0521875110 cambridge university press the nature of customary law legal historical and philosophical perspectives jun 2007

rationalisation of a set of customary practices having no special status orclaim to allegiance.This sceptical account of practical reason is reflected in many of thecontributions to this

This page intentionally left blank

T H E N A T U R E O F C U S T O M A R Y L A W

Some legal rules are not laid down by a legislator but grow instead from informal social practices In contract law, for example, the customs of merchants are used by courts to interpret the provisions of business contracts; in tort law, customs of best practice are used by courts to define professional responsibility Nowhere are customary rules of law more prominent than in international law The customs defining the obligations of each State to other States and, to some extent, to its own citizens, are often treated as legally binding However, unlike natural law and positive law, customary law has received very little scholarly analysis.

To remedy this neglect, a distinguished group of philosophers, historians and lawyers has been assembled to assess the nature and significance of customary law The book offers fresh new insights on this neglected and misunderstood form of law.

A M A N D A P E R R E A U - S A U S S I N E is a University Lecturer in Law at the University of Cambridge and a Fellow of Newnham College.

J A M E S B E R N A R D M U R P H Y is Professor of Government at Dartmouth College, Hanover, USA.

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

First published in print format

ISBN-13 978-0-521-87511-0

ISBN-13 978-0-511-27422-0

© Cambridge University Press 2007

2007

Information on this title: www.cambridge.org/9780521875110

This publication is in copyright Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.

ISBN-10 0-511-27422-X

ISBN-10 0-521-87511-0

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org

hardback

eBook (EBL) eBook (EBL) hardback

List of contributors page vii

Table of cases viii

The character of customary law: an introduction 1

P A R T II Custom and law: custom, common law andcustomary international law 149

7 Custom in medieval law 151

11 Custom, common law reasoning and the law of nations

in the nineteenth century 256

Christoph Kletzer, University Lecturer in Jurisprudence, University ofCambridge.

Randall Lesaffer, Professor of Legal History, Department of Jurisprudenceand Legal History, Tilburg University

Michael Lobban, Professor of Legal History, Queen Mary College ofLaw, University of London

James Bernard Murphy, Professor of Government, Dartmouth College,Hanover, New Hampshire

Amanda Perreau-Saussine, University Lecturer, Faculty of Law, and Fellow,Newnham College, University of Cambridge

Jean Porter, John A O’Brien Professor of Theology, University ofNotre Dame

Gerald Postema, Cary C Boshamer Professor of Philosophy andProfessor of Law, University of North Carolina, Chapel Hill

Frederick Schauer, Frank Stanton Professor of the First Amendment,John F Kennedy School of Government, Harvard University

John Tasioulas, CUF Lecturer in Philosophy, University of Oxford, andFellow and Tutor in Philosophy, Corpus Christi College, Oxford.Brian Tierney, Bowmar Professor of Humanistic Studies Emeritus,Cornell University

vii

TABLE OF CASES

Abercromby v Fermoy Town

Commissioners, 259

Attorney General v Mathias, 261

Attorney-General for Canada v.

Attorney General for Ontario, 271

Atwood v Seller, 263 , 264

Bastard v Smith, 259

Bate’s Case, 221

Bebb v Law Society, 258

Bechuanaland Exploration Co v.

London Trading Bank Ltd, 262

Dolder v Bank of England, 269

Duff Development Co Ltd v.

Government of Kelantan, 270

Duke of Brunswick v King of Hanover,

266 , 267

East India Company v Campbell, 272

Edelstein v Schuler & Co., 262 , 263

Edie and Laird v East India Company, 262

Jenkins v Harvey, 261

Johnson v Clark, 260

Jones v Garcia del Rio, 269

viii

King of Spain v Hullett, 274

King of the Two Sicilies v Willcox, 274

Maclaine Watson & Co v Department

of Trade and Industry, 271

Marquis of Salisbury v Gladstone, 261

Noble and Another v Kennoway, 264

North Sea Continental Shelf Case,

279 , 323

Novello v Toogood, 266

Nuclear Weapons Case, 333 –

Peru v Dreyfuss Brothers & Co., 275

Picker v London and County Banking

Rumball v Metropolitan Bank, 263

Secretary of State for India v.

Kamachee Boye Sahaba, 268

West Rand Central Gold Mining

Co Ltd v The King, 275 – , 277

The character of customary law: an introduction

A M A N D A P E R R E A U-S A U S S I N E A N D J A M E S B E R N A R D M U R P H Y

A book on customary law, many modern lawyers might say, can have norelevance for them And neither, many modern thinkers would echo,could it be of much interest On many influential modern accounts,reliance on customary practices is a mark of inadequacy: acceptance ofcustoms should be minimal and provisional since an unreflective attach-ment to customary ways of thinking is inimical both to practical thoughtand to political harmony Modern societies and their legal systemsdepend not on enslavement to customary habits and laws but on rea-soned principles and doctrines; customary laws grow up only wherelegislators have done a particularly poor job, leaving a need for elaboratestatutory construction and legislative gap-filling The more coherentand consistent a legal system, the less the need for such customaryrules and practices: an interest in customary law reflects at worst whatJeremy Bentham called the ‘sinister’ interests of self-interested reaction-aries, and at best the eccentric tastes of scholars, antiquarians and thosepurporting to be international lawyers who work in what, on suchaccounts, is really a lawless international world

This brief chapter introduces the diverse views of customary lawoffered in this collection of essays, showing how, despite this diversity,the thirteen contributors are united in arguing that such rejections of therelevance of customary law are wrong

Is custom all we have?

Some jurists and philosophers argue that customary practices are all wehave to guide us in aiming to solve practical questions: moral principles,written laws, legal doctrines and philosophical writing are all articula-tions of pre-existing customs Such accounts are deeply sceptical ofarguments in the name of reason, arguing that those who claim apriority for rational principles said to be manifest within a set of con-flicting customary practices are really claiming priority for their ownpreferred doctrines, doctrines which are themselves nothing but a

rationalisation of a set of customary practices having no special status orclaim to allegiance.

This sceptical account of practical reason is reflected in many of thecontributions to this book by legal historians As historians they areconcerned to avoid allowing contemporary concerns to drive their study

of earlier ideas and practices: instead they seek first to understand ‘thespecificity of a past situation’, leaving readers to ask whether and how far

‘the very specificity’ of that earlier situation gave rise to problemsanalogous to those arising in the contingencies of our own age.1ThusDavid Ibbetson frames his comparative study of customary elements inthe medieval laws of continental Europe and of England as a study of

‘the uses of the idea of custom’: his aim is to trace the different senses ofcustom in medieval law while prescinding from comment on the rela-tionship between those different usages.2 Such writers tend to treatdoctrine not as leading changes in customary practice but as followingand articulating the relevant changes in practice Thus, for example,Randall Lesaffer argues that more humane customary practices and rules

of siege warfare did not begin to be treated as binding rules in the earlymodern era as a result of doctrinal writings: ‘In the final analysis,doctrine acquiesced to the fact of life that customary law in reality wasnot and did not have to be in accordance with rationality and morality to

be accepted by states as constituting law.’

In modern societies, valid law is usually said to require democraticlegitimacy, exemplified by an elected legislature Many traditional juristsargued that custom is the only genuinely democratic mode of law-making, reflecting the actual convictions of the ordinary people whopractise them, people who vote by consenting to those customs Butthinkers and writers from within the sceptical tradition representedhere tend also to be sceptical about suggestions that customary practicesare binding and valuable because they serve ‘as a community buildingdevice for the group whose collective wisdom creates custom’.3Instead,these scholars argue that notions of customary law as a distillation ofpopular practices tend to be indefensible, and that the relevant customsprove to be those of an influential group of insiders Lesaffer argues that

‘the customs of war were still very much determined by the sameprofessional elite that had dominated them for ages’, and it was thenotions of this elite on the requirements of honour and reciprocity that

drove changes in the rules of siege warfare.4Most modern historians ofthe common law, including three contributors to this book, argueanalogously that the common law embodies a set of insiders’ customs,the product of lawyers’ practices – among those a claim that what is done

in the name of the common law reflects popular custom:5

At a very basic level, no doubt, the values espoused by the common law would have been generally recognised by people in England, but the detailed working out of the rules derived from these values would cer- tainly not have had any such populist grounding This was all the work of lawyers, customary in the sense that the communis opinio doctorum might have been.6

Where customs conflict, hard moral, political or legal cases arise Insolving such cases, one’s understanding of the nature of customarypractices or laws, and in particular of the relationship between practiceand legal doctrine, will become evident Does custom provide the tacitbut indispensable matrix for shared moral and legal reasoning or is itmerely the dead hand of the past? Is the selection or preference of onecustom over rival conflicting ones itself purely a matter of custom? And,whatever lawyers, judges and decision-makers claim, how far and inwhat ways (if at all) are they really constrained by past customarypractices?7

The relation between reason and customary moralityKant’s position illustrates an extreme approach to the relationshipbetween reason and custom For him, customary moral rules and prac-tices are only ever conditionally binding, forms of reasoning ‘private’

to those groups of unreflective, dependent people who accept as

6 Chapter 7 below, p 165.

7 See Frederick Schauer’s contribution to this volume, tracing five ‘sceptical’ questions, interpretative questions which ‘anyone seeking to develop a theory of customary inter- national law, or a theory of the role of custom in common law decision-making, must at least attempt to answer’ Chapter 1 below, p 14.

T H E C H A R A C T E R O F C U S T O M A R Y L A W : A N I N T R O D U C T I O N 3

authoritative the relevant practices.8‘Public’ practical reason is of valuenot least because it renders moral knowledge accessible and justifiable toreflective individuals without the need for a mediating tradition: prac-tical reason can pull itself up by its own boot-straps So moral solutions

to conflicts among customary practices are not to be found by seekingone winning principle incipient within the relevant customs Instead, aKantian aims to impose upon those practices a moral meaning conceived

in line with prior rational principles, principles one imposes upononeself because of their rationality This means that a moral interpreta-tion of customary practices may ‘appear to us as forced – and be oftenforced in fact; yet, if the text can at all bear it, it must be preferred to aliteral interpretation which either contains absolutely nothing for moral-ity, or even works counter to its incentives’.9

Such accounts of moral principles as imposed upon custom arechallenged by three contributors to this volume Writing within thetradition of Anglo-American analytical philosophy, Ross Harrisonoffers an argument designed to show that morality both requiresand reaches beyond convention James Bernard Murphy traces anAristotelian argument for why ‘our choice is not between reason andprejudice or between custom and law’, developing an account of custom

as both conventionalising human nature and naturalising humanconventions:

Custom, Janus-like, faces toward human nature and toward stipulated law Custom turns our natural propensities toward eating, competing, and mat- ing into complex conventions of dining, gaming, and marrying; custom also turns our deliberate rational and legal conventions of arguing, evaluating, and judging into tacit practices as spontaneous and fluid as natural instinct 10

8

See e.g Groundwork 4:408: ‘Nor could one give worse advice to morality than by wanting to derive it from examples For, every example of it represented to me must itself first be appraised in accordance with principles of morality, as to whether it is also worthy to serve as an original example, that is as a model; it can by no means authoritatively provide the concept of morality.’

9 Kant is writing here of the rational interpretation of scripture: Religion within the Limits

of Mere Reason 6:110 On Kant on interpretation in this context, see Allen Wood,

‘Rational Theology, Moral Faith, and Religion’, in The Cambridge Companion to Kant (ed Paul Guyer, Cambridge: Cambridge University Press, 1992), pp 394–416; and Onora O’Neill’s Tanner lectures, in The Tanner Lectures on Human Values, vol 18 (ed Grethe B Peterson, Salt Lake City: Utah University Press, 1997), pp 269–308 (also reproduced at www.tannerlectures.utah.edu/nopq.html).

10

Chapter 3 below, pp 78 and 58.

4 A M A N D A P E R R E A U - S A U S S I N E A N D J A M E S B E R N A R D M U R P H Y

While some jurists like Bentham argue that custom cloaks the sinisterinterests of a dominant elite, Savigny and his fellow jurists of thehistorical school argue that custom is morality made visible, that therecan be no further moral standard to erect over it In his contribution tothis volume, Christoph Kletzer defends Hegel’s attempt to transcendsuch polar views by arguing that reason and custom evolve togethertowards concrete universality Comparing the role of custom in Hegel’sphilosophy of right and Savigny’s legal science, Kletzer develops aHegelian argument that ‘Custom and habit are not social expressionsopposed to freedom, they are not expressions of the ‘‘daily grind’’ to beovercome by self-expressive, heroic subjectivity but they rather areconditions of this subjectivity, play-forms of freedom.’11

The relation between reason and customary law

Kant’s approach to the relation between reason and law again illustrates

an extreme position In strong contrast to his approach on moralreasoning, Kant argues that lawyers aiming to resolve conflicts betweenlegal rules and practices must not appeal to rational principles of justice:lawyers’ reasoning must remain exclusively within the reasoning inter-nal to legislative commands and authoritative customs If a faculty oflaw ‘presumes to mix with its teaching something it treats as derivedfrom reason, it offends against the authority of the government’; a jurist

‘as an authority on the text, does not look to his reason for the laws but to the code of laws that has been promulgated and sanctioned by thehighest authority (if, as he should, he acts as a civil servant)’.12

Kant’s position is one that many practising lawyers would find ing As one Kant scholar remarks, ‘it is hard to see how the practical tasks ofthe practising lawyer, and in particular the practical task of the judge, can befully guided by norms set by state authority That might be possible if legalrules were true algorithms – but it does not seem at all plausible to thinkthat any practical rules are algorithms: they may specify what is to be done,but always under-specify what is actually done.’13No written law can giveexhaustive directions on its own interpretation and application, so cus-tomary rules and practices will be needed, not just to resolve faults incodification, but to guide judicial interpretation – and these guiding

stagger-11 Chapter 6 below, p 138 12 Conflict of the Faculties 7:22–3.

13

Onora O’Neill, ‘Kant on Reason, Authority and Interpretation’ (unpublished ence presentation, Newnham College, Cambridge, September 2004), p 12.

confer-T H E C H A R A C confer-T E R O F C U S confer-T O M A R Y L A W : A N I N T R O D U C T I O N 5

customary rules and practices will themselves be subject to change anddevelopment through interpretation.14

While for many thinkers this is enough to show that customary rulesare an immanent part of any legal system, some would insist that insteadcustom is at best a source rather than a part of law and that a formal legalact such as a judicial decision is needed to convert custom into custom-ary law On the latter account, custom is not itself a valid part of law(akin to legislation) but at best the raw material out of which a legis-lature or a court might fashion genuine positive law Thus FrederickSchauer argues that ‘the important questions about customary law arequestions about formal law’s use of pre-legal normative practices asthe basis for legal norms’.15 And Michael Lobban offers a detailedstudy of the way in which nineteenth-century English common lawyersapproached customary international law in very much this spirit, work-ing on the assumption that ‘international law was a source of English lawwithout being itself part of it’.16

In reflecting on the nature of such customary rules and practices, whilethe question of how to resolve hard cases is important, it is at least asimportant – and as difficult – to understand ‘what it is that makes the easycases easy’.17This returns us to the question of what effect, if any, doctrine

or reason has on customary practices, and the contributors to this volumeoffer diverse responses As already seen above, the approach to the questiontaken by many legal historians is to offer an account of lawyers’ own views

14

Related arguments have been made against the positions of contemporary legal vists To argue that a particular formulation is the correct view of a rule of law, as do teachers, textbook-writers, judges and counsel, is, as Brian Simpson argues against

positi-H L A Hart, ‘to participate in the system, not simply to study it scientifically’ See

A W B Simpson, ‘The Common Law and Legal Theory’, in Oxford Essays in Jurisprudence (ed A W B Simpson, Oxford: Clarendon Press, 1973), p 97 Gerald Postema builds a powerful critique of Bentham’s position on a similar point: ‘what the courts do has an important (though not necessarily decisive) impact on what the law is and what it requires.’ See Gerald Postema, Jeremy Bentham and the Common Law Tradition (Oxford: Clarendon Press, 1986), pp 456–7.

15 Chapter 1 below, p 18 Schauer follows Raz’s reading of Hart in treating a rule as a

‘content-independent’ reason for action, and distinguishes a custom (such as waking at

6 a.m.) from a rule Taking the example of the contemporary prohibition on slavery, he also draws a sharp distinction between the morally right and ‘a series of national normative acts (not in the legal sense, and certainly not items of international law)’ Other contributors to this volume, notably Murphy (Chapter 3 below) and Harrison (Chapter 2 below), would contest such a disjunction between custom and morality.

16

Chapter 11 below, p 277. 17 Chapter 1 below, p 28n34.

6 A M A N D A P E R R E A U - S A U S S I N E A N D J A M E S B E R N A R D M U R P H Y

of the relation between practice and doctrine while aiming to avoid ing or relying on a view of their own In the most extreme cases, reason orlegal philosophy is rejected as ‘a waste of time’, an enterprise ‘of interestonly for people too idle to engage in the intricacies of the positive law’: thusSavigny writes sarcastically of how ‘until today we come across people whotake their own juristic concepts and opinions to be purely reasonable, onlybecause they lack knowledge of their genealogy’.18

impos-But, in his comparative study of Savigny and Hegel on customary law,Christoph Kletzer contends with Hegel that, if legal history understood

as a scholarly enterprise is to be rational, then legal history understood

as a series of events ‘must at least be understood as making the ality of this historical inquiry possible, as being the history of therationality of historical inquiry Now, historical research is not an iso-lated enterprise, but can be rational only in a context of freedom, i.e inthe modern rational state Thus, rational historical enquiry is theenquiry into the development of reason as such.’19

ration-And, in her study of Gratian’s Decretum, a text which attempted toshow how diverse and seemingly inconsistent canons could be inter-preted and applied in a consistent way, Jean Porter concludes inAristotelian fashion:

Because written laws serve to formulate and correct custom, they will mally supercede and override customary law; yet, because they find their context and point within a broader framework of customary law, the customs

nor-of a people will provide the necessary context for their interpretation What is more, written law will have no purchase on a community, unless it reflects the practices of that community in some way; even a law that sets out to correct custom will necessarily reflect other aspects of the customary practices of a community, or it will lack purchase in the community for which it is intended Far from being a minor adjunct to the law properly so called, custom is seen from this perspective as the one essential component of any legal system, sufficient to sustain a rule of law under some circumstances, and one essential component of the rule of law under any and every circumstance.20

18 Chapter 6 below, p 128, summarising Savigny’s position on legal philosophy: and quoting from Friedrich Carl von Savigny, Vom Beruf unserer Zeit fu¨r Gesetzgebung und Rechtswissenschaft (Hildesheim: Georg Olms, 1967), p 115.

19 Kletzer recognises that this line of thought makes sense only to one who believes, like Hegel, that ‘reason has already actualised itself in the world in the French Revolution, in the advent of the rational liberal state that guarantees mutual recognition and free citizenship

to all’ Chapter 6 below, p 145.

20

Chapter 4 below, p 100.

T H E C H A R A C T E R O F C U S T O M A R Y L A W : A N I N T R O D U C T I O N 7

The nature of customary international law

Codify it, repeal it, abolish it; some form of customary law will inevitablyreappear But how far, if at all, does a lawyer need to rely on reasonedargument in offering an account of rules of customary law? The issue ofdemocratic legitimacy is especially contested in the case of customaryinternational law, which some jurists claim threatens the democraticsovereignty of national law-making This is one of the broader questions

at stake in four of the contributions to this volume on customaryinternational law

Two of these essays focus mainly on English approaches to national law in the nineteenth century In chapter 10, Perreau-Saussineargues that nineteenth-century English treatises on the law of nationsreflect three distinctive accounts of the relationship between reasonedargument and the practices of states The question of the relationshipbetween reasoned argument and customary international law also plays

inter-a key role in Michinter-ael Lobbinter-an’s inter-account of the view of the linter-aw of ninter-ationstaken by English courts in the nineteenth century Lobban suggests thatthe attitude of English courts to the law of nations hinged both onnineteenth-century common lawyers’ own understanding of the com-mon law (as deriving not from custom itself but from judicial decisionand ultimately ‘artificial reason’) and on their understanding of how farthe relevant rule of customary international law was understood to berationally defensible:

As with their use of the law of nature, it was drawn on not for the moral content of its precepts, but as a means of reasoning on the nature of the problem In novel cases, where English law offered no clear answers, courts (particularly before the mid-nineteenth century) were content to draw on the classic natural law works of Grotius, Bynkershoek or Vattel However, insofar as the law of nations was made up of contingent and changing state practice, it was not regarded as of itself part of the common law.

For ‘sceptics’ who believe that custom is all we have, to suggest thatparticular jurists or treatise writers could have an attributable influence

on the development of international law is akin to suggesting thatassisting at the delivery of a child makes one a biological parent

A history of the influence of a particular writer or jurist can and must

be a history of the work of a professional tradition, of advocates’ andjudges’ ‘shared attempt at addressing and resolving the problematic of

8 A M A N D A P E R R E A U - S A U S S I N E A N D J A M E S B E R N A R D M U R P H Y

order in a diverse world’ On such accounts, ‘there is a fundamentalproblem with assigning and measuring influence in international law,which is the ultimately collective character of so much of the work’: thecollective work of international lawyers is rooted in a reflective profes-sional tradition whose customs have a long history Central to thistradition, it is usually argued, is a style and culture traceable toGrotius and other creators of modern international law and one ‘still-existing, and no longer merely European’ It is a tradition that indivi-duals ‘may influence but hardly decisively’, not least since ‘its outcomes

at any time, though expressed definitively in terms of current national law, are at the same time part of a process, and are to thatextent provisional’: ‘Rise and fall, rise and fall, that is its enduringsignificance.’21

inter-In contrast, the two final contributions to this volume defendaccounts of customary international law that do aim to reach beyondlegal practice to fundamental principles which it is argued are in somesense prior to and constraining of that practice Arguing that ‘humaninstitutions exist and are capable of acting intelligibly only insofar asthey and others recognize them as defined and governed by norms,capable of grasping and following norms as norms (rather than merelystrategic markers of the parameters of their anomic choices)’, GeraldPostema sketches a general account of custom as a ‘normative practice’,

an account which he suggests can ‘illuminate the nature and typical mode

of operation of customary international law’.22 And John Tasioulasargues that ‘the account of custom we should favour is that which isbest justified by a political morality that offers the most attractivespecification of the values served by international law’ Tasioulas offers

an interpretative understanding of customary international law in whichthe ethical appeal of a candidate rule of international law figures amongthe criteria for determining whether it is a valid rule: this account, heargues, can serve as ‘a template for guiding judicial decision-making andassessing its correctness’

While the studies in this book focus mainly on the common law and

on customary international law, customary practices underpin every

21 J Crawford, ‘Public International Law in Twentieth-Century England’, in Jurists Uprooted: German Speaking Emigre Lawyers in Twentieth Century Britain (ed.

J Beatson and R Zimmermann, Oxford: Oxford University Press, 2004), pp 692, 699 and 700–1.

22

Chapter 12 below, p 306.

T H E C H A R A C T E R O F C U S T O M A R Y L A W : A N I N T R O D U C T I O N 9

legal system Customary rules of interpretation play a part in any legalsystem, however codified: no written law can give exhaustive directions

on its own interpretation, so customary rules and practices inevitablyguide judicial interpretation And those customary rules and practicesthemselves in turn will be subject to change and development throughinterpretation Ancient and modern, international, civilian and com-mon law: every interpretation and application of a written law relies on acomplicated set of shared customs And, once given, each interpretationand application of a written law itself extends that same set of customs

As James Bernard Murphy writes, ‘Like a beaver, law is both adapted toits customary environment and transforms that environment Many

of our customs began as laws and all successful law eventually becomescustomary.’23

23

Chapter 3 below, p 77.

10 A M A N D A P E R R E A U - S A U S S I N E A N D J A M E S B E R N A R D M U R P H Y

P A R T I Custom and morality: natural law, customary

law and ius gentium

1 Pitfalls in the interpretation of customary law

F R E D E R I C K S C H A U E R*

Much has been written on the legal status of customary law, but siderably less attention has been devoted to the question of determiningthe content of the customary law whose legal status (or not) is at issue.Like any other source of law, customary law presents the question ofinterpreting, applying, and enforcing the emanations from that source,but interpreting customary law – or interpreting the custom that is to bepart of the law1 – presents issues arguably more complex than thosepresented when we are considering the interpretation of constitutions,statutes, regulations, treaties, and even the common law My goal here is

con-to explore these interpretive questions, and con-to do so with perhaps what of a skeptical attitude This is not to say that such skepticism willturn out at the end of the day to be justified It is to believe, however, that

some-* This paper was prepared for the Colloquium on Customary Law organized by Dartmouth College and by the Lauterpacht Research Centre for International Law at Cambridge University, and held in Cambridge on 14–16 September 2005 I am grateful for research support from the Joan Shorenstein Center on the Press, Politics and Public Policy, Harvard University, and from the University of Chicago Law School, where

I served as the Daniel R Fischel and Sylvia M Neil Distinguished Visiting Professor of Law during the time when this paper was written and presented.

1

There is a long-standing dispute about the status of customary law, with some (such as

C K Allen) holding that custom is an immanent part of law in any common law system, and others (most prominently Jeremy Bentham and John Austin) insisting that a formal legal act (such as a judicial decision) is necessary to convert custom into customary law See Rupert Cross, Precedent in English Law (3rd edn, Oxford: Clarendon Press, 1977),

pp 157–9; Gerald J Postema, Bentham and the Common Law Tradition (Oxford: Clarendon Press, 1986), pp 4–14 and 219–30 This is an important dispute, but nothing

I say in this paper depends on its resolution Nevertheless, both of these opposing positions should be distinguished from the sense in which a common law system just is itself a customary system of law, albeit not necessarily congruent with the pre- legal customs that the common law as a customary system may choose to adopt See

A W B Simpson, ‘‘The Common Law and Legal Theory,’’ in A W B Simpson, ed., Oxford Essays in Jurisprudence (Second Series) (Oxford: Clarendon Press, 1973), pp 77–99.

addressing such skeptical questions is an inevitable task for any tory account of the role of customary law in common law adjudication,and perhaps to an even greater extent with respect to the role ofcustomary international law as a part of international law more gene-rally So, although in this paper I will ask more questions than I answer,

satisfac-my goal is to put on the table those interpretive issues that anyoneseeking to develop a theory of customary international law, or a theory

of the role of custom in common law decision-making, must at leastattempt to answer

Indeed, one of my goals here is to connect questions about customarylaw with many of the enduring questions about legal interpretationmore generally, questions whose importance seems all-too-oftenignored by theorists of customary law And so at the outset it might beworthwhile noting five of these questions One is a question focusing onthe identification of those features of some previous decision that enablesubsequent decision-makers to reference that decision or to rely upon it.Thus, in a debate marked by the earlier contributions of Goodhart,Simpson, and Montrose, and furthered in more recent times by LarryAlexander, most prominently, the question was raised as to whether itwas the facts of a previous decision, or the decision itself, or the wordsused to describe that decision, that enabled such a decision to constitute

a precedent for some other decision.2This issue is plainly relevant to thequestion of customary law, for custom is itself the aggregate of a series ofpast acts or decisions, but in order to make sense out of these past acts ordecisions we need to know which features of those acts or decisions arethe ones that have the quasi-authoritative status necessary for customitself to have such a status

At a more extreme level, sorting out the status of customary lawrequires confronting the challenges of American Legal Realism, thetradition which has raised enduring questions about the extent towhich, if at all, previous acts, events, or decisions actually do constrain

2 Larry Alexander, ‘‘Constrained by Precedent,’’ Southern California Law Review, vol 63 (1989), pp 1–64; Arthur L Goodhart, ‘‘The Ratio Decidendi of a Case,’’ Modern Law Review, vol 22 (1959), pp 117–24; Arthur L Goodhart, ‘‘Determining the Ratio Decidendi of a Case,’’ Yale Law Journal, vol 40 (1930), pp 161–83; J L Montrose,

‘‘The Ratio Decidendi of a Case,’’ Modern Law Review, vol 20 (1957), pp 587–95;

J L Montrose, ‘‘Ratio Decidendi and the House of Lords,’’ Modern Law Review, vol 20 (1957); pp 124–30; A W B Simpson, ‘‘The Ratio Decidendi of a Case,’’ Modern Law Review, vol 21 (1958), pp 155–60; A W B Simpson, ‘‘The Ratio Decidendi of a Case,’’ Modern Law Review, vol 20 (1957), pp 413–15.

judges and other subsequent decision-makers.3Legal Realism, especially

at its extremes, may not be plausible, but nor may it be plausible, as anempirical proposition, to believe that the canon of authoritative law is asexclusive and as constraining as pre-Realist legal theory supposed it to

be It seems strange, therefore, to consider the actual (empirical) ity of customary law without considering a long-standing debate aboutthe empirical authority of legal norms, legal rules, and legal decisionsmore generally

author-Closely related to these debates about the status of precedent and thestatus of legal authority in general are contemporary debates about legalinterpretation inspired primarily by Ronald Dworkin.4 Is there, intheory if not in practice, one right answer to any legal question? Doesthe interpretation of law resemble in important ways the interpretation

of literature? Is legal interpretation ultimately a coherence-based andholistic practice, rather than one in which individual legal items deter-mine particular legal results? These are the questions that Dworkin has

so prominently placed on the jurisprudential agenda, and they are noless relevant when the question is the interpretation (and identification)

of customary law

In the United States, and increasingly in Canada, Australia, SouthAfrica, and other countries with written constitutions and aggressivejudicial review, many of these interpretive debates have played out asdebates over the proper way to interpret a written constitution.5Is the

3 See, for example, Jerome Frank, Law and the Modern Mind (New York: Brentano’s, 1930); Laura Kalman, Legal Realism at Yale, 1927–1960 (Chapel Hill, NC: University of North Carolina Press, 1986); Karl Nickerson Llewellyn, The Bramble Bush: Some Lectures

on Law and Its Study (New York: Columbia University School of Law, 1930); William Twining, Karl Llewellyn and the Realist Movement (London: Weidenfeld & Nicolson, 1973).

4

Especially in Ronald Dworkin, Law’s Empire (Cambridge, MA: Harvard University Press, 1986); Ronald Dworkin, A Matter of Principle (Cambridge, MA: Harvard University Press, 1985); Ronald Dworkin, Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1977).

5

The literature, especially in the United States, is vast A sample of the issues can be found

in John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, MA: Harvard University Press, 1980); Antonin Scalia, et al., A Matter of Interpretation (Amy Guttman ed., Princeton: Princeton University Press, 1997); Akhil Reed Amar,

‘‘Intratextualism,’’ Harvard Law Review, vol 112 (1999), pp 747–803; Richard H Fallon, Jr., ‘‘Judicially Manageable Standards and Constitutional Meaning,’’ Harvard Law Review, vol 119 (2006), pp 1274–332; Michael J Perry, ‘‘The Authority of Text, Tradition, and Reason: A Theory of Constitutional ‘Interpretation,’ ’’ Southern California Law Review, vol 58 (1985), pp 551–602; Frederick Schauer, ‘‘An Essay on Constitutional Language,’’ UCLA Law Review, vol 29 (1982), pp 797–832.

process of constitutional interpretation essentially a common law cess, or does it more resemble the interpretation of a statute, a regula-tion, a contract, or a will? Is the goal of such interpretation to interpretthe words of the document as ordinary language, or instead as technicallanguage, or as embodying the intentions of those who first wrote them,

pro-or in their best possible light in view of the demands of mpro-orality anddemocracy and policy? To purport to interpret customary law is to dealwith many of the same issues, especially since the significant indetermi-nacy of customary law bears a close affinity with the linguistic indeter-minacy of many of the most important and most disputed provisions inwritten constitutions

Finally, but perhaps most importantly, how can the insights ofdisciplines other than law inform our understanding of the legalinterpretive process? Are there lessons from philosophy, from psycho-logy, from behavioral economics, and from other disciplines andsub-disciplines that can help us to make sense of the process bywhich customary law is created and interpreted? Implicit in thispaper is an affirmative answer to this question, and thus my attemptshere to relate these debates and insights outside of law is but a largermanifestation of the guiding principle of this paper – that there aremany insights and challenges in the legal and non-legal literatureoutside of the literature on customary law and outside of the litera-ture on international law that can valuably inform questions about theinterpretation of customary law, and that have been less of a presence

in the customary law and international law literatures than might bedesirable

Many of these jurisprudential and philosophical debates revolvearound the respective roles of the creator of some norm and theinterpreter of that norm How much freedom do authoritative inter-preters actually have? When such interpreters purport to be describingcustomary law, are they simply engaged in an act of description, orare they doing something that is more interpretive and more creativethan many within the customary law tradition have been willing toadmit? This question arises in each of the five debates I have justmentioned, and thus in describing this paper as having a ‘‘skeptical’’cast my ultimate goal is to attempt to make the domains of customarylaw and customary international law less complacent than they attimes have appeared to be, and to confront the same challenges thatmost other areas of legal analysis have been confronting forgenerations

Clarifying the question

It is common ground that in some domains custom can be a source oflaw, and that reaching a legal conclusion based on custom can be aslegitimate as reaching a legal conclusion on the basis of a statute, a legalprecedent, a provision of a written constitution, or the opinion of anauthoritatively recognized secondary source With respect to such con-ventional sources of law, it is a trivial point that first we locate anormative rule, and then determine the extent to which, if at all, itapplies to the matter at hand All legal rules are expressed in or transla-table into an if-then form, and thus the application of any of theforegoing sources typically involves, to oversimplify, determiningwhether the facts we have perceived fall within the scope-designating

or ‘‘if’’ part of the rule, and, if so, then determining what the normativeconsequent – the ‘‘then’’ part of the rule – requires to be done.6

Seen from the perspective of this point about the basic structure of aprescriptive rule, one preliminary but key question about custom as asource of law is whether the customary source must be normative As

H L A Hart so plainly stressed in his discussion of habits,7and othershave analyzed in the context of descriptive rules,8not all regularities ofhuman behavior are based on normative or rule-guided considerations

It is my custom – I am accustomed – to wake up at 6 a.m., but no rule tells

me to do so, and no rule (not even my own) would be violated were

I not to do so So too with the behavior of institutions and governments.For a long time it was the practice of airlines to have names thathad either geographic or weighty and serious connotations, and some-times both, as with ‘‘British Airways’’ and ‘‘Air France’’ and ‘‘United’’and ‘‘Continental.’’ When airlines started calling themselves things like

‘‘Virgin Atlantic’’ and ‘‘Song’’ and ‘‘Ted,’’ the practice shifted, yet nonormative rule was broken Similarly, although it is a fact that themajority of the nations of the world have names that end in ‘‘a,’’ andthus it is a fact that nations generally have names ending in ‘‘a,’’ there is

no normative or prescriptive standard that is violated by Peru, NewZealand, Pakistan, and Portugal

6 For a lengthier discussion of such structural matters about rules, see Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-Based Decision-Making in Law and in Life (Oxford: Clarendon Press, 1991), pp 23–7.

7 H L A Hart, The Concept of Law (2nd edn., Oxford: Clarendon Press, 1994), pp 9–12 and 55–60.

8

Schauer, op cit note 6, pp 18–22.

Yet, although there are many pre-legal regularities that are not mative, there are many that are, and now we approach more closely theissues that surround customary law.9 It is not mere habit that leadswholesale diamond merchants in New York, Antwerp, Johannesburg,and Jerusalem to sell to retailers (or cutters) on a ‘‘take it or leave it’’basis for a package of diamonds, but rather a well-entrenched normativepractice within the industry, such that departure would be the occasionfor criticism or the imposition of a non-legal sanction.10So too with therules of etiquette, or the rules of non-governmental organizations Onecannot call the police when a person slurps his soup or, if insufficientlysenior, traverses the lawn of a Cambridge college, but there can be littledoubt that these practices are imbued with all of the trappings ofnormativity save for the state as the source of authority.11

nor-Thus, the important questions about customary law are questionsabout formal law’s use of pre-legal normative practices as the basis forlegal norms.12And, even more precisely, these important questions areones about the possibility, nature, and desirability of formal law’s taking

as legally authoritative some pre-legal normative and authoritativepractice And the limitation to the ‘‘normative’’ and the pre-legally

‘‘authoritative’’ is crucial It is always open for a law-maker exercisingdiscretion to decide to follow some existing pre-legal normative or non-normative practice, but this is no different from the law-maker consult-ing any other non-normative or non-authoritative source of wisdom.Only when pre-legal normative customs are taken as (not-necessarily-conclusive) content-independent sources of authority do the genuineissues arise, and thus my question is about determining the content of

11 On the existence of such normative customary practices, and on their predominance in less legally complex societies, see, for example, Lloyd Fallers, Law Without Precedent: Legal Ideas in Action in the Courts of Colonial Busoga (Chicago: University of Chicago Press, 1969), pp 310–14; Roberto Mangabeira Unger, Law in Modern Society (Cambridge: Cambridge University Press, 1976), pp 48–58.

12 On normative custom, and on the disagreements about how to understand it and assess

it, see, for example, Paul Bohannon, ‘‘The Differing Realms of Law,’’ American Anthropologist, vol 6 (1965), pp 33–42; Stanley Diamond, ‘‘The Rule of Law versus the Order of Custom,’’ Social Research, vol 38 (1971), pp 42–72.

those customary normative sources that have already been socially orculturally but not-yet-legally determined to be authoritative.13

This limitation may be slightly idiosyncratic, or at least at odds withsome aspects of positive law Consider, for example, Mercer v Dunne,14

in which ‘‘it was held in 1905 that the fishermen of Walmer were entitled

by a local custom to dry their nets on a particular stretch of sand,’’15despite the fact that, in the absence of the custom, the practice wouldhave constituted an unlawful trespass Here there is no indication thatthe customary practice was done under claim of right, and no indicationthat the practice itself became normative A fisherman new to theWalmer area would not have been subject to criticism, we suppose, fornot participating in the custom, and instead drying his nets somewhereelse Yet when, as in this example, the practice is not normative – indeedthe case would strike the American property lawyer as one of adversepossession, a matter of substantive law, and not of the application ofcustom – the issues are different The extent to which law does or shouldreflect the existing conditions of the world, an issue to which I shallreturn in conclusion, is important, but here, as in most of the literature

on customary law, I shall limit my inquiry to the arguably narrowerquestion of the extent to which law does or should reflect the pre-legalnormative world – a world in which some but not all practices areauthoritative – on which it is superimposed

The questions I want to pose are largely questions existing at thecontrast between statute or codified law, on the one hand, and commonlaw and customary law, on the other In practice, this means that thesequestions presuppose the (moderate) determinacy of language, variousand sundry literary theorists, French philosophers, and other decon-structionists notwithstanding Although it is obvious that the legal use

of terms like ‘‘justice,’’ ‘‘equal protection,’’ ‘‘reasonable,’’ ‘‘fair,’’ portionate,’’ and ‘‘necessary’’ provide little constraint on decision andallow much room for law-making under the rubric of ‘‘interpretation,’’

‘‘pro-13 On understanding authority in just this content-independent way, the locus classicus is

H L A Hart, ‘‘Commands and Authoritative Legal Reasons,’’ in Essays on Bentham: Jurisprudence and Political Theory (Oxford: Clarendon Press, 1982), pp 243–68 See also Joseph Raz, The Authority of Law: Essays on Law and Morality (Oxford: Clarendon Press, 1979); Donald H Regan, ‘‘Reasons, Authority, and the Meaning of ‘Obey’: Further Thoughts on Raz and Obedience to Law,’’ Canadian Journal of Law and Jurisprudence, vol 3 (1990), pp 3–28; Frederick Schauer, ‘‘The Questions of Authority,’’ Georgetown Law Journal, vol 81 (1992), pp 95–115.

14

[1905] 2 Ch 538. 15 Cross, op cit note 1 , p 162.

such is not the case, or at least is not necessarily the case, when law usesterms like ‘‘two,’’ ‘‘insect,’’ and ‘‘parliament.’’ It is true that most of theterms used by the law have, following Hart, a core of settled applicationand a fringe or penumbra of uncertainty.16Still, I take it as a given thatcommon linguistic usage, whether ordinary or technical, can and oftendoes serve satisfactorily to designate a core,17and that shared under-standings about linguistic meaning are what typically or standardlymake it possible for statutes often to generate ‘‘clear’’ or ‘‘easy’’ cases.18All of this may in some contexts be open to debate, but, if we are to try tofocus on the special problems of customary law, and on why customarylaw would often seem especially problematic, we need to assume, if onlyfor the sake of argument, that statute or codified law has the capacity togenerate unique or tightly clustered interpretations, that in mostadvanced legal systems it often does so, and that it does so by virtue ofthe ability of human beings to read off from a printed page a single ortightly clustered set of meanings for particular sentences, meanings thatare themselves a function of the similar capacities of individual words.Given this background assumption, I want to proceed by offering aseries of questions, each labeled with the name of a theorist who might

be said to have, or at least have for me, inspired the question Littleshould be made of the names, however, for my goal here is well removedfrom exegesis of this or that thinker The names should, however,provide a convenient way of designating particular and skeptical ques-tions about the practice of interpreting existing normative custom

Hanson’s questionInterpreting custom requires an interpreter And thus we can conceive

of the interpreter of custom as someone looking out over a vast sea ofhuman behavior and identifying the particular strands, patterns, andpractices that might constitute a normative custom In doing so, per-haps she is simply looking at the law Perhaps all normative customs are

16 Hart, op cit note 7 , pp 124–54.

17 As Lon Fuller pointed out against Hart, however, sometimes the core may be designated

by purpose and not by literal meaning, Lon L Fuller, ‘‘Positivism and Fidelity to Law – A Reply to Professor Hart,’’ Harvard Law Review, vol 71 (1958), pp 630–72, a point which Hart ultimately conceded in H L A Hart, Essays in Jurisprudence and Philosophy (Oxford: Clarendon Press, 1983), pp 6–8.

18

See Walter Sinnott-Armstrong, ‘‘Word Meaning in Legal Interpretation,’’ San Diego Law Review, vol 42 (2005), pp 465–83.

part of the relevant body of law, such that all of the normative customs

of England are immanent in the English common law, and all of thenormative customs of nations – and not just those that are understood

as legally normative under the doctrine of opinio juris – are part ofinternational law But, even if this is so, and certainly if it is not so, thenormative customs of a jurisdiction that are explicitly incorporatedwithin the law constitute but a subset of the totality of that jurisdiction’snormative customs And from this it follows that the task of interpreta-tion turns out to be, in significant part, the task of selection, the selectionfrom the universe of normative customs those normative customs thatwill explicitly be part of the law The relevant custom for some decisiondoes not simply leap out and grab the interpreter, but rather is selected

by that interpreter from among the entire field of that jurisdiction’snormative customs,

Once we see that the interpreter’s task includes the task of selection,however, then we are forced to consider the grounds on which thatselection is made The analogy here, and the analogy that explains thename of this section, is to the idea long referred to in the philosophy ofscience as theory-laden observation In his autobiography, Karl Popperrecounts the time at the end of a lecture when he gave an assignment

to the attendees, to be completed prior to the next lecture And thatassignment was simply to go out and observe Period Full stop Popper’spoint was that simple observation, without purpose and without theory,was impossible, and he wanted his students to recognize that the task ofobservation required them to have, use, or develop a theory of what theywere observing and why they were observing it.19This idea was devel-oped more fully by Norwood Hanson,20 and it is now more or lesscommonplace that the task of observing is not simply one of recordingthe world as it exists, but instead necessarily involves recording thoseparts of the world that are recorded for some reason, and then groupingthose parts into categories that once again reflect a goal, a purpose, or atheory

So too with the observation of custom The interpreter of custom isnot simply labeling all of the normative customs of the world or ofEngland or of whatever, but is identifying some customs for some

19 Karl R Popper, Unended Quest: An Intellectual Autobiography (London: Open Court, 1976).

20

Norwood R Hanson, Patterns of Discovery (Cambridge: Cambridge University Press, 1958).

purpose, much like the practice, as has been said of determining lative history in a non-parliamentary system with a large legislature, oflooking over a crowd and picking out your friends.21There are manymore normative customs than are explicitly incorporated into law forpurposes of making a legal decision, and thus anyone who seeks toexplain the incorporation of custom into law must address the question

legis-of which customs are selected, which are not selected, and what purpose,goal, theory, or whatever drives the selection of some customs whileothers remain disregarded Indeed, we would not be surprised to dis-cover that there was a custom among the non-fishermen of Walmer,especially after a few pints, to grumble about the presumptuousness ofthe fishermen, and the selection of the custom of the fishermen’s actionsrather than the custom of the non-fishermen residents’ grousing about itmay well reflect some unarticulated but no less real and no less causalfactor that led to the selection of the particular custom that determinedthe outcome

The selection of some customs but not others is not necessarily, as itmay have been in my hypothetical gloss on the case of the fishermen ofWalmer, driven by interpreter preferences for certain outcomes Theselection may be a function of the informational institutions that makesome customs more known to interpreters than others It may be afunction of dispute-resolution mechanisms that create incentives forsome customs to be pressed while others are not And it may be afunction of the behavioral aspects of human decision-making, and theway in which some facts, including some customs, are more salient forsome people than they are for others All of these factors, and undoubt-edly many more, will influence the selection of some customs for atten-tion while other customs are ignored, but the only point here is that wecannot fully understand the practice of interpreting custom until weunderstand that the very process of selecting the custom to be inter-preted reflects background facts and norms of the interpreter and herinstitutional environment The selection of what to interpret mayindeed be more important to an outcome than the interpretation ofwhat is selected, and attention to the impossibility of observation andselection that is not theory-laden helps focus our attention on thisomnipresent phenomenon

21 Conroy v Aniskoff, 507 US 511, 519 (1993), quoting United States federal judge Harold Laventhal And the point is similar to that implicit in the saying that ‘‘the Devil can cite Scripture for his purpose.’’ William Shakespeare, The Merchant of Venice, Act 1, Scene 3.

Wittgenstein’s questionShortly after the publication of Saul Kripke’s Wittgenstein on Rules andPrivate Language,22a number of American legal scholars attempted toconscript Kripke’s Wittgenstein into a radical indeterminacy agenda,claiming that Wittgenstein, as explained by Kripke, was challenging thevery possibility of determinate linguistic meaning.23 Once we under-stood Wittgenstein’s message as interpreted by Kripke, so the argumentwent, we would understand that law’s aspiration for determinate guid-ance removed from the ideological dimensions of particular decisionswas doomed to failure Because law’s pretensions to determinacy rested

on the determinacy of language, it was said, Wittgenstein and Kripkehad undercut the foundations of many of law’s claims to neutrality andimpersonality, aspirations that often parade under the banner of the

‘‘rule of law.’’

Those who made this argument succeeded in little more than strating that they understood neither Wittgenstein nor Kripke.24Thatlanguage may ultimately rest on agreement, and that such agreementmay in some way be contingent, says no more about the ability ofcontingently created language to constrain than it does about the ability

demon-of contingently created guns to cause fatal injury.25And the fact that theagreement can change over time once again says far less than it wasalleged to say, for just as I can stand on a moving train so too can

I actually use language and use its meanings even as language and itsconstituent meanings are ever so slowly shifting

Yet, although the original claims about the relevance of Kripke’sWittgenstein to legal interpretation were somewhere between exagger-ated and ridiculous, there remains an important lesson to be learnedfrom the literature on Wittgenstein and rule-following, and it is a lessonthat is especially relevant to common law and to custom Both common

22

Saul A Kripke, Wittgenstein on Rules and Private Language: An Elementary Exposition (Cambridge, MA: Harvard University Press, 1982).

23 A well-known example is Mark V Tushnet, ‘‘Following the Rules Laid Down: A Critique

of Interpretivism and Neutral Principles,’’ Harvard Law Review, vol 96 (1983),

pp 542–83.

24 I will leave to others the question whether Kripke understood Wittgenstein, although

I should note that I believe he did not, and that Wittgenstein’s point was simply that we

do not and cannot ask the very question that Kripke was asking See G P Baker and

P M S Hacker, Scepticism, Rules and Language (Oxford: Basil Blackwell, 1984).

25

Or so I have argued in Frederick Schauer, ‘‘Rules and the Rule-Following Argument,’’ Canadian Journal of Law and Jurisprudence, vol 3 (1990), pp 187–92.

law and custom are continuously developing, and one way to stand their development is as a progression or as a series So if we seecustom as a series of social decisions and (part of) the common law as aseries of judicial ones, we can then understand the act of making aparticular common law decision, and less obviously the act of deciding

under-in accordance with some custom, as beunder-ing a decision under-involvunder-ing theextension of a pre-existing series And this is where Wittgenstein andthe rule-following considerations enter the picture in a more seriousway When we have a series of decisions, and when the series up untilnow is not encrusted in a canonical linguistic definition of what thatseries stands for, the agent who extends the series has, in theory, a range

of options about the extension of that series, in just the way in which hewho seeks to extend the series 1000, 1002, 1004, 1006, may have arange of options about which number is to come next, options that areless determined by the previous items in the series than is commonlysupposed So insofar as the understandings that created the previousseries are more flexible than the understandings that create the rules oflanguage, a new interpretive opportunity may turn out to be less con-strained by previous members of the series than might be thought, andthan might be urged by the extender of the series

Like the lesson of theory-laden observation, this is not a lesson ofradical indeterminacy, but rather a lesson about the potential for choice(although not always on the part of an individual interpreter, who maywell be constrained by numerous factors external to the previous deci-sions26) built into the practices of interpretation, a potential that is oftendenied, especially in law, by those interpreters who wish to claim thatthey are doing nothing but following the dictates of the previous mem-bers of some series The sophisticated version of this argument – theargument that constraint in fact does come from the previous decision

in the series – is offered by Ronald Dworkin, especially as he uses thevivid metaphor of the chain novel to illustrate the nature of legalinterpretation.27 The author of the next chapter of the chain novel,Dworkin argues, or the next episode of a television soap opera, isconstrained by notions of interpretive integrity to continue the broadthemes established by the earlier chapters or episodes, even as thatinterpreter has a limited degree of freedom to shift the story, ever so

26 And that is why the person who says that 1013 is the next member of the series 1000,

1002, 1004, 1006 has simply given the wrong answer.

27

Ronald Dworkin, Law’s Empire (Cambridge, MA: Harvard University Press, 1986).

slightly, in one direction or another This degree of interpretive dom, argues Dworkin, is to be used to make the story ‘‘the best it can be,’’but there is a vast difference, he insists, between making this story thebest it can be and simply writing a new story or heading off in adramatically different direction.

free-There may not be that much in the distinction between Wittgenstein’srule-following lessons and Dworkin’s chain novel metaphor, becauseboth stand for the proposition that the past constrains but may not bindinexorably And, when the past is indeed perceived as binding inexor-ably, the important lesson is that the bindingness comes not solely fromthe items of the past, but how those past items are treated by theconstraints of the present And so too with the interpretation of customand common law Interpreters may have good reasons for denying thedegree of choice they actually have, and Dworkin may be right, as hestressed in his earlier work, to point out that interpreters may themselvesnot see the degree of choice they actually have,28but that does not meanthat the choice does not exist So when the interpreter of customaryinternational law, for example, says that in this instance the practices ofnations over some past period of years stand for this proposition now, itbecomes permissible to ask what else those customs of the past mightnow be taken to stand for, and then to ask further why the selectedinterpretation was selected rather than some number of others thatmight have been equally consistent with the past decisions and thepast acts, and thus, in the large, with the customs of the past

Consider, for example, the question of slavery under internationallaw, a question often posed independently of (or prior to) the explicitprohibition in Article 4 of the Universal Declaration of Human Rights of

1948.29 Although there was, even prior to 1948, a series of nationalnormative acts (not in the legal sense, and certainly not items of inter-national law) that would constitute a progression in the direction of

28

Ronald Dworkin, Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1977) Understanding Dworkin’s claim that there is ‘‘one right answer’’ to any legal question requires understanding that for Dworkin the important perspective is that of the judge making the decision When we read Dworkin charitably, we see that his ‘‘one right answer’’ claim is not an ontological one, but is rather about the processes by which the judge comes to what he believes to be the right answer And if we see the ‘‘one right answer’’ claim as one of judicial phenomenology and not one of legal ontology, the claim becomes far more plausible.

29 See Anthony D’Amato, The Concept of Custom in International Law (Ithaca, NY: Cornell University Press, 1971), pp 131–2 Dick W P Ruiter, Legal Institutions (Dordrecht: Kluwer Academic Publishers, 2001), pp 119–34.

treating slavery as impermissible, the conclusion that this progressionwould continue, and that slavery should thus be considered violative ofcustomary international law, presupposes a continuation of the pro-gression in a certain way That this was the morally right result is beyondquestion, but the interpretive issue is how much this result was dictated

by the progression itself, and how much by the determination, dently of the progression, that slavery simply is morally and thus legallyimpermissible And so too, in reverse, with questions of desuetude, forthe practice of assuming a continuing decline in the existence or nor-mativity of some norm from the decline that has taken place to the point

indepen-of decision may again be to impose upon a trend something that may beless inherent in the trend than is commonly assumed

None of this is to say that legal custom cannot proceed by imposingcertain second-order norms on the identification of first-order norms,and it could well be that an assumption about the continuation of somepast trend is one of those second-order norms But the selection of thosesecond-order norms will have to be based on some value or goal – thebelief that customary norms tend towards efficiency, for example30 –rather than being the product of something close to logical necessity.And thus Wittgenstein’s lesson for customary law is not a lesson of legalindeterminacy, but rather a lesson about the contingency and second-order normativity of the sources of short- or intermediate-term legaldeterminacy

Quine’s questionMuch the same question can be posed in terms of the relationshipbetween observations (the data) and the theory that might explainthem As W V O Quine – and Pierre Duhem – properly insisted, notheory or explanation or account of multiple data could ever be uniquelydetermined by that data Instead, extracting a theory from the data, orimposing an explanation on a set of data, requires the use of supple-mentary premises in order to wind up with the one explanation amongthose that fit the data in preference to others which are different from

30 See, for example, Ronald A Cass, ‘‘Economics and International Law,’’ New York University Journal of International Law and Politics, vol, 29 (1997), pp 473–502; Jeffrey L Dunoff and Joel P Trachtman, ‘‘Economic Analysis of International Law,’’ Yale Journal of International Law, vol 24 (1999), pp 1–54.

the one selected, and which have different extensions into the future, butwhich also fit the existing data.31

Any attempt to explain a past or existing custom is plagued by thesame phenomenon of under-determination Whether it be the decisions

of multiple previous courts on multiple different occasions, or themultiple foreign policy or human rights decisions of multiple nations

on multiple past occasions, or even just the identification of behavioralregularities of pairs of self-interested states,32the statement (the expla-nation) that the custom is such-and-such is a statement that could havebeen different but which even if different would still nonetheless haveexplained the same data In science, the norms and goals of science help

to make these decisions, as when we prefer simpler to more complexexplanations, or when we select those explanations that appear to havethe greatest future pragmatic or predictive value So too with history,where the explanation selected by the historian to explain the past is notthe only explanation that would explain the past, but may well be theexplanation the fits best with the present goals and norms of the parti-cular historian and the present goals and norms of history as a disciplineand a profession There are parallel individual and institutional normsfor law, but what are these norms? Is one explanation of past custombetter than another because it is morally preferable, as is perhaps the casewith the conclusion that a prohibition on slavery is part of customaryinternational law? Or is an explanation chosen over its logically equiva-lent competitors because it is better suited pragmatically to futureapplication and enforcement by some court, or by some internationaltribunal, or by some enforcement body? Or is it grounds for selecting anexplanation that the explanation selected will be more easily explainable

to the non-legal actors who are expected to be bound by it?

All of these and many other background norms pervade any use ofcustom as a source of law, whether domestic or international Thequestion for law, or the question for those who would use and interpretcustomary law, is consequently the question of what these backgroundnorms are, how they have been chosen, how they foster the goals of thelarger institutional setting in which they are employed, and how they

31 Pierre Maurice Marie Duhem, The Aim and Structure of Physical Theory (P P Wiener trans., Princeton: Princeton University Press, 1954); Willard Van Orman Quine, Word and Object (Cambridge, MA: MIT Press, 1960).

32

See Jack L Goldsmith and Eric A Posner, ‘‘A Theory of Customary International Law,’’ University of Chicago Law Review, vol 66 (1999), pp 1113–77.

lead to the selection of some explanations for past customary acts inpreference to others Like Wittgenstein’s question, and indeed to someextent like Hanson’s, Quine’s question is a question about unacknow-ledged options, and about the factors that led to the choice of one optionover the typically unacknowledged but in some sense equally valid orequally available alternatives And thus the question about some state-ment of a custom based on a series of events is the question of why adifferent statement, with different implications, but which would havebeen based on the same events, was not the one that in the final analysisemerged.

Llewellyn’s questionAlthough law is not only an adversarial practice, it is significantly so.And thus it is important to see the use of customary law in its typicalconcrete adversarial or litigation-based context.33Whenever one partyoffers an interpretation of customary law to support its side of anadversarial proceeding, the other side, assuming it does not immediatelyconfess error,34has a strong incentive to find an equally authoritative orequally persuasive source that would militate in favor of the oppositeresult Sometimes this will be a statute, sometimes it will be a reportedprecedent decision, and sometimes it will be the published opinion of an

33

Although much of international law arises outside of the context of formal litigation, it

is nevertheless often adversarial, with nations arguing their preferred interpretations

to international organizations, to human rights groups and other NGOs, and to the community of world public opinion.

34

The statement in the text is a joke Although the unwillingness to challenge settled law is

an omnipresent legal practice all too often ignored by court-obsessed legal theorists, cases almost never get to court unless the law is less clear, and thus the likelihood that a party who has gotten as far as actual litigation will simply give up in the face of strong legal precedent is essentially non-existent The phenomenon is generally known as the selection effect, see George Priest and William Klein, ‘‘The Selection of Disputes for Litigation,’’ Journal of Legal Studies, vol 13 (1984), pp 1–23; Frederick Schauer,

‘‘Judging in a Corner of the Law,’’ Southern California Law Review, vol 61 (1988),

pp 1717–33, and has important implications for thinking about the interpretation of customary law The instances of interpretation will not be a fair sample of the full scope

of customary law, and so, although there are difficult and important issues involved in the interpretation of custom, the difficulty of those issues may not tell us everything about the phenomenon of customary law itself How to resolve the hard cases is important, but equally important is to understand what it is that makes the easy cases easy.