin has suggested that the way forward in legal philosophy is to see thatthe concept of law is answerable to an ideal of legality. See Dworkin, "Hart'sPostscript and the Character of Political Philosophy," OxfordJournal ofLegalStudies 24, no. 1 (Spring 2004): 23-5. Note that there is renewed interest inthe topic of the rule of law after 9/11. Here Dworkin has played an importantpublic intellectual role in defending civilliberries in the pages of the New YorkReview ofBooks. I discuss the implications of 9/11 for debate about the rule oflaw in The Constitution ofLaw: Legality in a Time ofEmergency (Cambridge, UK:Cambridge UniversityPress, 2006).

3. See especially the posthumously published "Postscript" to H. L. A. Hart's TheConcept ofLaw, 2nd ed. (Oxford: Clarendon Press, 1994), 239-44.

4. I ignore in this essay the "incorporationist"or "soft" versions oflegal positivism.They not only share many of the problems of Dworkin's account of the ruleof law, but they also seem to subvert the positivist project, as I try to explainin "Positivism's StagnantResearch Programme," OxfordJournalofLegalStudies20, no. 4 (2000): 703-22 and in "The Genealogy ofLegal Positivism," OxfordJournal ofLegal Studies 24, no. 1 (Spring 2004): 39--67. Some contemporarypositivists deny that positivism subscribes to the separation thesis; see JohnGardner, "The Legality ofLaw," Ratio Juris 17, no. 2 aune 2004): 168-81.

5. At least it is not a conceptual concern. It is a political concern, as I mentionlater, because of issues about political accountability - issues that conceptuallydriven positivists consign to political theory.6. See Dworkin, 1itkingRights Seriously (London: Duckworth, 1978),vi i (hereafterTRS). My principal focus in what follows is the "Introduction" and Chapters 2,3, and 4 ofTRS. In my view, every important element ofDworkin's account ofthe rule of law is to be found in these chapters, which explains why they hadsuch an immediate and important influence.

7. Accordingto Dworkin, the two streams sometimes converge in the conservativeconstitutional theory advanced primarily by Robert Bork; see Dworkin, "Law'sAmbitions for Itself," Vil-ginia Law Review 71, no. 2 (March 1985): 173-87, andDworkin, Fl"eed011t's Law (Cambridge, MA: Harvard University Press, 1996),especially the "Introduction" and the essays dealing with Bork in Section III.On pages 33 and 34of this book, Dworkin suggests that judgesmaynot providethe only official site for the moral elaboration of the requirements of the ruleoflaw.


25/28

78 David Dyzenhaus

8. Dworkin, TRS, xii.9. Dworkin, TRS, xii.10. Dworkin, TRS, xii.11. The historic compromise between positivism and the common law was struckby John Austin, precisely because he wanted to find a role for judges as -abulwark against democracy; see Austin, Lectllres on }misprlldence, vol. I, 5thed. (London: John Murray, 1885),218; and Austin, Lect1m!s on Jmisp17ldence,vol. 2, 5th ed. (London: John Murray, 1885), 532-3, 641-7; both may profitably be read together with his pamphlet, A Plea for the Constitution (London,1859). The incorporationist or soft positivism referred to in note 4 is perhapsbest understood as providing the theoretical spin on how to accommodate thiscompromise.12. ThusDworkin's argument in "Equality, Democracy, and Constitntion:WeThe

People in Court,"AlbertaLawReview 28, no. 2 (1990): 324-46,while presentinga sophisticated argument for a "nonstatistical" account ofdemocracy, does notdeal at allwith the question ofwhy the law already embodies that account.Thesame judgment holds for the essays on political equality and democracy collected in Ronald Dworkin, Sove1'eign Virtue: The Them] andPmctice ofEf]1Iality(Cambridge, MA: Harvard University Press, 2000). It is worth noting in thisconnection Richard Pildes's argument that Dworkin's idea of rights as trumpsconstrainsanyattempt tonnderstand the fruitful role judges can play in enhancing the democratic character ofpolitical and legal institutions; seePildes, "WhyRightsAreNotTrumps: SocialMeanings, ExpressiveHarms, andConstitutionalism," J01l171al OfLegalStltdies 27, no. 2 aune 1998): 725-63. JeremyWaldronhas suggested that Pildes misunderstands Dworkin's understanding of rights,sinceDworkin in fact opts for Pildes'spreferred understandingofrights as "constraints on the kinds of reasons that government may legitimately act upon"rather than as trumps or "simple protections for certain individual interestsagainst the demands of the common good." Waldron, "Pildes on Dworkin'sTheory of Rights," J01l171al OfLegal Studies 29, no. 1 aanuary 2000): 301. Butthis responsemakes little difference to the substanceofPildes's argument. EvenifWaldron is right, Dworkin also supposes that the principal constrainton reasons is a requirement that government must treat each individual with equalconcern and respect.

13. Dworkin, TRS, 82.14. Dworkin, TRS, 84-7.15. Dworkin, TRS,86-7.16. Dworkin. TRS, 87-8.17. Dworkin, TRS, 93.18. Dworkin, TRS, 101.19. Dworkin, TRS, 101.20. Dworkin, TRS, 105.21. Dworkin, TRS, 21.22. Dworkin, TRS, 108.23. Dworkin, TRS, 109-10.


26/28


24. Dworkin, IRS, 111-13, at 113.25. The terms "fit" and "soundness"are drawn from later work; see Dworkin, Law'sEmpire.26. Dworkin, TRS, 121.27. Dworkin, TRS,122-3.28. Dworkin puts things somewhat more ambiguously in another earlier formulation of this maxim, saying that a decision could be wrong because "unfair,within the community's own conception of fairness" (ibid). For an illuminatingdiscussion of this ambiguity, seeAnthonyJ. Sebok, LegalPositivism in AmericanJttrisp17ldence (Cambridge, UK: Cambridge UniversityPress, 1998): 238ff.29. Dworkin, TRS, 123.30. Dworkin, IRS, 105.31. Dworkin, TRS, 126.32. Dworkin, TRS, 127-8.33. See Raymond Waclrs, "Judges and Injustice," South African Law]ounlal101,no. 2 (May 1984): 266-85. I deal with these issues in Hard Cases in WickedLegal Systems: South African Lnw in the Perspective ofLegal Philosophy (Oxford:Clarendon Press, 1991) andJtldging the Jtldges, Jtldging Ourselves: 7i-uth, Reconciliation and the ApartheidLegal Orde7' (Oxford: Hart Publishing, 1997).34. Dworkin, IRS,326-7.35. SeeHart,&says in Jtl77sprudence and Philosophy (Oxford: ClarendonPress, 1983),

9, and&says on Bentham: Jtlrisprudence and Political Tbeol'Y (Oxford: ClarendonPress, 1982), 150-3.36. Ronald Dworkin, "AReply," in Ronald Dworkin &Contemporll1Y Jtlrisprtldence,ed. Marshall Cohen (London: Duckworth, 1984), 260.37. Dworkin, "AReply," 299 4n.38. Dworkin, "AReply," 258.39. Dworkin, "AReply," 260.40. For the ideas oflegislativeand interpretativemonopolyinDicey, seePaulCraig,Public 'Law and Democracy in tbe United Kingdom and the United States ofA11l1!11ca(Oxford: Clarendon Press, 1990), 12-55.41. Lord Hewart, The New Despotisnt (London: ErnestBenn Limited, 1929).42. E A., Hayek, The Rond to Serfdom, 50th Anniversary ed. (Chicago: University ofChicago Press, 1994).43. See Dworkin, "Principle, Policy, Procedure," inA Mlltte7' ofPrinciple, 99-101.44. The problem here is akin to the one Roberto Unger detects when he describesthe "two dirty little secrets of contemporary jurisprudence." The first is a"rightwing Hegelian view of social and legal history," one which emphasizes

the "cunning ofhistory in developing rational order . . . out of the unpromisingstuffof historical conflict and compromise." The second is a "discomfortwithdemocracy" thatleads to themarginalizationofpolitics in legal theory, includingtheworkofdemocratic legislatures, and a focus onwhat "higher judges" do. Thetwosecretswork togethersince the "restraints on democracy" thathave becomethe focus oflegal theory "open the space in which the self-fulfilling prophecies


27/28

80 David Oyzenhau5

of rightwing Hegelianism c:an come to pass." See Roberto M:angabieraUnger, What Should Legal Analysis Become? (London: Verso, 1996), 72-4, :and.also JeremyWaldron, Law and Disagreement (Oxford: Clarendon Press, 1999),especiallyChapter 2.For a sustained :andverypowerfulargument that:anunderst:anding of the rule oflawrequires a Dworkini:an account ofthe commongood,see All:an, C01zstitlttional Jzzstice: A Liberal Theory ofthe Rule ofLaw. .

45. The imperial project is subject to the principle of local priority, which allows"departments of law" to deviate somewhat from pure integrity; see Dworkin,Law sEmpil'e, 250-4. But all such deviations are a matterofregret, and given theplace of judges in the imperial project, it is highly unlikely that a judge shoUldpause to take seriously :an administrative agency's sense that the principles of,say, the common law of contract should not govern the interpretation of acollective agreement.

46. See Dworkin's discussion of the distinction between moral harm :and bareharm, "Principle, Policy, Procedure," 84ff.,:and see Genevra Richardson, "TheLegal RegulationofProcess," inAdminist:rativeLaw and GovenlmentAction, eds.Genevra Richardson :and Hazel Genu (Oxford: Clarendon Press, 1994), 105.Richardson states, "This is apowerful argumentforprocedural rights in relationto substantive rights. The corollary is, however, that there are no proceduralrights in relation to bare interests. The harm which may result from interference with bare interests is bare harm; there is no moral harm since nothinghas been lost to which there is a right, :and it is only moral harm that attractsprocedural protection. It seems that for Dworkin, once outside the sphere ofsubstantive rights, procedures are essentially matters of policy with :any claimto specific procedures being so weak as to be negligible" (ibid., 112-13).

47. These are set out in detail in Lon L. Fuller, The Morality ofLaw, rev. ed. (NewHaven: Yale University Press, 1969).

48. Fuller, The Morality ofLaw, especiallyChapter 2.49. SeeHart, Essays in Jzwisprudence and Philosophy, 350-2. For a more recent state

ment, see]oseph Raz, "The Rule ofLaw :and its VIrtue," in his The Authority ofLaw: Essays on Law and Morality (Oxford: Clarendon Press, 1983),210-29.50. Dworkin, "Philosophy, Morality, :and Law: Observations Prompted by Pro-

fessor Fuller's Novel Claim," University ofPemzsylvania Law Review 113, no. 5(March 1965): 668-90.51. Dworkin, "Philosophy, Morality, :and Law," 671.52. Dworkin, "Philosophy, Morality, and Law," 672.53. Dworkin, "Philosophy,Morality,:and Law," 672.54. Dworkin, "Philosophy,Morality, :and Law," 684.55. Dworkin, "Philosophy, Morality, :and Law," 694 3On; and seeHart, The Concept

ofLaw, 171-3.56. See especially Fuller, Anat01JlY ofLaw (New York: FrederickA. Praeger, 1968).57. Fuller identified his own starting point as a faith in the moral resources of

law; in particular, he suggested that when people "are compelled to explain:and justify their decisions, the effect will generally be to pull those decisions


28/28

The Rule of Law as the RuLe of Liberal Principle 81towards goodness, bywhatever standards ofultimate goodness there are." SeeFuller, "Positivism and Fidelity to Law: A Reply to Professor Hart," Harvm'dLaw Review 71, no. 4 (February 1958): 632.

58. Such links creep eveninto the account oflaw of that allegedlyuncompromisingadvocate of political absolutism, Thomas Hobbes. For further discussion seeDavidDyzenhaus, "Hobbes and the LegitimacyofLaw," Law and Philosophy20,no. 5 (September 2001): 461-98.

59. See David Dyzenhaus, "Form and Substance in the Rule ofLaw: ADemocraticJustification for Judicial Review," in Judicial Review & the C01lstitltti01l, ed. C.Forsyth (Oxford: Hart Publishing, 2000), 141-72. The term "culture of justification" was coined by the late South African administrative lawyer, EtienneMureinik.

60. SeeJeremyWaldron, Law and Disagreement as well as The Dignit:y o/Legislation(Cambridge, UK: Cambridge University Press, 1999); Jeffrey Goldsworthy,The Sove1'eignt:y ofPadiament: History and Philosophy (Oxford: Clarendon Press,1999); andTomD. Campbell, The Legal Theory ofEthicalPositivism (Aldershot:Dartmouth, 1996).

61. The nature of this challenge is best articulated by Stephen Perry's work onmethodology; see Perry, "Hart's Methodological Positivism," Legal Theo1"'Y 4(1998): 427-67.62. See, for example, Dworkin's discussion of integrity, his more recent label for

articulate consistency, in Law's Empi1-e. At 176--7, integrity seems to requirejudicial fidelity to the law rather than to a substantive conception of justice.But at 190ff. - Dworkin's exploration of the "puzzle of legitimacy" - he seemsundecided on this issue. And, i f we read this exploration in the light of hisdiscussion of the wicked legal system problem at 101-13, a similar indecisionreprodnces the problems in this regard, which I outlined earlier. In both cases,Dworkin seems to move on by presuming that ultimately the problems areresolved by political rather than legal theory.Seen in this light, the claim thatDworkin's theory ofadjudication turns out tobe parochial need not be considered disrespectful Although I cannotmake thisargumenthere, I suspect thatall theories ofadjudication are parochial or local,but that their fruitfulness as theories will be judged not only by the way theyhelp US understand what happens at the local level, but also by how they assistdebates about more general theories of law_ And theywill so assist by castingother theories of adjudication and other ways of designing legal order intosharper relief, thus focusing our attention on the most fundamental questionsabout the nature oflaw_ But for there to be this-reciprocal relationship betweenthe local and the more general, there must be some common terms to thedebate about the question "What is law?" Those common terms are set by anagreement that the question is to be answered politically, not conceptually.One mightnaturally raise here a challenge that, ifsuccessful, undermines myconclusion - namely, that the constraints I identify do not amount to amorality.

But that challenge assumes a particular conception ofmorality, for example, aparticular liberal list of individual rights and liberties.

DAVID DYZENHAUS - Rule Law as Rule Liberal Principle

Documents

Transcript of DAVID DYZENHAUS - Rule Law as Rule Liberal Principle